diff --git "a/grafted/wandb/run-20241127_215015-torch-grafted-redux/files/output.log" "b/grafted/wandb/run-20241127_215015-torch-grafted-redux/files/output.log"
new file mode 100644--- /dev/null
+++ "b/grafted/wandb/run-20241127_215015-torch-grafted-redux/files/output.log"
@@ -0,0 +1,34352 @@
+I1127 21:50:15.810525 137274321021824 train.py:125] NOTE: Initializing train dataset...
+I1127 21:50:15.810664 137274321021824 train.py:125] NOTE: Global batch size 1024 on 1 hosts results in 1024 local batch size. With 1 dev per host (1 dev total), that's a 1024 per-device batch size.
+I1127 21:50:16.040117 137274321021824 dataset_info.py:707] Load dataset info from /data/tensorflow_datasets/imagenet2012/5.1.0
+I1127 21:50:16.058943 137274321021824 reader.py:261] Creating a tf.data.Dataset reading 1024 files located in folders: /data/tensorflow_datasets/imagenet2012/5.1.0.
+WARNING:tensorflow:From /home/jason-chou/.pyenv/versions/3.11.10/lib/python3.11/site-packages/tensorflow_datasets/core/reader.py:101: CounterV2 (from tensorflow.python.data.experimental.ops.counter) is deprecated and will be removed in a future version.
+Instructions for updating:
+Use `tf.data.Dataset.counter(...)` instead.
+W1127 21:50:16.093780 137274321021824 deprecation.py:50] From /home/jason-chou/.pyenv/versions/3.11.10/lib/python3.11/site-packages/tensorflow_datasets/core/reader.py:101: CounterV2 (from tensorflow.python.data.experimental.ops.counter) is deprecated and will be removed in a future version.
+Instructions for updating:
+Use `tf.data.Dataset.counter(...)` instead.
+I1127 21:50:16.116578 137274321021824 logging_logger.py:49] Constructing tf.data.Dataset imagenet2012 for split _EvenSplit(split='train', index=0, count=1, drop_remainder=False), from /data/tensorflow_datasets/imagenet2012/5.1.0
+I1127 21:50:16.183868 137274321021824 api.py:460] Data before pre-processing:
+{'file_name': <tf.Tensor 'args_1:0' shape=() dtype=string>, 'image': <tf.Tensor 'args_2:0' shape=() dtype=string>, 'label': <tf.Tensor 'args_3:0' shape=() dtype=int64>, 'tfds_id': <tf.Tensor 'args_4:0' shape=() dtype=string>, '_id': <tf.Tensor 'args_0:0' shape=() dtype=int32>}
+INFO:tensorflow:Using RandAug.
+I1127 21:50:16.447511 137274321021824 api.py:460] Using RandAug.
+WARNING:tensorflow:From /home/jason-chou/.pyenv/versions/3.11.10/lib/python3.11/site-packages/tensorflow/python/util/dispatch.py:1260: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
+Instructions for updating:
+Use `tf.cast` instead.
+W1127 21:50:16.625000 137274321021824 deprecation.py:50] From /home/jason-chou/.pyenv/versions/3.11.10/lib/python3.11/site-packages/tensorflow/python/util/dispatch.py:1260: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
+Instructions for updating:
+Use `tf.cast` instead.
+I1127 21:50:18.860011 137274321021824 api.py:460] Data after pre-processing:
+{'image': <tf.Tensor 'add:0' shape=(224, 224, 3) dtype=float32>, 'labels': <tf.Tensor 'one_hot:0' shape=(1000,) dtype=float32>}
+I1127 21:50:18.941294 137274321021824 train.py:125] NOTE: Running for 112603 steps, that means 90.000345 epochs
+I1127 21:50:19.589837 137274321021824 train.py:125] NOTE: Creating model...
+Weight decay for: conv_proj.weight
+Weight decay for: encoder.layers.encoder_layer_0.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_0.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_0.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_0.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_1.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_1.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_1.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_1.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_2.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_2.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_2.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_2.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_3.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_3.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_3.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_3.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_4.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_4.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_4.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_4.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_5.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_5.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_5.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_5.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_6.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_6.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_6.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_6.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_7.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_7.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_7.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_7.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_8.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_8.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_8.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_8.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_9.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_9.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_9.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_9.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_10.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_10.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_10.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_10.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_11.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_11.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_11.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_11.mlp.3.weight
+Weight decay for: heads.head.weight
+Weight decay for: conv_proj.weight
+Weight decay for: encoder.layers.encoder_layer_0.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_0.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_0.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_0.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_1.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_1.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_1.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_1.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_2.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_2.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_2.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_2.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_3.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_3.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_3.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_3.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_4.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_4.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_4.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_4.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_5.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_5.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_5.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_5.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_6.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_6.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_6.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_6.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_7.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_7.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_7.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_7.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_8.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_8.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_8.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_8.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_9.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_9.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_9.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_9.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_10.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_10.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_10.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_10.mlp.3.weight
+Weight decay for: encoder.layers.encoder_layer_11.self_attention.in_proj_weight
+Weight decay for: encoder.layers.encoder_layer_11.self_attention.out_proj.weight
+Weight decay for: encoder.layers.encoder_layer_11.mlp.0.weight
+Weight decay for: encoder.layers.encoder_layer_11.mlp.3.weight
+Weight decay for: heads.head.weight
+I1127 21:50:20.530162 137274321021824 train.py:125] NOTE: Running initial or final evals...
+I1127 21:50:20.530552 137274321021824 train.py:125] NOTE: Init evaluator: val…
+Steps:0/112603 [0.0%]
+I1127 21:50:20.532502 137274321021824 reader.py:261] Creating a tf.data.Dataset reading 64 files located in folders: /data/tensorflow_datasets/imagenet2012/5.1.0.
+I1127 21:50:20.564504 137274321021824 logging_logger.py:49] Constructing tf.data.Dataset imagenet2012 for split _EvenSplit(split='validation', index=0, count=1, drop_remainder=False), from /data/tensorflow_datasets/imagenet2012/5.1.0
+I1127 21:50:20.599688 137274321021824 api.py:460] Data before pre-processing:
+{'file_name': <tf.Tensor 'args_1:0' shape=() dtype=string>, 'image': <tf.Tensor 'args_2:0' shape=() dtype=string>, 'label': <tf.Tensor 'args_3:0' shape=() dtype=int64>, 'tfds_id': <tf.Tensor 'args_4:0' shape=() dtype=string>, '_id': <tf.Tensor 'args_0:0' shape=() dtype=int32>}
+I1127 21:50:20.775087 137274321021824 api.py:460] Data after pre-processing:
+{'image': <tf.Tensor 'add:0' shape=(224, 224, 3) dtype=float32>, 'labels': <tf.Tensor 'one_hot:0' shape=(1000,) dtype=float32>}
+I1127 21:50:20.880522 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:0/112603 [0.0%]
+I1127 21:51:41.671772 137274321021824 utils.py:1231] [0] val/acc@1 = 0.0045041454081632655
+I1127 21:51:41.671978 137274321021824 utils.py:1231] [0] val/loss = 6.883525214633163
+I1127 21:51:41.672120 137274321021824 utils.py:1231] [0] z/secs/eval/val = 80.79130142200302
+I1127 21:51:41.672218 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 80.79130142200302
+I1127 21:51:41.672323 137274321021824 train.py:125] NOTE: Starting training loop, compiling the first step...
+I1127 21:52:09.610654 137274321021824 utils.py:1231] [1] l2_params = 207.73155841164476
+I1127 21:52:09.610920 137274321021824 utils.py:1231] [1] train/loss = 6.907755374908447
+I1127 21:52:09.611017 137274321021824 utils.py:1231] [1] l2_grads = 0.593871533870697
+I1127 21:52:09.611075 137274321021824 utils.py:1231] [1] lr = 0.0
+I1127 21:52:09.611147 137274321021824 utils.py:1231] [1] uptime = 118.97350603800442
+I1127 21:52:09.611201 137274321021824 utils.py:1231] [1] examples_seen = 1024.0
+I1127 21:52:09.611250 137274321021824 utils.py:1231] [1] progress = 8.88075806150813e-06
+I1127 21:52:09.611297 137274321021824 utils.py:1231] [1] epoch = 0.0007992712893791364
+I1127 21:52:09.611342 137274321021824 train.py:125] NOTE: Steps:0/112603 [0.0%]
+I1127 21:52:14.180502 137274321021824 utils.py:1231] [2] l2_params = 207.73155841164967
+I1127 21:52:14.180751 137274321021824 utils.py:1231] [2] train/loss = 6.907755196094513
+I1127 21:52:14.180845 137274321021824 utils.py:1231] [2] l2_grads = 0.5919386744499207
+I1127 21:52:14.180950 137274321021824 utils.py:1231] [2] lr = 1.0000000000000001e-07
+I1127 21:52:14.181006 137274321021824 utils.py:1231] [2] uptime = 123.54336848400271
+I1127 21:52:14.181068 137274321021824 utils.py:1231] [2] examples_seen = 2048.0
+I1127 21:52:14.181142 137274321021824 utils.py:1231] [2] progress = 1.776151612301626e-05
+I1127 21:52:14.181201 137274321021824 utils.py:1231] [2] epoch = 0.0015985425787582727
+I1127 21:52:14.181257 137274321021824 train.py:125] NOTE: Steps:0/112603 [0.0%]
+I1127 21:56:15.091482 137274321021824 utils.py:1231] [50] l2_params = 207.73017658172895
+I1127 21:56:15.091721 137274321021824 utils.py:1231] [50] train/loss = 6.907232761383057
+I1127 21:56:15.091821 137274321021824 utils.py:1231] [50] l2_grads = 0.6004149317741394
+I1127 21:56:15.091898 137274321021824 utils.py:1231] [50] lr = 4.9e-06
+I1127 21:56:15.091959 137274321021824 utils.py:1231] [50] uptime = 364.45432087200606
+I1127 21:56:15.092023 137274321021824 utils.py:1231] [50] examples_seen = 51200.0
+I1127 21:56:15.092082 137274321021824 utils.py:1231] [50] progress = 0.0004440379030754065
+I1127 21:56:15.092149 137274321021824 utils.py:1231] [50] epoch = 0.03996356446895682
+I1127 21:56:15.092220 137274321021824 utils.py:1231] [50] img/sec/core = 204.0255933272697
+I1127 21:56:15.092284 137274321021824 utils.py:1231] [50] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.06691970899666759
+I1127 21:56:15.092350 137274321021824 utils.py:1231] [50] core_hours = 0.06691970899666759
+I1127 21:56:15.092432 137274321021824 train.py:125] NOTE: Steps:50/112603 [0.0%]
+Walltime:6m4s (0s eval)
+ETA:6d12h55m
+Total train time:6d12h59m
+I1127 22:01:00.140702 137274321021824 utils.py:1231] [100] l2_params = 207.72771226368462
+I1127 22:01:00.140950 137274321021824 utils.py:1231] [100] train/loss = 6.902469277381897
+I1127 22:01:00.141052 137274321021824 utils.py:1231] [100] l2_grads = 0.6215766668319702
+I1127 22:01:00.141118 137274321021824 utils.py:1231] [100] lr = 9.900000000000002e-06
+I1127 22:01:00.141187 137274321021824 utils.py:1231] [100] uptime = 649.5035454970057
+I1127 22:01:00.141262 137274321021824 utils.py:1231] [100] examples_seen = 102400.0
+I1127 22:01:00.141312 137274321021824 utils.py:1231] [100] progress = 0.000888075806150813
+I1127 22:01:00.141361 137274321021824 utils.py:1231] [100] epoch = 0.07992712893791364
+I1127 22:01:00.141412 137274321021824 utils.py:1231] [100] img/sec/core = 179.61809953125413
+I1127 22:01:00.141473 137274321021824 utils.py:1231] [100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.1461000491702786
+I1127 22:01:00.141521 137274321021824 utils.py:1231] [100] core_hours = 0.1461000491702786
+I1127 22:01:00.141584 137274321021824 train.py:125] NOTE: Steps:100/112603 [0.1%]
+Walltime:10m50s (0s eval)
+ETA:6d23h43m
+Total train time:6d23h52m
+I1127 22:05:59.572687 137274321021824 utils.py:1231] [150] l2_params = 207.7270025501309
+I1127 22:05:59.573049 137274321021824 utils.py:1231] [150] train/loss = 6.895652115345001
+I1127 22:05:59.573174 137274321021824 utils.py:1231] [150] l2_grads = 0.624903678894043
+I1127 22:05:59.573244 137274321021824 utils.py:1231] [150] lr = 1.49e-05
+I1127 22:05:59.573301 137274321021824 utils.py:1231] [150] uptime = 948.9356624190041
+I1127 22:05:59.573358 137274321021824 utils.py:1231] [150] examples_seen = 153600.0
+I1127 22:05:59.573409 137274321021824 utils.py:1231] [150] progress = 0.0013321137092262196
+I1127 22:05:59.573462 137274321021824 utils.py:1231] [150] epoch = 0.11989069340687046
+I1127 22:05:59.573513 137274321021824 utils.py:1231] [150] img/sec/core = 170.9903417385835
+I1127 22:05:59.581576 137274321021824 utils.py:1231] [150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.22927563720416705
+I1127 22:05:59.581773 137274321021824 utils.py:1231] [150] core_hours = 0.22927563720416705
+I1127 22:05:59.581887 137274321021824 train.py:125] NOTE: Steps:150/112603 [0.1%]
+Walltime:15m49s (0s eval)
+ETA:7d6h12m
+Total train time:7d6h26m
+I1127 22:10:58.279794 137274321021824 utils.py:1231] [200] l2_params = 207.73410395098549
+I1127 22:10:58.280144 137274321021824 utils.py:1231] [200] train/loss = 6.8778902888298035
+I1127 22:10:58.280297 137274321021824 utils.py:1231] [200] l2_grads = 0.5932494401931763
+I1127 22:10:58.280376 137274321021824 utils.py:1231] [200] lr = 1.9900000000000003e-05
+I1127 22:10:58.280436 137274321021824 utils.py:1231] [200] uptime = 1247.6427969760043
+I1127 22:10:58.280492 137274321021824 utils.py:1231] [200] examples_seen = 204800.0
+I1127 22:10:58.280541 137274321021824 utils.py:1231] [200] progress = 0.001776151612301626
+I1127 22:10:58.280598 137274321021824 utils.py:1231] [200] epoch = 0.15985425787582727
+I1127 22:10:58.280652 137274321021824 utils.py:1231] [200] img/sec/core = 171.40534683221586
+I1127 22:10:58.280709 137274321021824 utils.py:1231] [200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.31224984124777827
+I1127 22:10:58.280756 137274321021824 utils.py:1231] [200] core_hours = 0.31224984124777827
+I1127 22:10:58.280819 137274321021824 train.py:125] NOTE: Steps:200/112603 [0.2%]
+Walltime:20m48s (0s eval)
+ETA:7d9h15m
+Total train time:7d9h34m
+I1127 22:16:05.591229 137274321021824 utils.py:1231] [250] l2_params = 207.74705519488188
+I1127 22:16:05.592353 137274321021824 utils.py:1231] [250] train/loss = 6.847655951976776
+I1127 22:16:05.592540 137274321021824 utils.py:1231] [250] l2_grads = 0.664771556854248
+I1127 22:16:05.592607 137274321021824 utils.py:1231] [250] lr = 2.49e-05
+I1127 22:16:05.592680 137274321021824 utils.py:1231] [250] uptime = 1554.9550419620064
+I1127 22:16:05.592737 137274321021824 utils.py:1231] [250] examples_seen = 256000.0
+I1127 22:16:05.592808 137274321021824 utils.py:1231] [250] progress = 0.0022201895153770327
+I1127 22:16:05.592859 137274321021824 utils.py:1231] [250] epoch = 0.1998178223447841
+I1127 22:16:05.592914 137274321021824 utils.py:1231] [250] img/sec/core = 166.60579210676144
+I1127 22:16:05.592977 137274321021824 utils.py:1231] [250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.3976143537438899
+I1127 22:16:05.593025 137274321021824 utils.py:1231] [250] core_hours = 0.3976143537438899
+I1127 22:16:05.593089 137274321021824 train.py:125] NOTE: Steps:250/112603 [0.2%]
+Walltime:25m55s (0s eval)
+ETA:7d12h8m
+Total train time:7d12h32m
+I1127 22:21:12.323120 137274321021824 utils.py:1231] [300] l2_params = 207.7565011336316
+I1127 22:21:12.323461 137274321021824 utils.py:1231] [300] train/loss = 6.828816175460815
+I1127 22:21:12.323624 137274321021824 utils.py:1231] [300] l2_grads = 0.609754204750061
+I1127 22:21:12.323709 137274321021824 utils.py:1231] [300] lr = 2.9900000000000002e-05
+I1127 22:21:12.323783 137274321021824 utils.py:1231] [300] uptime = 1861.6861436750041
+I1127 22:21:12.323863 137274321021824 utils.py:1231] [300] examples_seen = 307200.0
+I1127 22:21:12.323931 137274321021824 utils.py:1231] [300] progress = 0.0026642274184524393
+I1127 22:21:12.323988 137274321021824 utils.py:1231] [300] epoch = 0.23978138681374092
+I1127 22:21:12.324047 137274321021824 utils.py:1231] [300] img/sec/core = 166.92144915877108
+I1127 22:21:12.324110 137274321021824 utils.py:1231] [300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.48281743755305595
+I1127 22:21:12.324165 137274321021824 utils.py:1231] [300] core_hours = 0.48281743755305595
+I1127 22:21:12.324233 137274321021824 train.py:125] NOTE: Steps:300/112603 [0.3%]
+Walltime:31m2s (0s eval)
+ETA:7d13h57m
+Total train time:7d14h26m
+I1127 22:26:21.756703 137274321021824 utils.py:1231] [350] l2_params = 207.76526475960875
+I1127 22:26:21.756976 137274321021824 utils.py:1231] [350] train/loss = 6.785471975803375
+I1127 22:26:21.757179 137274321021824 utils.py:1231] [350] l2_grads = 0.6754442453384399
+I1127 22:26:21.757295 137274321021824 utils.py:1231] [350] lr = 3.49e-05
+I1127 22:26:21.757396 137274321021824 utils.py:1231] [350] uptime = 2171.1197443970013
+I1127 22:26:21.757480 137274321021824 utils.py:1231] [350] examples_seen = 358400.0
+I1127 22:26:21.757563 137274321021824 utils.py:1231] [350] progress = 0.003108265321527846
+I1127 22:26:21.757624 137274321021824 utils.py:1231] [350] epoch = 0.2797449512826977
+I1127 22:26:21.757702 137274321021824 utils.py:1231] [350] img/sec/core = 165.46360796156506
+I1127 22:26:21.757805 137274321021824 utils.py:1231] [350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.5687712155313885
+I1127 22:26:21.757872 137274321021824 utils.py:1231] [350] core_hours = 0.5687712155313885
+I1127 22:26:21.757956 137274321021824 train.py:125] NOTE: Steps:350/112603 [0.3%]
+Walltime:36m11s (0s eval)
+ETA:7d15h27m
+Total train time:7d16h2m
+I1127 22:31:33.568876 137274321021824 utils.py:1231] [400] l2_params = 207.77527953635794
+I1127 22:31:33.569155 137274321021824 utils.py:1231] [400] train/loss = 6.791165471076965
+I1127 22:31:33.569335 137274321021824 utils.py:1231] [400] l2_grads = 0.6224698424339294
+I1127 22:31:33.569420 137274321021824 utils.py:1231] [400] lr = 3.99e-05
+I1127 22:31:33.569509 137274321021824 utils.py:1231] [400] uptime = 2482.9318652940055
+I1127 22:31:33.569585 137274321021824 utils.py:1231] [400] examples_seen = 409600.0
+I1127 22:31:33.569664 137274321021824 utils.py:1231] [400] progress = 0.003552303224603252
+I1127 22:31:33.569734 137274321021824 utils.py:1231] [400] epoch = 0.31970851575165454
+I1127 22:31:33.569800 137274321021824 utils.py:1231] [400] img/sec/core = 164.20144237084375
+I1127 22:31:33.569876 137274321021824 utils.py:1231] [400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.6553856935583341
+I1127 22:31:33.569952 137274321021824 utils.py:1231] [400] core_hours = 0.6553856935583341
+I1127 22:31:33.570021 137274321021824 train.py:125] NOTE: Steps:400/112603 [0.4%]
+Walltime:41m23s (0s eval)
+ETA:7d16h45m
+Total train time:7d17h25m
+I1127 22:36:45.345535 137274321021824 utils.py:1231] [450] l2_params = 207.78328983158377
+I1127 22:36:45.345828 137274321021824 utils.py:1231] [450] train/loss = 6.783650696277618
+I1127 22:36:45.345985 137274321021824 utils.py:1231] [450] l2_grads = 0.8215329051017761
+I1127 22:36:45.346062 137274321021824 utils.py:1231] [450] lr = 4.49e-05
+I1127 22:36:45.346121 137274321021824 utils.py:1231] [450] uptime = 2794.708481599002
+I1127 22:36:45.346181 137274321021824 utils.py:1231] [450] examples_seen = 460800.0
+I1127 22:36:45.346237 137274321021824 utils.py:1231] [450] progress = 0.003996341127678659
+I1127 22:36:45.346293 137274321021824 utils.py:1231] [450] epoch = 0.35967208022061137
+I1127 22:36:45.346349 137274321021824 utils.py:1231] [450] img/sec/core = 164.2201413524658
+I1127 22:36:45.346409 137274321021824 utils.py:1231] [450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.7419903091986109
+I1127 22:36:45.346465 137274321021824 utils.py:1231] [450] core_hours = 0.7419903091986109
+I1127 22:36:45.346530 137274321021824 train.py:125] NOTE: Steps:450/112603 [0.4%]
+Walltime:46m35s (0s eval)
+ETA:7d17h45m
+Total train time:7d18h29m
+I1127 22:41:57.151323 137274321021824 utils.py:1231] [500] l2_params = 207.79007993992906
+I1127 22:41:57.151543 137274321021824 utils.py:1231] [500] train/loss = 6.704731106758118
+I1127 22:41:57.151649 137274321021824 utils.py:1231] [500] l2_grads = 0.7645918726921082
+I1127 22:41:57.151712 137274321021824 utils.py:1231] [500] lr = 4.99e-05
+I1127 22:41:57.151763 137274321021824 utils.py:1231] [500] uptime = 3106.5141257210053
+I1127 22:41:57.151814 137274321021824 utils.py:1231] [500] examples_seen = 512000.0
+I1127 22:41:57.151864 137274321021824 utils.py:1231] [500] progress = 0.0044403790307540655
+I1127 22:41:57.151916 137274321021824 utils.py:1231] [500] epoch = 0.3996356446895682
+I1127 22:41:57.151965 137274321021824 utils.py:1231] [500] img/sec/core = 164.2048531359056
+I1127 22:41:57.152019 137274321021824 utils.py:1231] [500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.8286029881213895
+I1127 22:41:57.152066 137274321021824 utils.py:1231] [500] core_hours = 0.8286029881213895
+I1127 22:41:57.152125 137274321021824 train.py:125] NOTE: Steps:500/112603 [0.4%]
+Walltime:51m47s (0s eval)
+ETA:7d18h31m
+Total train time:7d19h21m
+I1127 22:47:31.454324 137274321021824 utils.py:1231] [550] l2_params = 207.79588389057972
+I1127 22:47:31.454663 137274321021824 utils.py:1231] [550] train/loss = 6.698488473892212
+I1127 22:47:31.454876 137274321021824 utils.py:1231] [550] l2_grads = 0.7601398229598999
+I1127 22:47:31.454976 137274321021824 utils.py:1231] [550] lr = 5.49e-05
+I1127 22:47:31.455046 137274321021824 utils.py:1231] [550] uptime = 3440.8174063270053
+I1127 22:47:31.455113 137274321021824 utils.py:1231] [550] examples_seen = 563200.0
+I1127 22:47:31.455175 137274321021824 utils.py:1231] [550] progress = 0.004884416933829472
+I1127 22:47:31.455236 137274321021824 utils.py:1231] [550] epoch = 0.439599209158525
+I1127 22:47:31.455300 137274321021824 utils.py:1231] [550] img/sec/core = 153.15434508207176
+I1127 22:47:31.455367 137274321021824 utils.py:1231] [550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 0.9214650105119452
+I1127 22:47:31.455431 137274321021824 utils.py:1231] [550] core_hours = 0.9214650105119452
+I1127 22:47:31.455505 137274321021824 train.py:125] NOTE: Steps:550/112603 [0.5%]
+Walltime:57m21s (0s eval)
+ETA:7d20h25m
+Total train time:7d21h20m
+I1127 22:54:56.721297 137274321021824 utils.py:1231] [600] l2_params = 207.79270339023807
+I1127 22:54:56.721582 137274321021824 utils.py:1231] [600] train/loss = 6.653766095638275
+I1127 22:54:56.721740 137274321021824 utils.py:1231] [600] l2_grads = 0.9278506636619568
+I1127 22:54:56.721822 137274321021824 utils.py:1231] [600] lr = 5.9900000000000006e-05
+I1127 22:54:56.721878 137274321021824 utils.py:1231] [600] uptime = 3886.0842395590007
+I1127 22:54:56.721942 137274321021824 utils.py:1231] [600] examples_seen = 614400.0
+I1127 22:54:56.721994 137274321021824 utils.py:1231] [600] progress = 0.0053284548369048786
+I1127 22:54:56.722051 137274321021824 utils.py:1231] [600] epoch = 0.47956277362748184
+I1127 22:54:56.722104 137274321021824 utils.py:1231] [600] img/sec/core = 114.98723052952721
+I1127 22:54:56.722162 137274321021824 utils.py:1231] [600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.0451502419652772
+I1127 22:54:56.722215 137274321021824 utils.py:1231] [600] core_hours = 1.0451502419652772
+I1127 22:54:56.722290 137274321021824 train.py:125] NOTE: Steps:600/112603 [0.5%]
+Walltime:1h4m (0s eval)
+ETA:8d3h45m
+Total train time:8d4h48m
+I1127 23:01:06.842322 137274321021824 utils.py:1231] [650] l2_params = 207.7897209326118
+I1127 23:01:06.842595 137274321021824 utils.py:1231] [650] train/loss = 6.733096897602081
+I1127 23:01:06.842751 137274321021824 utils.py:1231] [650] l2_grads = 1.0579862594604492
+I1127 23:01:06.842834 137274321021824 utils.py:1231] [650] lr = 6.49e-05
+I1127 23:01:06.842898 137274321021824 utils.py:1231] [650] uptime = 4256.205249565006
+I1127 23:01:06.842952 137274321021824 utils.py:1231] [650] examples_seen = 665600.0
+I1127 23:01:06.843002 137274321021824 utils.py:1231] [650] progress = 0.005772492739980285
+I1127 23:01:06.843051 137274321021824 utils.py:1231] [650] epoch = 0.5195263380964387
+I1127 23:01:06.843101 137274321021824 utils.py:1231] [650] img/sec/core = 138.33313596320642
+I1127 23:01:06.843158 137274321021824 utils.py:1231] [650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.147961633633612
+I1127 23:01:06.843210 137274321021824 utils.py:1231] [650] core_hours = 1.147961633633612
+I1127 23:01:06.843278 137274321021824 train.py:125] NOTE: Steps:650/112603 [0.6%]
+Walltime:1h10m (0s eval)
+ETA:8d6h19m
+Total train time:8d7h28m
+I1127 23:06:18.687170 137274321021824 utils.py:1231] [700] l2_params = 207.77615965801652
+I1127 23:06:18.687424 137274321021824 utils.py:1231] [700] train/loss = 6.587809264659882
+I1127 23:06:18.687557 137274321021824 utils.py:1231] [700] l2_grads = 1.3097805976867676
+I1127 23:06:18.687627 137274321021824 utils.py:1231] [700] lr = 6.99e-05
+I1127 23:06:18.687678 137274321021824 utils.py:1231] [700] uptime = 4568.0500407490035
+I1127 23:06:18.687728 137274321021824 utils.py:1231] [700] examples_seen = 716800.0
+I1127 23:06:18.687774 137274321021824 utils.py:1231] [700] progress = 0.006216530643055692
+I1127 23:06:18.687820 137274321021824 utils.py:1231] [700] epoch = 0.5594899025653954
+I1127 23:06:18.687869 137274321021824 utils.py:1231] [700] img/sec/core = 164.18423987653043
+I1127 23:06:18.687928 137274321021824 utils.py:1231] [700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.2345851867402782
+I1127 23:06:18.687975 137274321021824 utils.py:1231] [700] core_hours = 1.2345851867402782
+I1127 23:06:18.688031 137274321021824 train.py:125] NOTE: Steps:700/112603 [0.6%]
+Walltime:1h16m (0s eval)
+ETA:8d5h55m
+Total train time:8d7h9m
+I1127 23:13:00.792864 137274321021824 utils.py:1231] [750] l2_params = 207.76650943558795
+I1127 23:13:00.793225 137274321021824 utils.py:1231] [750] train/loss = 6.590098857879639
+I1127 23:13:00.793406 137274321021824 utils.py:1231] [750] l2_grads = 1.2309309244155884
+I1127 23:13:00.793488 137274321021824 utils.py:1231] [750] lr = 7.489999999999999e-05
+I1127 23:13:00.793580 137274321021824 utils.py:1231] [750] uptime = 4970.155940940007
+I1127 23:13:00.793637 137274321021824 utils.py:1231] [750] examples_seen = 768000.0
+I1127 23:13:00.793692 137274321021824 utils.py:1231] [750] progress = 0.006660568546131097
+I1127 23:13:00.793744 137274321021824 utils.py:1231] [750] epoch = 0.5994534670343523
+I1127 23:13:00.793798 137274321021824 utils.py:1231] [750] img/sec/core = 127.3296412106353
+I1127 23:13:00.793871 137274321021824 utils.py:1231] [750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.3462812701266678
+I1127 23:13:00.793950 137274321021824 utils.py:1231] [750] core_hours = 1.3462812701266678
+I1127 23:13:00.794018 137274321021824 train.py:125] NOTE: Steps:750/112603 [0.7%]
+Walltime:1h22m (0s eval)
+ETA:8d9h19m
+Total train time:8d10h40m
+I1127 23:20:12.608306 137274321021824 utils.py:1231] [800] l2_params = 207.7551484633955
+I1127 23:20:12.608594 137274321021824 utils.py:1231] [800] train/loss = 6.539799153804779
+I1127 23:20:12.608778 137274321021824 utils.py:1231] [800] l2_grads = 1.0596776008605957
+I1127 23:20:12.608873 137274321021824 utils.py:1231] [800] lr = 7.99e-05
+I1127 23:20:12.608959 137274321021824 utils.py:1231] [800] uptime = 5401.9713164870045
+I1127 23:20:12.609026 137274321021824 utils.py:1231] [800] examples_seen = 819200.0
+I1127 23:20:12.609090 137274321021824 utils.py:1231] [800] progress = 0.007104606449206504
+I1127 23:20:12.609155 137274321021824 utils.py:1231] [800] epoch = 0.6394170315033091
+I1127 23:20:12.609220 137274321021824 utils.py:1231] [800] img/sec/core = 118.5691916021817
+I1127 23:20:12.609291 137274321021824 utils.py:1231] [800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.4662299855563894
+I1127 23:20:12.609361 137274321021824 utils.py:1231] [800] core_hours = 1.4662299855563894
+I1127 23:20:12.609438 137274321021824 train.py:125] NOTE: Steps:800/112603 [0.7%]
+Walltime:1h30m (0s eval)
+ETA:8d13h25m
+Total train time:8d14h53m
+I1127 23:27:52.181164 137274321021824 utils.py:1231] [850] l2_params = 207.73746241177625
+I1127 23:27:52.181538 137274321021824 utils.py:1231] [850] train/loss = 6.499816000461578
+I1127 23:27:52.181746 137274321021824 utils.py:1231] [850] l2_grads = 1.3088393211364746
+I1127 23:27:52.181850 137274321021824 utils.py:1231] [850] lr = 8.49e-05
+I1127 23:27:52.181946 137274321021824 utils.py:1231] [850] uptime = 5861.544298759007
+I1127 23:27:52.182066 137274321021824 utils.py:1231] [850] examples_seen = 870400.0
+I1127 23:27:52.182167 137274321021824 utils.py:1231] [850] progress = 0.00754864435228191
+I1127 23:27:52.182274 137274321021824 utils.py:1231] [850] epoch = 0.679380595972266
+I1127 23:27:52.182389 137274321021824 utils.py:1231] [850] img/sec/core = 111.4077675908651
+I1127 23:27:52.182492 137274321021824 utils.py:1231] [850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.593889147298612
+I1127 23:27:52.182577 137274321021824 utils.py:1231] [850] core_hours = 1.593889147298612
+I1127 23:27:52.182675 137274321021824 train.py:125] NOTE: Steps:850/112603 [0.8%]
+Walltime:1h37m (0s eval)
+ETA:8d18h2m
+Total train time:8d19h38m
+I1127 23:35:24.598136 137274321021824 utils.py:1231] [900] l2_params = 207.71625120760436
+I1127 23:35:24.598461 137274321021824 utils.py:1231] [900] train/loss = 6.470226347446442
+I1127 23:35:24.598605 137274321021824 utils.py:1231] [900] l2_grads = 1.5435563325881958
+I1127 23:35:24.598684 137274321021824 utils.py:1231] [900] lr = 8.989999999999999e-05
+I1127 23:35:24.598754 137274321021824 utils.py:1231] [900] uptime = 6313.961116041006
+I1127 23:35:24.598818 137274321021824 utils.py:1231] [900] examples_seen = 921600.0
+I1127 23:35:24.598879 137274321021824 utils.py:1231] [900] progress = 0.007992682255357318
+I1127 23:35:24.598948 137274321021824 utils.py:1231] [900] epoch = 0.7193441604412227
+I1127 23:35:24.599020 137274321021824 utils.py:1231] [900] img/sec/core = 113.1699752179773
+I1127 23:35:24.599086 137274321021824 utils.py:1231] [900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.719560485432501
+I1127 23:35:24.599143 137274321021824 utils.py:1231] [900] core_hours = 1.719560485432501
+I1127 23:35:24.599213 137274321021824 train.py:125] NOTE: Steps:900/112603 [0.8%]
+Walltime:1h45m (0s eval)
+ETA:8d21h53m
+Total train time:8d23h37m
+I1127 23:42:51.874147 137274321021824 utils.py:1231] [950] l2_params = 207.68659764895742
+I1127 23:42:51.874503 137274321021824 utils.py:1231] [950] train/loss = 6.487657129764557
+I1127 23:42:51.874706 137274321021824 utils.py:1231] [950] l2_grads = 1.2876965999603271
+I1127 23:42:51.874822 137274321021824 utils.py:1231] [950] lr = 9.49e-05
+I1127 23:42:51.874920 137274321021824 utils.py:1231] [950] uptime = 6761.237278460001
+I1127 23:42:51.875008 137274321021824 utils.py:1231] [950] examples_seen = 972800.0
+I1127 23:42:51.875103 137274321021824 utils.py:1231] [950] progress = 0.008436720158432724
+I1127 23:42:51.875180 137274321021824 utils.py:1231] [950] epoch = 0.7593077249101795
+I1127 23:42:51.875257 137274321021824 utils.py:1231] [950] img/sec/core = 114.4706655572611
+I1127 23:42:51.875342 137274321021824 utils.py:1231] [950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.8438038638822218
+I1127 23:42:51.875432 137274321021824 utils.py:1231] [950] core_hours = 1.8438038638822218
+I1127 23:42:51.875537 137274321021824 train.py:125] NOTE: Steps:950/112603 [0.8%]
+Walltime:1h52m (0s eval)
+ETA:9d1h9m
+Total train time:9d3h0m
+I1127 23:50:48.415564 137274321021824 utils.py:1231] [1000] l2_params = 207.65768646187283
+I1127 23:50:48.415898 137274321021824 utils.py:1231] [1000] train/loss = 6.7580578327178955
+I1127 23:50:48.416078 137274321021824 utils.py:1231] [1000] l2_grads = 1.1302751302719116
+I1127 23:50:48.416161 137274321021824 utils.py:1231] [1000] lr = 9.99e-05
+I1127 23:50:48.416224 137274321021824 utils.py:1231] [1000] uptime = 7237.778585568005
+I1127 23:50:48.416287 137274321021824 utils.py:1231] [1000] examples_seen = 1024000.0
+I1127 23:50:48.416343 137274321021824 utils.py:1231] [1000] progress = 0.008880758061508131
+I1127 23:50:48.416397 137274321021824 utils.py:1231] [1000] epoch = 0.7992712893791364
+I1127 23:50:48.416462 137274321021824 utils.py:1231] [1000] img/sec/core = 107.44084350361673
+I1127 23:50:48.416526 137274321021824 utils.py:1231] [1000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 1.9761764491900005
+I1127 23:50:48.416584 137274321021824 utils.py:1231] [1000] core_hours = 1.9761764491900005
+I1127 23:50:48.416651 137274321021824 train.py:125] NOTE: Steps:1000/112603 [0.9%]
+Walltime:2h0m (0s eval)
+ETA:9d4h59m
+Total train time:9d6h58m
+I1127 23:58:50.823723 137274321021824 utils.py:1231] [1050] l2_params = 207.6260948337967
+I1127 23:58:50.824141 137274321021824 utils.py:1231] [1050] train/loss = 6.763529419898987
+I1127 23:58:50.824344 137274321021824 utils.py:1231] [1050] l2_grads = 2.0321569442749023
+I1127 23:58:50.824424 137274321021824 utils.py:1231] [1050] lr = 0.0001049
+I1127 23:58:50.824487 137274321021824 utils.py:1231] [1050] uptime = 7720.186849218
+I1127 23:58:50.824562 137274321021824 utils.py:1231] [1050] examples_seen = 1075200.0
+I1127 23:58:50.824614 137274321021824 utils.py:1231] [1050] progress = 0.009324795964583537
+I1127 23:58:50.824667 137274321021824 utils.py:1231] [1050] epoch = 0.8392348538480932
+I1127 23:58:50.824719 137274321021824 utils.py:1231] [1050] img/sec/core = 106.13416862433242
+I1127 23:58:50.824778 137274321021824 utils.py:1231] [1050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.1101787446483327
+I1127 23:58:50.824831 137274321021824 utils.py:1231] [1050] core_hours = 2.1101787446483327
+I1127 23:58:50.824902 137274321021824 train.py:125] NOTE: Steps:1050/112603 [0.9%]
+Walltime:2h8m (0s eval)
+ETA:9d8h36m
+Total train time:9d10h43m
+I1128 00:05:55.433187 137274321021824 utils.py:1231] [1100] l2_params = 207.60244732521355
+I1128 00:05:55.433564 137274321021824 utils.py:1231] [1100] train/loss = 6.701339960098267
+I1128 00:05:55.433721 137274321021824 utils.py:1231] [1100] l2_grads = 1.3682069778442383
+I1128 00:05:55.433809 137274321021824 utils.py:1231] [1100] lr = 0.0001099
+I1128 00:05:55.433875 137274321021824 utils.py:1231] [1100] uptime = 8144.7962345880005
+I1128 00:05:55.433947 137274321021824 utils.py:1231] [1100] examples_seen = 1126400.0
+I1128 00:05:55.434006 137274321021824 utils.py:1231] [1100] progress = 0.009768833867658944
+I1128 00:05:55.434064 137274321021824 utils.py:1231] [1100] epoch = 0.87919841831705
+I1128 00:05:55.434124 137274321021824 utils.py:1231] [1100] img/sec/core = 120.58141379843696
+I1128 00:05:55.434191 137274321021824 utils.py:1231] [1100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.228125796139999
+I1128 00:05:55.434249 137274321021824 utils.py:1231] [1100] core_hours = 2.228125796139999
+I1128 00:05:55.434327 137274321021824 train.py:125] NOTE: Steps:1100/112603 [1.0%]
+Walltime:2h15m (0s eval)
+ETA:9d10h16m
+Total train time:9d12h30m
+I1128 00:12:46.566779 137274321021824 utils.py:1231] [1150] l2_params = 207.57732299623007
+I1128 00:12:46.567140 137274321021824 utils.py:1231] [1150] train/loss = 6.459486126899719
+I1128 00:12:46.567326 137274321021824 utils.py:1231] [1150] l2_grads = 1.3197449445724487
+I1128 00:12:46.567397 137274321021824 utils.py:1231] [1150] lr = 0.0001149
+I1128 00:12:46.567457 137274321021824 utils.py:1231] [1150] uptime = 8555.929818535005
+I1128 00:12:46.567514 137274321021824 utils.py:1231] [1150] examples_seen = 1177600.0
+I1128 00:12:46.567565 137274321021824 utils.py:1231] [1150] progress = 0.01021287177073435
+I1128 00:12:46.567615 137274321021824 utils.py:1231] [1150] epoch = 0.9191619827860068
+I1128 00:12:46.567668 137274321021824 utils.py:1231] [1150] img/sec/core = 124.53373307153542
+I1128 00:12:46.567730 137274321021824 utils.py:1231] [1150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.3423295694586117
+I1128 00:12:46.567781 137274321021824 utils.py:1231] [1150] core_hours = 2.3423295694586117
+I1128 00:12:46.567845 137274321021824 train.py:125] NOTE: Steps:1150/112603 [1.0%]
+Walltime:2h22m (0s eval)
+ETA:9d11h24m
+Total train time:9d13h45m
+I1128 00:19:43.339123 137274321021824 utils.py:1231] [1200] l2_params = 207.55305156635518
+I1128 00:19:43.339637 137274321021824 utils.py:1231] [1200] train/loss = 6.483503639698029
+I1128 00:19:43.339931 137274321021824 utils.py:1231] [1200] l2_grads = 1.108984351158142
+I1128 00:19:43.340048 137274321021824 utils.py:1231] [1200] lr = 0.00011990000000000001
+I1128 00:19:43.340174 137274321021824 utils.py:1231] [1200] uptime = 8972.702521565006
+I1128 00:19:43.340306 137274321021824 utils.py:1231] [1200] examples_seen = 1228800.0
+I1128 00:19:43.340387 137274321021824 utils.py:1231] [1200] progress = 0.010656909673809757
+I1128 00:19:43.340456 137274321021824 utils.py:1231] [1200] epoch = 0.9591255472549637
+I1128 00:19:43.340530 137274321021824 utils.py:1231] [1200] img/sec/core = 122.84873656016381
+I1128 00:19:43.340640 137274321021824 utils.py:1231] [1200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.4580997647447234
+I1128 00:19:43.340711 137274321021824 utils.py:1231] [1200] core_hours = 2.4580997647447234
+I1128 00:19:43.340792 137274321021824 train.py:125] NOTE: Steps:1200/112603 [1.1%]
+Walltime:2h29m (0s eval)
+ETA:9d12h34m
+Total train time:9d15h2m
+I1128 00:27:43.765471 137274321021824 utils.py:1231] [1250] l2_params = 207.53805502968402
+I1128 00:27:43.765803 137274321021824 utils.py:1231] [1250] train/loss = 6.374241292476654
+I1128 00:27:43.765994 137274321021824 utils.py:1231] [1250] l2_grads = 1.4652392864227295
+I1128 00:27:43.766076 137274321021824 utils.py:1231] [1250] lr = 0.0001249
+I1128 00:27:43.766141 137274321021824 utils.py:1231] [1250] uptime = 9453.128501962005
+I1128 00:27:43.766208 137274321021824 utils.py:1231] [1250] examples_seen = 1280000.0
+I1128 00:27:43.766269 137274321021824 utils.py:1231] [1250] progress = 0.011100947576885163
+I1128 00:27:43.766328 137274321021824 utils.py:1231] [1250] epoch = 0.9990891117239205
+I1128 00:27:43.766392 137274321021824 utils.py:1231] [1250] img/sec/core = 106.57208829066882
+I1128 00:27:43.766459 137274321021824 utils.py:1231] [1250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.5915514259661117
+I1128 00:27:43.766518 137274321021824 utils.py:1231] [1250] core_hours = 2.5915514259661117
+I1128 00:27:43.766587 137274321021824 train.py:125] NOTE: Steps:1250/112603 [1.1%]
+Walltime:2h37m (0s eval)
+ETA:9d15h13m
+Total train time:9d17h49m
+I1128 00:33:40.701765 137274321021824 utils.py:1231] [1300] l2_params = 207.51472465655615
+I1128 00:33:40.702080 137274321021824 utils.py:1231] [1300] train/loss = 6.408080995082855
+I1128 00:33:40.702234 137274321021824 utils.py:1231] [1300] l2_grads = 1.2163749933242798
+I1128 00:33:40.702311 137274321021824 utils.py:1231] [1300] lr = 0.00012989999999999999
+I1128 00:33:40.702374 137274321021824 utils.py:1231] [1300] uptime = 9810.064730899001
+I1128 00:33:40.702436 137274321021824 utils.py:1231] [1300] examples_seen = 1331200.0
+I1128 00:33:40.702495 137274321021824 utils.py:1231] [1300] progress = 0.01154498547996057
+I1128 00:33:40.702547 137274321021824 utils.py:1231] [1300] epoch = 1.0390526761928773
+I1128 00:33:40.702597 137274321021824 utils.py:1231] [1300] img/sec/core = 143.4429902296004
+I1128 00:33:40.702654 137274321021824 utils.py:1231] [1300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.6907003784486108
+I1128 00:33:40.702706 137274321021824 utils.py:1231] [1300] core_hours = 2.6907003784486108
+I1128 00:33:40.702769 137274321021824 train.py:125] NOTE: Steps:1300/112603 [1.2%]
+Walltime:2h43m (0s eval)
+ETA:9d14h43m
+Total train time:9d17h25m
+I1128 00:39:26.817270 137274321021824 utils.py:1231] [1350] l2_params = 207.48464312073216
+I1128 00:39:26.817578 137274321021824 utils.py:1231] [1350] train/loss = 6.366620421409607
+I1128 00:39:26.817758 137274321021824 utils.py:1231] [1350] l2_grads = 1.726142406463623
+I1128 00:39:26.817841 137274321021824 utils.py:1231] [1350] lr = 0.0001349
+I1128 00:39:26.817922 137274321021824 utils.py:1231] [1350] uptime = 10156.180283673006
+I1128 00:39:26.817978 137274321021824 utils.py:1231] [1350] examples_seen = 1382400.0
+I1128 00:39:26.818030 137274321021824 utils.py:1231] [1350] progress = 0.011989023383035976
+I1128 00:39:26.818079 137274321021824 utils.py:1231] [1350] epoch = 1.079016240661834
+I1128 00:39:26.818130 137274321021824 utils.py:1231] [1350] img/sec/core = 147.92747563537228
+I1128 00:39:26.818190 137274321021824 utils.py:1231] [1350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.786843587552501
+I1128 00:39:26.818241 137274321021824 utils.py:1231] [1350] core_hours = 2.786843587552501
+I1128 00:39:26.818304 137274321021824 train.py:125] NOTE: Steps:1350/112603 [1.2%]
+Walltime:2h49m (0s eval)
+ETA:9d14h0m
+Total train time:9d16h47m
+I1128 00:45:11.995209 137274321021824 utils.py:1231] [1400] l2_params = 207.45982386375564
+I1128 00:45:11.995480 137274321021824 utils.py:1231] [1400] train/loss = 6.255841851234436
+I1128 00:45:11.995601 137274321021824 utils.py:1231] [1400] l2_grads = 1.5057644844055176
+I1128 00:45:11.995674 137274321021824 utils.py:1231] [1400] lr = 0.0001399
+I1128 00:45:11.995740 137274321021824 utils.py:1231] [1400] uptime = 10501.358100953003
+I1128 00:45:11.995820 137274321021824 utils.py:1231] [1400] examples_seen = 1433600.0
+I1128 00:45:11.995907 137274321021824 utils.py:1231] [1400] progress = 0.012433061286111383
+I1128 00:45:11.995988 137274321021824 utils.py:1231] [1400] epoch = 1.1189798051307909
+I1128 00:45:11.996048 137274321021824 utils.py:1231] [1400] img/sec/core = 148.3293463162155
+I1128 00:45:11.996112 137274321021824 utils.py:1231] [1400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.882726314574722
+I1128 00:45:11.996169 137274321021824 utils.py:1231] [1400] core_hours = 2.882726314574722
+I1128 00:45:11.996237 137274321021824 train.py:125] NOTE: Steps:1400/112603 [1.2%]
+Walltime:2h55m (0s eval)
+ETA:9d13h18m
+Total train time:9d16h11m
+I1128 00:50:55.753236 137274321021824 utils.py:1231] [1450] l2_params = 207.44141633080247
+I1128 00:50:55.753543 137274321021824 utils.py:1231] [1450] train/loss = 6.718242585659027
+I1128 00:50:55.753711 137274321021824 utils.py:1231] [1450] l2_grads = 1.0811831951141357
+I1128 00:50:55.753798 137274321021824 utils.py:1231] [1450] lr = 0.0001449
+I1128 00:50:55.753861 137274321021824 utils.py:1231] [1450] uptime = 10845.116222141005
+I1128 00:50:55.753934 137274321021824 utils.py:1231] [1450] examples_seen = 1484800.0
+I1128 00:50:55.754008 137274321021824 utils.py:1231] [1450] progress = 0.012877099189186789
+I1128 00:50:55.754081 137274321021824 utils.py:1231] [1450] epoch = 1.1589433695997478
+I1128 00:50:55.754144 137274321021824 utils.py:1231] [1450] img/sec/core = 148.9419357513842
+I1128 00:50:55.754230 137274321021824 utils.py:1231] [1450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 2.9782146815713895
+I1128 00:50:55.754288 137274321021824 utils.py:1231] [1450] core_hours = 2.9782146815713895
+I1128 00:50:55.754357 137274321021824 train.py:125] NOTE: Steps:1450/112603 [1.3%]
+Walltime:3h0m (0s eval)
+ETA:9d12h37m
+Total train time:9d15h35m
+I1128 00:56:39.310253 137274321021824 utils.py:1231] [1500] l2_params = 207.41887292189125
+I1128 00:56:39.310533 137274321021824 utils.py:1231] [1500] train/loss = 6.311974346637726
+I1128 00:56:39.310684 137274321021824 utils.py:1231] [1500] l2_grads = 1.6365958452224731
+I1128 00:56:39.310748 137274321021824 utils.py:1231] [1500] lr = 0.0001499
+I1128 00:56:39.310802 137274321021824 utils.py:1231] [1500] uptime = 11188.673164077001
+I1128 00:56:39.310857 137274321021824 utils.py:1231] [1500] examples_seen = 1536000.0
+I1128 00:56:39.310917 137274321021824 utils.py:1231] [1500] progress = 0.013321137092262195
+I1128 00:56:39.310967 137274321021824 utils.py:1231] [1500] epoch = 1.1989069340687046
+I1128 00:56:39.311017 137274321021824 utils.py:1231] [1500] img/sec/core = 149.0291528137377
+I1128 00:56:39.311074 137274321021824 utils.py:1231] [1500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.0736471654425
+I1128 00:56:39.311122 137274321021824 utils.py:1231] [1500] core_hours = 3.0736471654425
+I1128 00:56:39.311183 137274321021824 train.py:125] NOTE: Steps:1500/112603 [1.3%]
+Walltime:3h6m (0s eval)
+ETA:9d11h57m
+Total train time:9d15h2m
+I1128 01:02:28.313591 137274321021824 utils.py:1231] [1550] l2_params = 207.39348108529677
+I1128 01:02:28.313916 137274321021824 utils.py:1231] [1550] train/loss = 6.322002172470093
+I1128 01:02:28.314093 137274321021824 utils.py:1231] [1550] l2_grads = 1.4804465770721436
+I1128 01:02:28.314211 137274321021824 utils.py:1231] [1550] lr = 0.00015490000000000002
+I1128 01:02:28.314280 137274321021824 utils.py:1231] [1550] uptime = 11537.676639811005
+I1128 01:02:28.314339 137274321021824 utils.py:1231] [1550] examples_seen = 1587200.0
+I1128 01:02:28.314392 137274321021824 utils.py:1231] [1550] progress = 0.013765174995337602
+I1128 01:02:28.314450 137274321021824 utils.py:1231] [1550] epoch = 1.2388704985376613
+I1128 01:02:28.314510 137274321021824 utils.py:1231] [1550] img/sec/core = 146.7034100228348
+I1128 01:02:28.314589 137274321021824 utils.py:1231] [1550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.1705925753686115
+I1128 01:02:28.314644 137274321021824 utils.py:1231] [1550] core_hours = 3.1705925753686115
+I1128 01:02:28.314723 137274321021824 train.py:125] NOTE: Steps:1550/112603 [1.4%]
+Walltime:3h12m (0s eval)
+ETA:9d11h27m
+Total train time:9d14h37m
+I1128 01:09:35.742632 137274321021824 utils.py:1231] [1600] l2_params = 207.3730930144862
+I1128 01:09:35.742987 137274321021824 utils.py:1231] [1600] train/loss = 6.275707244873047
+I1128 01:09:35.743200 137274321021824 utils.py:1231] [1600] l2_grads = 1.5797171592712402
+I1128 01:09:35.743353 137274321021824 utils.py:1231] [1600] lr = 0.00015989999999999998
+I1128 01:09:35.743452 137274321021824 utils.py:1231] [1600] uptime = 11965.105805746003
+I1128 01:09:35.743574 137274321021824 utils.py:1231] [1600] examples_seen = 1638400.0
+I1128 01:09:35.743684 137274321021824 utils.py:1231] [1600] progress = 0.014209212898413008
+I1128 01:09:35.743781 137274321021824 utils.py:1231] [1600] epoch = 1.2788340630066182
+I1128 01:09:35.743871 137274321021824 utils.py:1231] [1600] img/sec/core = 119.7859296475484
+I1128 01:09:35.743984 137274321021824 utils.py:1231] [1600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.2893228992394445
+I1128 01:09:35.744072 137274321021824 utils.py:1231] [1600] core_hours = 3.2893228992394445
+I1128 01:09:35.744179 137274321021824 train.py:125] NOTE: Steps:1600/112603 [1.4%]
+Walltime:3h19m (0s eval)
+ETA:9d12h29m
+Total train time:9d15h46m
+I1128 01:16:25.645528 137274321021824 utils.py:1231] [1650] l2_params = 207.34219168344475
+I1128 01:16:25.645855 137274321021824 utils.py:1231] [1650] train/loss = 6.1983612179756165
+I1128 01:16:25.646047 137274321021824 utils.py:1231] [1650] l2_grads = 1.2041071653366089
+I1128 01:16:25.646146 137274321021824 utils.py:1231] [1650] lr = 0.0001649
+I1128 01:16:25.646218 137274321021824 utils.py:1231] [1650] uptime = 12375.008578881003
+I1128 01:16:25.646281 137274321021824 utils.py:1231] [1650] examples_seen = 1689600.0
+I1128 01:16:25.646347 137274321021824 utils.py:1231] [1650] progress = 0.014653250801488415
+I1128 01:16:25.646405 137274321021824 utils.py:1231] [1650] epoch = 1.318797627475575
+I1128 01:16:25.646476 137274321021824 utils.py:1231] [1650] img/sec/core = 124.90766922217796
+I1128 01:16:25.646554 137274321021824 utils.py:1231] [1650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.4031847806658333
+I1128 01:16:25.646624 137274321021824 utils.py:1231] [1650] core_hours = 3.4031847806658333
+I1128 01:16:25.646692 137274321021824 train.py:125] NOTE: Steps:1650/112603 [1.5%]
+Walltime:3h26m (0s eval)
+ETA:9d13h7m
+Total train time:9d16h31m
+I1128 01:23:21.065704 137274321021824 utils.py:1231] [1700] l2_params = 207.3151576535917
+I1128 01:23:21.066060 137274321021824 utils.py:1231] [1700] train/loss = 6.270511269569397
+I1128 01:23:21.066247 137274321021824 utils.py:1231] [1700] l2_grads = 1.3517533540725708
+I1128 01:23:21.066343 137274321021824 utils.py:1231] [1700] lr = 0.0001699
+I1128 01:23:21.066404 137274321021824 utils.py:1231] [1700] uptime = 12790.428764201002
+I1128 01:23:21.066464 137274321021824 utils.py:1231] [1700] examples_seen = 1740800.0
+I1128 01:23:21.066522 137274321021824 utils.py:1231] [1700] progress = 0.01509728870456382
+I1128 01:23:21.066577 137274321021824 utils.py:1231] [1700] epoch = 1.358761191944532
+I1128 01:23:21.066635 137274321021824 utils.py:1231] [1700] img/sec/core = 123.24870530920519
+I1128 01:23:21.066699 137274321021824 utils.py:1231] [1700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.5185792765880555
+I1128 01:23:21.066762 137274321021824 utils.py:1231] [1700] core_hours = 3.5185792765880555
+I1128 01:23:21.066833 137274321021824 train.py:125] NOTE: Steps:1700/112603 [1.5%]
+Walltime:3h33m (0s eval)
+ETA:9d13h48m
+Total train time:9d17h20m
+I1128 01:30:33.557964 137274321021824 utils.py:1231] [1750] l2_params = 207.27943121688565
+I1128 01:30:33.558262 137274321021824 utils.py:1231] [1750] train/loss = 6.233032405376434
+I1128 01:30:33.558431 137274321021824 utils.py:1231] [1750] l2_grads = 1.7007148265838623
+I1128 01:30:33.558508 137274321021824 utils.py:1231] [1750] lr = 0.0001749
+I1128 01:30:33.558575 137274321021824 utils.py:1231] [1750] uptime = 13222.920935892005
+I1128 01:30:33.558641 137274321021824 utils.py:1231] [1750] examples_seen = 1792000.0
+I1128 01:30:33.558703 137274321021824 utils.py:1231] [1750] progress = 0.015541326607639228
+I1128 01:30:33.558764 137274321021824 utils.py:1231] [1750] epoch = 1.3987247564134886
+I1128 01:30:33.558826 137274321021824 utils.py:1231] [1750] img/sec/core = 118.3836456503083
+I1128 01:30:33.558904 137274321021824 utils.py:1231] [1750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.6387159909466673
+I1128 01:30:33.558965 137274321021824 utils.py:1231] [1750] core_hours = 3.6387159909466673
+I1128 01:30:33.559040 137274321021824 train.py:125] NOTE: Steps:1750/112603 [1.6%]
+Walltime:3h40m (0s eval)
+ETA:9d14h45m
+Total train time:9d18h23m
+I1128 01:38:49.056309 137274321021824 utils.py:1231] [1800] l2_params = 207.26007591526636
+I1128 01:38:49.056535 137274321021824 utils.py:1231] [1800] train/loss = 6.125150680541992
+I1128 01:38:49.056637 137274321021824 utils.py:1231] [1800] l2_grads = 1.5387941598892212
+I1128 01:38:49.056700 137274321021824 utils.py:1231] [1800] lr = 0.0001799
+I1128 01:38:49.056754 137274321021824 utils.py:1231] [1800] uptime = 13718.419115913006
+I1128 01:38:49.056806 137274321021824 utils.py:1231] [1800] examples_seen = 1843200.0
+I1128 01:38:49.056855 137274321021824 utils.py:1231] [1800] progress = 0.015985364510714636
+I1128 01:38:49.056914 137274321021824 utils.py:1231] [1800] epoch = 1.4386883208824455
+I1128 01:38:49.056965 137274321021824 utils.py:1231] [1800] img/sec/core = 103.33034926148464
+I1128 01:38:49.057022 137274321021824 utils.py:1231] [1800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.7763543742858343
+I1128 01:38:49.057100 137274321021824 utils.py:1231] [1800] core_hours = 3.7763543742858343
+I1128 01:38:49.057167 137274321021824 train.py:125] NOTE: Steps:1800/112603 [1.6%]
+Walltime:3h48m (0s eval)
+ETA:9d16h43m
+Total train time:9d20h30m
+I1128 01:45:11.332852 137274321021824 utils.py:1231] [1850] l2_params = 207.22998261993223
+I1128 01:45:11.333223 137274321021824 utils.py:1231] [1850] train/loss = 6.112380564212799
+I1128 01:45:11.333401 137274321021824 utils.py:1231] [1850] l2_grads = 1.3338249921798706
+I1128 01:45:11.333470 137274321021824 utils.py:1231] [1850] lr = 0.00018490000000000002
+I1128 01:45:11.333527 137274321021824 utils.py:1231] [1850] uptime = 14100.695887994007
+I1128 01:45:11.333583 137274321021824 utils.py:1231] [1850] examples_seen = 1894400.0
+I1128 01:45:11.333636 137274321021824 utils.py:1231] [1850] progress = 0.01642940241379004
+I1128 01:45:11.333688 137274321021824 utils.py:1231] [1850] epoch = 1.4786518853514024
+I1128 01:45:11.333742 137274321021824 utils.py:1231] [1850] img/sec/core = 133.93437357253615
+I1128 01:45:11.333811 137274321021824 utils.py:1231] [1850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 3.8825423665305565
+I1128 01:45:11.333863 137274321021824 utils.py:1231] [1850] core_hours = 3.8825423665305565
+I1128 01:45:11.333943 137274321021824 train.py:125] NOTE: Steps:1850/112603 [1.6%]
+Walltime:3h55m (0s eval)
+ETA:9d16h41m
+Total train time:9d20h34m
+I1128 01:52:40.351019 137274321021824 utils.py:1231] [1900] l2_params = 207.2118821790164
+I1128 01:52:40.351297 137274321021824 utils.py:1231] [1900] train/loss = 6.361689031124115
+I1128 01:52:40.351412 137274321021824 utils.py:1231] [1900] l2_grads = 1.4877737760543823
+I1128 01:52:40.351500 137274321021824 utils.py:1231] [1900] lr = 0.0001899
+I1128 01:52:40.351563 137274321021824 utils.py:1231] [1900] uptime = 14549.713924040007
+I1128 01:52:40.351650 137274321021824 utils.py:1231] [1900] examples_seen = 1945600.0
+I1128 01:52:40.351707 137274321021824 utils.py:1231] [1900] progress = 0.016873440316865447
+I1128 01:52:40.351757 137274321021824 utils.py:1231] [1900] epoch = 1.518615449820359
+I1128 01:52:40.351811 137274321021824 utils.py:1231] [1900] img/sec/core = 114.02660002449164
+I1128 01:52:40.351869 137274321021824 utils.py:1231] [1900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.007269598765556
+I1128 01:52:40.351933 137274321021824 utils.py:1231] [1900] core_hours = 4.007269598765556
+I1128 01:52:40.351995 137274321021824 train.py:125] NOTE: Steps:1900/112603 [1.7%]
+Walltime:4h2m (0s eval)
+ETA:9d17h43m
+Total train time:9d21h44m
+I1128 01:58:12.660506 137274321021824 utils.py:1231] [1950] l2_params = 207.20008097193164
+I1128 01:58:12.660737 137274321021824 utils.py:1231] [1950] train/loss = 6.143209993839264
+I1128 01:58:12.660845 137274321021824 utils.py:1231] [1950] l2_grads = 1.6467852592468262
+I1128 01:58:12.660926 137274321021824 utils.py:1231] [1950] lr = 0.0001949
+I1128 01:58:12.660987 137274321021824 utils.py:1231] [1950] uptime = 14882.023348463
+I1128 01:58:12.661047 137274321021824 utils.py:1231] [1950] examples_seen = 1996800.0
+I1128 01:58:12.661104 137274321021824 utils.py:1231] [1950] progress = 0.017317478219940854
+I1128 01:58:12.661164 137274321021824 utils.py:1231] [1950] epoch = 1.558579014289316
+I1128 01:58:12.661222 137274321021824 utils.py:1231] [1950] img/sec/core = 154.07327098501997
+I1128 01:58:12.661285 137274321021824 utils.py:1231] [1950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.099577772216389
+I1128 01:58:12.661343 137274321021824 utils.py:1231] [1950] core_hours = 4.099577772216389
+I1128 01:58:12.661410 137274321021824 train.py:125] NOTE: Steps:1950/112603 [1.7%]
+Walltime:4h8m (0s eval)
+ETA:9d16h52m
+Total train time:9d20h58m
+I1128 02:03:28.924434 137274321021824 utils.py:1231] [2000] l2_params = 207.1952094067897
+I1128 02:03:28.924708 137274321021824 utils.py:1231] [2000] train/loss = 6.1439860463142395
+I1128 02:03:28.924838 137274321021824 utils.py:1231] [2000] l2_grads = 1.655167818069458
+I1128 02:03:28.924926 137274321021824 utils.py:1231] [2000] lr = 0.0001999
+I1128 02:03:28.924982 137274321021824 utils.py:1231] [2000] uptime = 15198.287344043005
+I1128 02:03:28.925045 137274321021824 utils.py:1231] [2000] examples_seen = 2048000.0
+I1128 02:03:28.925096 137274321021824 utils.py:1231] [2000] progress = 0.017761516123016262
+I1128 02:03:28.925148 137274321021824 utils.py:1231] [2000] epoch = 1.5985425787582728
+I1128 02:03:28.925209 137274321021824 utils.py:1231] [2000] img/sec/core = 161.89006878921862
+I1128 02:03:28.925269 137274321021824 utils.py:1231] [2000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.187428882099723
+I1128 02:03:28.925330 137274321021824 utils.py:1231] [2000] core_hours = 4.187428882099723
+I1128 02:03:28.925405 137274321021824 train.py:125] NOTE: Steps:2000/112603 [1.8%]
+Walltime:4h13m (0s eval)
+ETA:9d15h48m
+Total train time:9d19h59m
+I1128 02:08:46.348965 137274321021824 utils.py:1231] [2050] l2_params = 207.18143717034707
+I1128 02:08:46.349179 137274321021824 utils.py:1231] [2050] train/loss = 6.019332587718964
+I1128 02:08:46.349290 137274321021824 utils.py:1231] [2050] l2_grads = 1.6528494358062744
+I1128 02:08:46.349367 137274321021824 utils.py:1231] [2050] lr = 0.0002049
+I1128 02:08:46.349437 137274321021824 utils.py:1231] [2050] uptime = 15515.711797928001
+I1128 02:08:46.349509 137274321021824 utils.py:1231] [2050] examples_seen = 2099200.0
+I1128 02:08:46.349567 137274321021824 utils.py:1231] [2050] progress = 0.018205554026091666
+I1128 02:08:46.349627 137274321021824 utils.py:1231] [2050] epoch = 1.6385061432272296
+I1128 02:08:46.349685 137274321021824 utils.py:1231] [2050] img/sec/core = 161.29822190243073
+I1128 02:08:46.349752 137274321021824 utils.py:1231] [2050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.275602341512222
+I1128 02:08:46.349817 137274321021824 utils.py:1231] [2050] core_hours = 4.275602341512222
+I1128 02:08:46.349910 137274321021824 train.py:125] NOTE: Steps:2050/112603 [1.8%]
+Walltime:4h18m (0s eval)
+ETA:9d14h48m
+Total train time:9d19h4m
+I1128 02:13:58.194667 137274321021824 utils.py:1231] [2100] l2_params = 207.17265938288506
+I1128 02:13:58.194897 137274321021824 utils.py:1231] [2100] train/loss = 6.36241602897644
+I1128 02:13:58.195001 137274321021824 utils.py:1231] [2100] l2_grads = 1.6057418584823608
+I1128 02:13:58.195066 137274321021824 utils.py:1231] [2100] lr = 0.0002099
+I1128 02:13:58.195126 137274321021824 utils.py:1231] [2100] uptime = 15827.557488217004
+I1128 02:13:58.195191 137274321021824 utils.py:1231] [2100] examples_seen = 2150400.0
+I1128 02:13:58.195245 137274321021824 utils.py:1231] [2100] progress = 0.018649591929167073
+I1128 02:13:58.195301 137274321021824 utils.py:1231] [2100] epoch = 1.6784697076961863
+I1128 02:13:58.195354 137274321021824 utils.py:1231] [2100] img/sec/core = 164.18376650499934
+I1128 02:13:58.195415 137274321021824 utils.py:1231] [2100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.362226144370278
+I1128 02:13:58.195469 137274321021824 utils.py:1231] [2100] core_hours = 4.362226144370278
+I1128 02:13:58.195533 137274321021824 train.py:125] NOTE: Steps:2100/112603 [1.9%]
+Walltime:4h23m (0s eval)
+ETA:9d13h45m
+Total train time:9d18h7m
+I1128 02:19:10.036418 137274321021824 utils.py:1231] [2150] l2_params = 207.1646477248363
+I1128 02:19:10.036759 137274321021824 utils.py:1231] [2150] train/loss = 6.034305214881897
+I1128 02:19:10.036936 137274321021824 utils.py:1231] [2150] l2_grads = 1.5363245010375977
+I1128 02:19:10.037014 137274321021824 utils.py:1231] [2150] lr = 0.00021490000000000002
+I1128 02:19:10.037088 137274321021824 utils.py:1231] [2150] uptime = 16139.399447122007
+I1128 02:19:10.037143 137274321021824 utils.py:1231] [2150] examples_seen = 2201600.0
+I1128 02:19:10.037194 137274321021824 utils.py:1231] [2150] progress = 0.01909362983224248
+I1128 02:19:10.037242 137274321021824 utils.py:1231] [2150] epoch = 1.7184332721651432
+I1128 02:19:10.037317 137274321021824 utils.py:1231] [2150] img/sec/core = 164.18573106641242
+I1128 02:19:10.037377 137274321021824 utils.py:1231] [2150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.448848910732779
+I1128 02:19:10.037430 137274321021824 utils.py:1231] [2150] core_hours = 4.448848910732779
+I1128 02:19:10.037489 137274321021824 train.py:125] NOTE: Steps:2150/112603 [1.9%]
+Walltime:4h28m (0s eval)
+ETA:9d12h45m
+Total train time:9d17h13m
+I1128 02:24:21.867450 137274321021824 utils.py:1231] [2200] l2_params = 207.17710374624775
+I1128 02:24:21.867678 137274321021824 utils.py:1231] [2200] train/loss = 6.034717500209808
+I1128 02:24:21.867777 137274321021824 utils.py:1231] [2200] l2_grads = 1.7383131980895996
+I1128 02:24:21.867843 137274321021824 utils.py:1231] [2200] lr = 0.0002199
+I1128 02:24:21.867901 137274321021824 utils.py:1231] [2200] uptime = 16451.230262622004
+I1128 02:24:21.867954 137274321021824 utils.py:1231] [2200] examples_seen = 2252800.0
+I1128 02:24:21.868004 137274321021824 utils.py:1231] [2200] progress = 0.019537667735317888
+I1128 02:24:21.868052 137274321021824 utils.py:1231] [2200] epoch = 1.7583968366341
+I1128 02:24:21.868102 137274321021824 utils.py:1231] [2200] img/sec/core = 164.191598312388
+I1128 02:24:21.868157 137274321021824 utils.py:1231] [2200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.5354685817050004
+I1128 02:24:21.868207 137274321021824 utils.py:1231] [2200] core_hours = 4.5354685817050004
+I1128 02:24:21.868268 137274321021824 train.py:125] NOTE: Steps:2200/112603 [2.0%]
+Walltime:4h34m (0s eval)
+ETA:9d11h48m
+Total train time:9d16h21m
+I1128 02:29:33.673537 137274321021824 utils.py:1231] [2250] l2_params = 207.18169355325588
+I1128 02:29:33.673834 137274321021824 utils.py:1231] [2250] train/loss = 6.008518576622009
+I1128 02:29:33.674026 137274321021824 utils.py:1231] [2250] l2_grads = 1.5938533544540405
+I1128 02:29:33.674125 137274321021824 utils.py:1231] [2250] lr = 0.0002249
+I1128 02:29:33.674201 137274321021824 utils.py:1231] [2250] uptime = 16763.036562411005
+I1128 02:29:33.674268 137274321021824 utils.py:1231] [2250] examples_seen = 2304000.0
+I1128 02:29:33.674343 137274321021824 utils.py:1231] [2250] progress = 0.019981705638393292
+I1128 02:29:33.674412 137274321021824 utils.py:1231] [2250] epoch = 1.7983604011030567
+I1128 02:29:33.674479 137274321021824 utils.py:1231] [2250] img/sec/core = 164.20450784556624
+I1128 02:29:33.674543 137274321021824 utils.py:1231] [2250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.622081442757501
+I1128 02:29:33.674605 137274321021824 utils.py:1231] [2250] core_hours = 4.622081442757501
+I1128 02:29:33.674697 137274321021824 train.py:125] NOTE: Steps:2250/112603 [2.0%]
+Walltime:4h39m (0s eval)
+ETA:9d10h53m
+Total train time:9d15h31m
+I1128 02:34:45.457891 137274321021824 utils.py:1231] [2300] l2_params = 207.18389839068232
+I1128 02:34:45.458157 137274321021824 utils.py:1231] [2300] train/loss = 6.650797367095947
+I1128 02:34:45.458338 137274321021824 utils.py:1231] [2300] l2_grads = 1.316536784172058
+I1128 02:34:45.458422 137274321021824 utils.py:1231] [2300] lr = 0.0002299
+I1128 02:34:45.458508 137274321021824 utils.py:1231] [2300] uptime = 17074.82086916
+I1128 02:34:45.458575 137274321021824 utils.py:1231] [2300] examples_seen = 2355200.0
+I1128 02:34:45.458649 137274321021824 utils.py:1231] [2300] progress = 0.0204257435414687
+I1128 02:34:45.458721 137274321021824 utils.py:1231] [2300] epoch = 1.8383239655720136
+I1128 02:34:45.458784 137274321021824 utils.py:1231] [2300] img/sec/core = 164.21609071305485
+I1128 02:34:45.458858 137274321021824 utils.py:1231] [2300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.708688194632222
+I1128 02:34:45.458921 137274321021824 utils.py:1231] [2300] core_hours = 4.708688194632222
+I1128 02:34:45.458989 137274321021824 train.py:125] NOTE: Steps:2300/112603 [2.0%]
+Walltime:4h44m (0s eval)
+ETA:9d10h0m
+Total train time:9d14h43m
+I1128 02:39:57.236364 137274321021824 utils.py:1231] [2350] l2_params = 207.21255889650226
+I1128 02:39:57.236603 137274321021824 utils.py:1231] [2350] train/loss = 6.020943224430084
+I1128 02:39:57.236727 137274321021824 utils.py:1231] [2350] l2_grads = 1.3464710712432861
+I1128 02:39:57.236806 137274321021824 utils.py:1231] [2350] lr = 0.0002349
+I1128 02:39:57.236858 137274321021824 utils.py:1231] [2350] uptime = 17386.599219810007
+I1128 02:39:57.236925 137274321021824 utils.py:1231] [2350] examples_seen = 2406400.0
+I1128 02:39:57.236975 137274321021824 utils.py:1231] [2350] progress = 0.020869781444544107
+I1128 02:39:57.237024 137274321021824 utils.py:1231] [2350] epoch = 1.8782875300409705
+I1128 02:39:57.237072 137274321021824 utils.py:1231] [2350] img/sec/core = 164.2192278368803
+I1128 02:39:57.237127 137274321021824 utils.py:1231] [2350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.795293292035001
+I1128 02:39:57.237176 137274321021824 utils.py:1231] [2350] core_hours = 4.795293292035001
+I1128 02:39:57.237236 137274321021824 train.py:125] NOTE: Steps:2350/112603 [2.1%]
+Walltime:4h49m (0s eval)
+ETA:9d9h10m
+Total train time:9d13h58m
+I1128 02:45:09.023460 137274321021824 utils.py:1231] [2400] l2_params = 207.23469161378694
+I1128 02:45:09.023673 137274321021824 utils.py:1231] [2400] train/loss = 5.943787753582001
+I1128 02:45:09.023766 137274321021824 utils.py:1231] [2400] l2_grads = 1.6398128271102905
+I1128 02:45:09.023838 137274321021824 utils.py:1231] [2400] lr = 0.0002399
+I1128 02:45:09.023907 137274321021824 utils.py:1231] [2400] uptime = 17698.386267566006
+I1128 02:45:09.023963 137274321021824 utils.py:1231] [2400] examples_seen = 2457600.0
+I1128 02:45:09.024011 137274321021824 utils.py:1231] [2400] progress = 0.021313819347619514
+I1128 02:45:09.024058 137274321021824 utils.py:1231] [2400] epoch = 1.9182510945099274
+I1128 02:45:09.024109 137274321021824 utils.py:1231] [2400] img/sec/core = 164.21464704354403
+I1128 02:45:09.024162 137274321021824 utils.py:1231] [2400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.881900805300557
+I1128 02:45:09.024209 137274321021824 utils.py:1231] [2400] core_hours = 4.881900805300557
+I1128 02:45:09.024265 137274321021824 train.py:125] NOTE: Steps:2400/112603 [2.1%]
+Walltime:4h54m (0s eval)
+ETA:9d8h21m
+Total train time:9d13h14m
+I1128 02:50:20.805113 137274321021824 utils.py:1231] [2450] l2_params = 207.2472588507702
+I1128 02:50:20.805345 137274321021824 utils.py:1231] [2450] train/loss = 5.856095552444458
+I1128 02:50:20.805452 137274321021824 utils.py:1231] [2450] l2_grads = 1.5205655097961426
+I1128 02:50:20.805514 137274321021824 utils.py:1231] [2450] lr = 0.0002449
+I1128 02:50:20.805565 137274321021824 utils.py:1231] [2450] uptime = 18010.167926972004
+I1128 02:50:20.805616 137274321021824 utils.py:1231] [2450] examples_seen = 2508800.0
+I1128 02:50:20.805666 137274321021824 utils.py:1231] [2450] progress = 0.021757857250694918
+I1128 02:50:20.805713 137274321021824 utils.py:1231] [2450] epoch = 1.958214658978884
+I1128 02:50:20.805764 137274321021824 utils.py:1231] [2450] img/sec/core = 164.21748507447617
+I1128 02:50:20.805817 137274321021824 utils.py:1231] [2450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 4.968506821802222
+I1128 02:50:20.805865 137274321021824 utils.py:1231] [2450] core_hours = 4.968506821802222
+I1128 02:50:20.805934 137274321021824 train.py:125] NOTE: Steps:2450/112603 [2.2%]
+Walltime:5h0m (0s eval)
+ETA:9d7h34m
+Total train time:9d12h32m
+I1128 02:55:32.611287 137274321021824 utils.py:1231] [2500] l2_params = 207.28731570163444
+I1128 02:55:32.611526 137274321021824 utils.py:1231] [2500] train/loss = 5.933801591396332
+I1128 02:55:32.611629 137274321021824 utils.py:1231] [2500] l2_grads = 1.6542280912399292
+I1128 02:55:32.611712 137274321021824 utils.py:1231] [2500] lr = 0.0002499
+I1128 02:55:32.611786 137274321021824 utils.py:1231] [2500] uptime = 18321.974147591005
+I1128 02:55:32.611851 137274321021824 utils.py:1231] [2500] examples_seen = 2560000.0
+I1128 02:55:32.611915 137274321021824 utils.py:1231] [2500] progress = 0.022201895153770326
+I1128 02:55:32.611974 137274321021824 utils.py:1231] [2500] epoch = 1.998178223447841
+I1128 02:55:32.612028 137274321021824 utils.py:1231] [2500] img/sec/core = 164.20454953835517
+I1128 02:55:32.612088 137274321021824 utils.py:1231] [2500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.055119660863056
+I1128 02:55:32.612151 137274321021824 utils.py:1231] [2500] core_hours = 5.055119660863056
+I1128 02:55:32.612231 137274321021824 train.py:125] NOTE: Steps:2500/112603 [2.2%]
+Walltime:5h5m (0s eval)
+ETA:9d6h48m
+Total train time:9d11h52m
+I1128 02:55:32.612354 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:2500/112603 [2.2%]
+Walltime:5h5m (0s eval)
+ETA:9d6h48m
+Total train time:9d11h52m
+I1128 02:57:15.180923 137274321021824 utils.py:1231] [2500] val/acc@1 = 0.059490593112244895
+I1128 02:57:15.181278 137274321021824 utils.py:1231] [2500] val/loss = 5.516728479035047
+I1128 02:57:15.181587 137274321021824 utils.py:1231] [2500] z/secs/eval/val = 102.56066390599881
+I1128 02:57:15.181736 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 102.56066390599881
+I1128 03:02:26.686961 137274321021824 utils.py:1231] [2550] l2_params = 207.32724373311453
+I1128 03:02:26.687144 137274321021824 utils.py:1231] [2550] train/loss = 5.900575578212738
+I1128 03:02:26.687226 137274321021824 utils.py:1231] [2550] l2_grads = 1.799676537513733
+I1128 03:02:26.687279 137274321021824 utils.py:1231] [2550] lr = 0.0002549
+I1128 03:02:26.687329 137274321021824 utils.py:1231] [2550] uptime = 18736.049691382002
+I1128 03:02:26.687377 137274321021824 utils.py:1231] [2550] examples_seen = 2611200.0
+I1128 03:02:26.687421 137274321021824 utils.py:1231] [2550] progress = 0.022645933056845733
+I1128 03:02:26.687465 137274321021824 utils.py:1231] [2550] epoch = 2.0381417879167976
+I1128 03:02:26.687511 137274321021824 utils.py:1231] [2550] img/sec/core = 123.648935001684
+I1128 03:02:26.687562 137274321021824 utils.py:1231] [2550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.170140645249445
+I1128 03:02:26.687606 137274321021824 utils.py:1231] [2550] core_hours = 5.170140645249445
+I1128 03:02:26.687661 137274321021824 train.py:125] NOTE: Steps:2550/112603 [2.3%]
+Walltime:5h12m (0s eval)
+ETA:9d7h18m
+Total train time:9d12h28m
+I1128 03:07:38.479702 137274321021824 utils.py:1231] [2600] l2_params = 207.37759473091086
+I1128 03:07:38.479936 137274321021824 utils.py:1231] [2600] train/loss = 6.6590588092803955
+I1128 03:07:38.480026 137274321021824 utils.py:1231] [2600] l2_grads = 1.844239354133606
+I1128 03:07:38.480089 137274321021824 utils.py:1231] [2600] lr = 0.00025990000000000003
+I1128 03:07:38.480139 137274321021824 utils.py:1231] [2600] uptime = 19047.842501572006
+I1128 03:07:38.480189 137274321021824 utils.py:1231] [2600] examples_seen = 2662400.0
+I1128 03:07:38.480234 137274321021824 utils.py:1231] [2600] progress = 0.02308997095992114
+I1128 03:07:38.480278 137274321021824 utils.py:1231] [2600] epoch = 2.0781053523857547
+I1128 03:07:38.480324 137274321021824 utils.py:1231] [2600] img/sec/core = 164.2116120920147
+I1128 03:07:38.480377 137274321021824 utils.py:1231] [2600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.256749759191112
+I1128 03:07:38.480423 137274321021824 utils.py:1231] [2600] core_hours = 5.256749759191112
+I1128 03:07:38.480480 137274321021824 train.py:125] NOTE: Steps:2600/112603 [2.3%]
+Walltime:5h17m (0s eval)
+ETA:9d6h34m
+Total train time:9d11h50m
+I1128 03:12:50.268852 137274321021824 utils.py:1231] [2650] l2_params = 207.44467500517382
+I1128 03:12:50.269105 137274321021824 utils.py:1231] [2650] train/loss = 5.840724050998688
+I1128 03:12:50.269349 137274321021824 utils.py:1231] [2650] l2_grads = 1.358041524887085
+I1128 03:12:50.269463 137274321021824 utils.py:1231] [2650] lr = 0.00026490000000000004
+I1128 03:12:50.269536 137274321021824 utils.py:1231] [2650] uptime = 19359.631895032006
+I1128 03:12:50.269621 137274321021824 utils.py:1231] [2650] examples_seen = 2713600.0
+I1128 03:12:50.269726 137274321021824 utils.py:1231] [2650] progress = 0.023534008862996544
+I1128 03:12:50.269801 137274321021824 utils.py:1231] [2650] epoch = 2.1180689168547113
+I1128 03:12:50.269879 137274321021824 utils.py:1231] [2650] img/sec/core = 164.21341159755835
+I1128 03:12:50.269982 137274321021824 utils.py:1231] [2650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.343357924041112
+I1128 03:12:50.270068 137274321021824 utils.py:1231] [2650] core_hours = 5.343357924041112
+I1128 03:12:50.270167 137274321021824 train.py:125] NOTE: Steps:2650/112603 [2.4%]
+Walltime:5h22m (0s eval)
+ETA:9d5h52m
+Total train time:9d11h13m
+I1128 03:18:02.053124 137274321021824 utils.py:1231] [2700] l2_params = 207.4879494861996
+I1128 03:18:02.053465 137274321021824 utils.py:1231] [2700] train/loss = 5.800882399082184
+I1128 03:18:02.053662 137274321021824 utils.py:1231] [2700] l2_grads = 1.5212534666061401
+I1128 03:18:02.053756 137274321021824 utils.py:1231] [2700] lr = 0.0002699
+I1128 03:18:02.053826 137274321021824 utils.py:1231] [2700] uptime = 19671.416187132003
+I1128 03:18:02.053907 137274321021824 utils.py:1231] [2700] examples_seen = 2764800.0
+I1128 03:18:02.053966 137274321021824 utils.py:1231] [2700] progress = 0.023978046766071952
+I1128 03:18:02.054023 137274321021824 utils.py:1231] [2700] epoch = 2.158032481323668
+I1128 03:18:02.054081 137274321021824 utils.py:1231] [2700] img/sec/core = 164.21609842864956
+I1128 03:18:02.054139 137274321021824 utils.py:1231] [2700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.429964671846666
+I1128 03:18:02.054200 137274321021824 utils.py:1231] [2700] core_hours = 5.429964671846666
+I1128 03:18:02.054266 137274321021824 train.py:125] NOTE: Steps:2700/112603 [2.4%]
+Walltime:5h27m (0s eval)
+ETA:9d5h11m
+Total train time:9d10h37m
+I1128 03:23:13.659164 137274321021824 utils.py:1231] [2750] l2_params = 207.54383556189742
+I1128 03:23:13.659422 137274321021824 utils.py:1231] [2750] train/loss = 6.114029824733734
+I1128 03:23:13.659548 137274321021824 utils.py:1231] [2750] l2_grads = 1.2425663471221924
+I1128 03:23:13.659645 137274321021824 utils.py:1231] [2750] lr = 0.00027489999999999996
+I1128 03:23:13.659726 137274321021824 utils.py:1231] [2750] uptime = 19983.022087208003
+I1128 03:23:13.659786 137274321021824 utils.py:1231] [2750] examples_seen = 2816000.0
+I1128 03:23:13.659853 137274321021824 utils.py:1231] [2750] progress = 0.02442208466914736
+I1128 03:23:13.659914 137274321021824 utils.py:1231] [2750] epoch = 2.197996045792625
+I1128 03:23:13.659990 137274321021824 utils.py:1231] [2750] img/sec/core = 164.31011090455118
+I1128 03:23:13.660053 137274321021824 utils.py:1231] [2750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.516521866312222
+I1128 03:23:13.660114 137274321021824 utils.py:1231] [2750] core_hours = 5.516521866312222
+I1128 03:23:13.660196 137274321021824 train.py:125] NOTE: Steps:2750/112603 [2.4%]
+Walltime:5h33m (0s eval)
+ETA:9d4h31m
+Total train time:9d10h2m
+I1128 03:28:25.452361 137274321021824 utils.py:1231] [2800] l2_params = 207.59415409566645
+I1128 03:28:25.452571 137274321021824 utils.py:1231] [2800] train/loss = 5.844701826572418
+I1128 03:28:25.452673 137274321021824 utils.py:1231] [2800] l2_grads = 1.364916205406189
+I1128 03:28:25.452740 137274321021824 utils.py:1231] [2800] lr = 0.0002799
+I1128 03:28:25.452800 137274321021824 utils.py:1231] [2800] uptime = 20294.815162376006
+I1128 03:28:25.452857 137274321021824 utils.py:1231] [2800] examples_seen = 2867200.0
+I1128 03:28:25.452916 137274321021824 utils.py:1231] [2800] progress = 0.024866122572222767
+I1128 03:28:25.452973 137274321021824 utils.py:1231] [2800] epoch = 2.2379596102615817
+I1128 03:28:25.453026 137274321021824 utils.py:1231] [2800] img/sec/core = 164.21147253643122
+I1128 03:28:25.453085 137274321021824 utils.py:1231] [2800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.60313105385889
+I1128 03:28:25.453138 137274321021824 utils.py:1231] [2800] core_hours = 5.60313105385889
+I1128 03:28:25.453200 137274321021824 train.py:125] NOTE: Steps:2800/112603 [2.5%]
+Walltime:5h38m (0s eval)
+ETA:9d3h53m
+Total train time:9d9h29m
+I1128 03:33:37.232022 137274321021824 utils.py:1231] [2850] l2_params = 207.66386961447284
+I1128 03:33:37.232226 137274321021824 utils.py:1231] [2850] train/loss = 5.757157146930695
+I1128 03:33:37.232328 137274321021824 utils.py:1231] [2850] l2_grads = 1.2538517713546753
+I1128 03:33:37.232390 137274321021824 utils.py:1231] [2850] lr = 0.0002849
+I1128 03:33:37.232442 137274321021824 utils.py:1231] [2850] uptime = 20606.594803810003
+I1128 03:33:37.232493 137274321021824 utils.py:1231] [2850] examples_seen = 2918400.0
+I1128 03:33:37.232542 137274321021824 utils.py:1231] [2850] progress = 0.02531016047529817
+I1128 03:33:37.232592 137274321021824 utils.py:1231] [2850] epoch = 2.277923174730539
+I1128 03:33:37.232643 137274321021824 utils.py:1231] [2850] img/sec/core = 164.21854796070406
+I1128 03:33:37.232699 137274321021824 utils.py:1231] [2850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.689736509812778
+I1128 03:33:37.232750 137274321021824 utils.py:1231] [2850] core_hours = 5.689736509812778
+I1128 03:33:37.232814 137274321021824 train.py:125] NOTE: Steps:2850/112603 [2.5%]
+Walltime:5h43m (0s eval)
+ETA:9d3h15m
+Total train time:9d8h57m
+I1128 03:38:49.018586 137274321021824 utils.py:1231] [2900] l2_params = 207.7457213945987
+I1128 03:38:49.018850 137274321021824 utils.py:1231] [2900] train/loss = 6.439901292324066
+I1128 03:38:49.019000 137274321021824 utils.py:1231] [2900] l2_grads = 1.4877070188522339
+I1128 03:38:49.019083 137274321021824 utils.py:1231] [2900] lr = 0.0002899
+I1128 03:38:49.019149 137274321021824 utils.py:1231] [2900] uptime = 20918.381511299005
+I1128 03:38:49.019213 137274321021824 utils.py:1231] [2900] examples_seen = 2969600.0
+I1128 03:38:49.019262 137274321021824 utils.py:1231] [2900] progress = 0.025754198378373578
+I1128 03:38:49.019322 137274321021824 utils.py:1231] [2900] epoch = 2.3178867391994955
+I1128 03:38:49.019379 137274321021824 utils.py:1231] [2900] img/sec/core = 164.21482625844803
+I1128 03:38:49.019450 137274321021824 utils.py:1231] [2900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.776343928559723
+I1128 03:38:49.019499 137274321021824 utils.py:1231] [2900] core_hours = 5.776343928559723
+I1128 03:38:49.019559 137274321021824 train.py:125] NOTE: Steps:2900/112603 [2.6%]
+Walltime:5h48m (0s eval)
+ETA:9d2h39m
+Total train time:9d8h26m
+I1128 03:44:00.812181 137274321021824 utils.py:1231] [2950] l2_params = 207.84872412941138
+I1128 03:44:00.812455 137274321021824 utils.py:1231] [2950] train/loss = 6.545375287532806
+I1128 03:44:00.812677 137274321021824 utils.py:1231] [2950] l2_grads = 1.9666757583618164
+I1128 03:44:00.812790 137274321021824 utils.py:1231] [2950] lr = 0.0002949
+I1128 03:44:00.812868 137274321021824 utils.py:1231] [2950] uptime = 21230.175226562
+I1128 03:44:00.812963 137274321021824 utils.py:1231] [2950] examples_seen = 3020800.0
+I1128 03:44:00.813065 137274321021824 utils.py:1231] [2950] progress = 0.026198236281448985
+I1128 03:44:00.813159 137274321021824 utils.py:1231] [2950] epoch = 2.357850303668452
+I1128 03:44:00.813262 137274321021824 utils.py:1231] [2950] img/sec/core = 164.21113541949694
+I1128 03:44:00.813362 137274321021824 utils.py:1231] [2950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.862953293910555
+I1128 03:44:00.813452 137274321021824 utils.py:1231] [2950] core_hours = 5.862953293910555
+I1128 03:44:00.813551 137274321021824 train.py:125] NOTE: Steps:2950/112603 [2.6%]
+Walltime:5h53m (0s eval)
+ETA:9d2h4m
+Total train time:9d7h56m
+I1128 03:49:12.598361 137274321021824 utils.py:1231] [3000] l2_params = 207.91980171388886
+I1128 03:49:12.598568 137274321021824 utils.py:1231] [3000] train/loss = 6.103371977806091
+I1128 03:49:12.598697 137274321021824 utils.py:1231] [3000] l2_grads = 1.2588528394699097
+I1128 03:49:12.598808 137274321021824 utils.py:1231] [3000] lr = 0.0002999
+I1128 03:49:12.598889 137274321021824 utils.py:1231] [3000] uptime = 21541.961246208004
+I1128 03:49:12.598955 137274321021824 utils.py:1231] [3000] examples_seen = 3072000.0
+I1128 03:49:12.599010 137274321021824 utils.py:1231] [3000] progress = 0.02664227418452439
+I1128 03:49:12.599062 137274321021824 utils.py:1231] [3000] epoch = 2.3978138681374093
+I1128 03:49:12.599116 137274321021824 utils.py:1231] [3000] img/sec/core = 164.21518853902361
+I1128 03:49:12.599171 137274321021824 utils.py:1231] [3000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 5.9495605215900005
+I1128 03:49:12.599221 137274321021824 utils.py:1231] [3000] core_hours = 5.9495605215900005
+I1128 03:49:12.599281 137274321021824 train.py:125] NOTE: Steps:3000/112603 [2.7%]
+Walltime:5h59m (0s eval)
+ETA:9d1h30m
+Total train time:9d7h27m
+I1128 03:54:24.756111 137274321021824 utils.py:1231] [3050] l2_params = 208.0004749661547
+I1128 03:54:24.756406 137274321021824 utils.py:1231] [3050] train/loss = 6.3073049783706665
+I1128 03:54:24.756638 137274321021824 utils.py:1231] [3050] l2_grads = 1.3921799659729004
+I1128 03:54:24.756746 137274321021824 utils.py:1231] [3050] lr = 0.0003049
+I1128 03:54:24.756847 137274321021824 utils.py:1231] [3050] uptime = 21854.119199730005
+I1128 03:54:24.756943 137274321021824 utils.py:1231] [3050] examples_seen = 3123200.0
+I1128 03:54:24.757040 137274321021824 utils.py:1231] [3050] progress = 0.027086312087599797
+I1128 03:54:24.757127 137274321021824 utils.py:1231] [3050] epoch = 2.437777432606366
+I1128 03:54:24.757224 137274321021824 utils.py:1231] [3050] img/sec/core = 164.01952736530689
+I1128 03:54:24.757330 137274321021824 utils.py:1231] [3050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.036271064235001
+I1128 03:54:24.757411 137274321021824 utils.py:1231] [3050] core_hours = 6.036271064235001
+I1128 03:54:24.757500 137274321021824 train.py:125] NOTE: Steps:3050/112603 [2.7%]
+Walltime:6h4m (0s eval)
+ETA:9d0h57m
+Total train time:9d6h59m
+I1128 03:59:36.537946 137274321021824 utils.py:1231] [3100] l2_params = 208.08434747192854
+I1128 03:59:36.538180 137274321021824 utils.py:1231] [3100] train/loss = 6.424351811408997
+I1128 03:59:36.538283 137274321021824 utils.py:1231] [3100] l2_grads = 1.309248685836792
+I1128 03:59:36.538373 137274321021824 utils.py:1231] [3100] lr = 0.0003099
+I1128 03:59:36.538439 137274321021824 utils.py:1231] [3100] uptime = 22165.900797647002
+I1128 03:59:36.538497 137274321021824 utils.py:1231] [3100] examples_seen = 3174400.0
+I1128 03:59:36.538552 137274321021824 utils.py:1231] [3100] progress = 0.027530349990675204
+I1128 03:59:36.538606 137274321021824 utils.py:1231] [3100] epoch = 2.4777409970753226
+I1128 03:59:36.538661 137274321021824 utils.py:1231] [3100] img/sec/core = 164.2175174611508
+I1128 03:59:36.538721 137274321021824 utils.py:1231] [3100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.1228770636563885
+I1128 03:59:36.538775 137274321021824 utils.py:1231] [3100] core_hours = 6.1228770636563885
+I1128 03:59:36.538837 137274321021824 train.py:125] NOTE: Steps:3100/112603 [2.8%]
+Walltime:6h9m (0s eval)
+ETA:9d0h25m
+Total train time:9d6h32m
+I1128 04:04:48.326585 137274321021824 utils.py:1231] [3150] l2_params = 208.1789630801884
+I1128 04:04:48.326797 137274321021824 utils.py:1231] [3150] train/loss = 6.506803572177887
+I1128 04:04:48.326907 137274321021824 utils.py:1231] [3150] l2_grads = 1.565660834312439
+I1128 04:04:48.326980 137274321021824 utils.py:1231] [3150] lr = 0.0003149
+I1128 04:04:48.327030 137274321021824 utils.py:1231] [3150] uptime = 22477.689392613007
+I1128 04:04:48.327081 137274321021824 utils.py:1231] [3150] examples_seen = 3225600.0
+I1128 04:04:48.327129 137274321021824 utils.py:1231] [3150] progress = 0.02797438789375061
+I1128 04:04:48.327180 137274321021824 utils.py:1231] [3150] epoch = 2.5177045615442797
+I1128 04:04:48.327236 137274321021824 utils.py:1231] [3150] img/sec/core = 164.2138321498978
+I1128 04:04:48.327303 137274321021824 utils.py:1231] [3150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.209485006702502
+I1128 04:04:48.327351 137274321021824 utils.py:1231] [3150] core_hours = 6.209485006702502
+I1128 04:04:48.327434 137274321021824 train.py:125] NOTE: Steps:3150/112603 [2.8%]
+Walltime:6h14m (0s eval)
+ETA:8d23h53m
+Total train time:9d6h6m
+I1128 04:10:00.109024 137274321021824 utils.py:1231] [3200] l2_params = 208.24913056932874
+I1128 04:10:00.109259 137274321021824 utils.py:1231] [3200] train/loss = 5.886916220188141
+I1128 04:10:00.109360 137274321021824 utils.py:1231] [3200] l2_grads = 1.609678030014038
+I1128 04:10:00.109421 137274321021824 utils.py:1231] [3200] lr = 0.0003199
+I1128 04:10:00.109472 137274321021824 utils.py:1231] [3200] uptime = 22789.471833664
+I1128 04:10:00.109522 137274321021824 utils.py:1231] [3200] examples_seen = 3276800.0
+I1128 04:10:00.109570 137274321021824 utils.py:1231] [3200] progress = 0.028418425796826016
+I1128 04:10:00.109615 137274321021824 utils.py:1231] [3200] epoch = 2.5576681260132363
+I1128 04:10:00.109663 137274321021824 utils.py:1231] [3200] img/sec/core = 164.2170733778625
+I1128 04:10:00.109716 137274321021824 utils.py:1231] [3200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.296091240327778
+I1128 04:10:00.109763 137274321021824 utils.py:1231] [3200] core_hours = 6.296091240327778
+I1128 04:10:00.109820 137274321021824 train.py:125] NOTE: Steps:3200/112603 [2.8%]
+Walltime:6h19m (0s eval)
+ETA:8d23h23m
+Total train time:9d5h41m
+I1128 04:15:11.885889 137274321021824 utils.py:1231] [3250] l2_params = 208.342183278545
+I1128 04:15:11.886092 137274321021824 utils.py:1231] [3250] train/loss = 5.683818876743317
+I1128 04:15:11.886193 137274321021824 utils.py:1231] [3250] l2_grads = 1.3292096853256226
+I1128 04:15:11.886260 137274321021824 utils.py:1231] [3250] lr = 0.00032490000000000004
+I1128 04:15:11.886319 137274321021824 utils.py:1231] [3250] uptime = 23101.248679596007
+I1128 04:15:11.886392 137274321021824 utils.py:1231] [3250] examples_seen = 3328000.0
+I1128 04:15:11.886446 137274321021824 utils.py:1231] [3250] progress = 0.028862463699901423
+I1128 04:15:11.886516 137274321021824 utils.py:1231] [3250] epoch = 2.597631690482193
+I1128 04:15:11.886575 137274321021824 utils.py:1231] [3250] img/sec/core = 164.2200204025601
+I1128 04:15:11.886635 137274321021824 utils.py:1231] [3250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.382695919753334
+I1128 04:15:11.886685 137274321021824 utils.py:1231] [3250] core_hours = 6.382695919753334
+I1128 04:15:11.886747 137274321021824 train.py:125] NOTE: Steps:3250/112603 [2.9%]
+Walltime:6h25m (0s eval)
+ETA:8d22h53m
+Total train time:9d5h16m
+I1128 04:20:23.655164 137274321021824 utils.py:1231] [3300] l2_params = 208.4695742974759
+I1128 04:20:23.655439 137274321021824 utils.py:1231] [3300] train/loss = 5.7239843010902405
+I1128 04:20:23.655554 137274321021824 utils.py:1231] [3300] l2_grads = 1.4334020614624023
+I1128 04:20:23.655631 137274321021824 utils.py:1231] [3300] lr = 0.00032990000000000005
+I1128 04:20:23.655699 137274321021824 utils.py:1231] [3300] uptime = 23413.018060487004
+I1128 04:20:23.655778 137274321021824 utils.py:1231] [3300] examples_seen = 3379200.0
+I1128 04:20:23.655834 137274321021824 utils.py:1231] [3300] progress = 0.02930650160297683
+I1128 04:20:23.655897 137274321021824 utils.py:1231] [3300] epoch = 2.63759525495115
+I1128 04:20:23.655954 137274321021824 utils.py:1231] [3300] img/sec/core = 164.22395250513995
+I1128 04:20:23.656014 137274321021824 utils.py:1231] [3300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.469298525556389
+I1128 04:20:23.656070 137274321021824 utils.py:1231] [3300] core_hours = 6.469298525556389
+I1128 04:20:23.656133 137274321021824 train.py:125] NOTE: Steps:3300/112603 [2.9%]
+Walltime:6h30m (0s eval)
+ETA:8d22h24m
+Total train time:9d4h52m
+I1128 04:25:35.428113 137274321021824 utils.py:1231] [3350] l2_params = 208.63425709389668
+I1128 04:25:35.428388 137274321021824 utils.py:1231] [3350] train/loss = 5.682764530181885
+I1128 04:25:35.428590 137274321021824 utils.py:1231] [3350] l2_grads = 1.9380338191986084
+I1128 04:25:35.428668 137274321021824 utils.py:1231] [3350] lr = 0.0003349
+I1128 04:25:35.428731 137274321021824 utils.py:1231] [3350] uptime = 23724.791090983003
+I1128 04:25:35.428794 137274321021824 utils.py:1231] [3350] examples_seen = 3430400.0
+I1128 04:25:35.428857 137274321021824 utils.py:1231] [3350] progress = 0.029750539506052238
+I1128 04:25:35.428923 137274321021824 utils.py:1231] [3350] epoch = 2.6775588194201068
+I1128 04:25:35.428985 137274321021824 utils.py:1231] [3350] img/sec/core = 164.22203010486857
+I1128 04:25:35.429050 137274321021824 utils.py:1231] [3350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.555902145138611
+I1128 04:25:35.429106 137274321021824 utils.py:1231] [3350] core_hours = 6.555902145138611
+I1128 04:25:35.429176 137274321021824 train.py:125] NOTE: Steps:3350/112603 [3.0%]
+Walltime:6h35m (0s eval)
+ETA:8d21h56m
+Total train time:9d4h29m
+I1128 04:30:47.203135 137274321021824 utils.py:1231] [3400] l2_params = 208.77220791433678
+I1128 04:30:47.203333 137274321021824 utils.py:1231] [3400] train/loss = 5.606162071228027
+I1128 04:30:47.203424 137274321021824 utils.py:1231] [3400] l2_grads = 1.3051503896713257
+I1128 04:30:47.203484 137274321021824 utils.py:1231] [3400] lr = 0.00033989999999999997
+I1128 04:30:47.203538 137274321021824 utils.py:1231] [3400] uptime = 24036.565900237
+I1128 04:30:47.203590 137274321021824 utils.py:1231] [3400] examples_seen = 3481600.0
+I1128 04:30:47.203640 137274321021824 utils.py:1231] [3400] progress = 0.03019457740912764
+I1128 04:30:47.203688 137274321021824 utils.py:1231] [3400] epoch = 2.717522383889064
+I1128 04:30:47.203739 137274321021824 utils.py:1231] [3400] img/sec/core = 164.22109317461906
+I1128 04:30:47.203796 137274321021824 utils.py:1231] [3400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.642506258820277
+I1128 04:30:47.203845 137274321021824 utils.py:1231] [3400] core_hours = 6.642506258820277
+I1128 04:30:47.203910 137274321021824 train.py:125] NOTE: Steps:3400/112603 [3.0%]
+Walltime:6h40m (0s eval)
+ETA:8d21h28m
+Total train time:9d4h7m
+I1128 04:35:58.996622 137274321021824 utils.py:1231] [3450] l2_params = 208.88224626119202
+I1128 04:35:58.996980 137274321021824 utils.py:1231] [3450] train/loss = 5.688660025596619
+I1128 04:35:58.997174 137274321021824 utils.py:1231] [3450] l2_grads = 1.5036791563034058
+I1128 04:35:58.997249 137274321021824 utils.py:1231] [3450] lr = 0.0003449
+I1128 04:35:58.997306 137274321021824 utils.py:1231] [3450] uptime = 24348.359668689
+I1128 04:35:58.997354 137274321021824 utils.py:1231] [3450] examples_seen = 3532800.0
+I1128 04:35:58.997401 137274321021824 utils.py:1231] [3450] progress = 0.03063861531220305
+I1128 04:35:58.997446 137274321021824 utils.py:1231] [3450] epoch = 2.7574859483580205
+I1128 04:35:58.997492 137274321021824 utils.py:1231] [3450] img/sec/core = 164.2111074066642
+I1128 04:35:58.997545 137274321021824 utils.py:1231] [3450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.729115638945832
+I1128 04:35:58.997590 137274321021824 utils.py:1231] [3450] core_hours = 6.729115638945832
+I1128 04:35:58.997647 137274321021824 train.py:125] NOTE: Steps:3450/112603 [3.1%]
+Walltime:6h45m (0s eval)
+ETA:8d21h1m
+Total train time:9d3h45m
+I1128 04:41:10.778661 137274321021824 utils.py:1231] [3500] l2_params = 209.0262061973573
+I1128 04:41:10.778941 137274321021824 utils.py:1231] [3500] train/loss = 5.645868599414825
+I1128 04:41:10.779069 137274321021824 utils.py:1231] [3500] l2_grads = 1.4707714319229126
+I1128 04:41:10.779138 137274321021824 utils.py:1231] [3500] lr = 0.0003499
+I1128 04:41:10.779203 137274321021824 utils.py:1231] [3500] uptime = 24660.141565278005
+I1128 04:41:10.779259 137274321021824 utils.py:1231] [3500] examples_seen = 3584000.0
+I1128 04:41:10.779308 137274321021824 utils.py:1231] [3500] progress = 0.031082653215278457
+I1128 04:41:10.779360 137274321021824 utils.py:1231] [3500] epoch = 2.797449512826977
+I1128 04:41:10.779409 137274321021824 utils.py:1231] [3500] img/sec/core = 164.21736014869626
+I1128 04:41:10.779465 137274321021824 utils.py:1231] [3500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.815721721331667
+I1128 04:41:10.779524 137274321021824 utils.py:1231] [3500] core_hours = 6.815721721331667
+I1128 04:41:10.779582 137274321021824 train.py:125] NOTE: Steps:3500/112603 [3.1%]
+Walltime:6h51m (0s eval)
+ETA:8d20h34m
+Total train time:9d3h24m
+I1128 04:46:22.503247 137274321021824 utils.py:1231] [3550] l2_params = 209.18645885250066
+I1128 04:46:22.503436 137274321021824 utils.py:1231] [3550] train/loss = 5.590174376964569
+I1128 04:46:22.503529 137274321021824 utils.py:1231] [3550] l2_grads = 1.4766831398010254
+I1128 04:46:22.503592 137274321021824 utils.py:1231] [3550] lr = 0.0003549
+I1128 04:46:22.503654 137274321021824 utils.py:1231] [3550] uptime = 24971.866015950007
+I1128 04:46:22.503708 137274321021824 utils.py:1231] [3550] examples_seen = 3635200.0
+I1128 04:46:22.503763 137274321021824 utils.py:1231] [3550] progress = 0.031526691118353864
+I1128 04:46:22.503814 137274321021824 utils.py:1231] [3550] epoch = 2.8374130772959343
+I1128 04:46:22.503865 137274321021824 utils.py:1231] [3550] img/sec/core = 164.24762282722872
+I1128 04:46:22.503927 137274321021824 utils.py:1231] [3550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.902311846518335
+I1128 04:46:22.503978 137274321021824 utils.py:1231] [3550] core_hours = 6.902311846518335
+I1128 04:46:22.504033 137274321021824 train.py:125] NOTE: Steps:3550/112603 [3.2%]
+Walltime:6h56m (0s eval)
+ETA:8d20h9m
+Total train time:9d3h3m
+I1128 04:51:34.291948 137274321021824 utils.py:1231] [3600] l2_params = 209.33164468791426
+I1128 04:51:34.292239 137274321021824 utils.py:1231] [3600] train/loss = 5.540681302547455
+I1128 04:51:34.292404 137274321021824 utils.py:1231] [3600] l2_grads = 1.4787309169769287
+I1128 04:51:34.292475 137274321021824 utils.py:1231] [3600] lr = 0.0003599
+I1128 04:51:34.292538 137274321021824 utils.py:1231] [3600] uptime = 25283.654899235
+I1128 04:51:34.292597 137274321021824 utils.py:1231] [3600] examples_seen = 3686400.0
+I1128 04:51:34.292658 137274321021824 utils.py:1231] [3600] progress = 0.03197072902142927
+I1128 04:51:34.292716 137274321021824 utils.py:1231] [3600] epoch = 2.877376641764891
+I1128 04:51:34.292772 137274321021824 utils.py:1231] [3600] img/sec/core = 164.213680297255
+I1128 04:51:34.292833 137274321021824 utils.py:1231] [3600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 6.9889198696530554
+I1128 04:51:34.292899 137274321021824 utils.py:1231] [3600] core_hours = 6.9889198696530554
+I1128 04:51:34.292992 137274321021824 train.py:125] NOTE: Steps:3600/112603 [3.2%]
+Walltime:7h1m (0s eval)
+ETA:8d19h43m
+Total train time:9d2h43m
+I1128 04:56:46.069494 137274321021824 utils.py:1231] [3650] l2_params = 209.46431624810504
+I1128 04:56:46.069699 137274321021824 utils.py:1231] [3650] train/loss = 6.467800498008728
+I1128 04:56:46.069801 137274321021824 utils.py:1231] [3650] l2_grads = 1.268568754196167
+I1128 04:56:46.069900 137274321021824 utils.py:1231] [3650] lr = 0.00036490000000000003
+I1128 04:56:46.069979 137274321021824 utils.py:1231] [3650] uptime = 25595.432337651
+I1128 04:56:46.070056 137274321021824 utils.py:1231] [3650] examples_seen = 3737600.0
+I1128 04:56:46.070132 137274321021824 utils.py:1231] [3650] progress = 0.03241476692450468
+I1128 04:56:46.070202 137274321021824 utils.py:1231] [3650] epoch = 2.9173402062338476
+I1128 04:56:46.070277 137274321021824 utils.py:1231] [3650] img/sec/core = 164.2197083282356
+I1128 04:56:46.070361 137274321021824 utils.py:1231] [3650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.075524713657499
+I1128 04:56:46.070436 137274321021824 utils.py:1231] [3650] core_hours = 7.075524713657499
+I1128 04:56:46.070528 137274321021824 train.py:125] NOTE: Steps:3650/112603 [3.2%]
+Walltime:7h6m (0s eval)
+ETA:8d19h19m
+Total train time:9d2h24m
+I1128 05:01:57.863904 137274321021824 utils.py:1231] [3700] l2_params = 209.64027833183258
+I1128 05:01:57.864137 137274321021824 utils.py:1231] [3700] train/loss = 6.042769074440002
+I1128 05:01:57.864255 137274321021824 utils.py:1231] [3700] l2_grads = 1.6913701295852661
+I1128 05:01:57.864326 137274321021824 utils.py:1231] [3700] lr = 0.0003699
+I1128 05:01:57.864434 137274321021824 utils.py:1231] [3700] uptime = 25907.226796104005
+I1128 05:01:57.864488 137274321021824 utils.py:1231] [3700] examples_seen = 3788800.0
+I1128 05:01:57.864536 137274321021824 utils.py:1231] [3700] progress = 0.03285880482758008
+I1128 05:01:57.864583 137274321021824 utils.py:1231] [3700] epoch = 2.9573037707028047
+I1128 05:01:57.864632 137274321021824 utils.py:1231] [3700] img/sec/core = 164.2107440075529
+I1128 05:01:57.864686 137274321021824 utils.py:1231] [3700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.1621342854500005
+I1128 05:01:57.864734 137274321021824 utils.py:1231] [3700] core_hours = 7.1621342854500005
+I1128 05:01:57.864792 137274321021824 train.py:125] NOTE: Steps:3700/112603 [3.3%]
+Walltime:7h11m (0s eval)
+ETA:8d18h55m
+Total train time:9d2h5m
+I1128 05:07:09.643614 137274321021824 utils.py:1231] [3750] l2_params = 209.82515497630604
+I1128 05:07:09.643840 137274321021824 utils.py:1231] [3750] train/loss = 5.4262049198150635
+I1128 05:07:09.643949 137274321021824 utils.py:1231] [3750] l2_grads = 1.4200491905212402
+I1128 05:07:09.644011 137274321021824 utils.py:1231] [3750] lr = 0.0003749
+I1128 05:07:09.644065 137274321021824 utils.py:1231] [3750] uptime = 26219.006426479005
+I1128 05:07:09.644116 137274321021824 utils.py:1231] [3750] examples_seen = 3840000.0
+I1128 05:07:09.644166 137274321021824 utils.py:1231] [3750] progress = 0.03330284273065549
+I1128 05:07:09.644214 137274321021824 utils.py:1231] [3750] epoch = 2.9972673351717614
+I1128 05:07:09.644265 137274321021824 utils.py:1231] [3750] img/sec/core = 164.21855378562745
+I1128 05:07:09.644329 137274321021824 utils.py:1231] [3750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.248739738331945
+I1128 05:07:09.644378 137274321021824 utils.py:1231] [3750] core_hours = 7.248739738331945
+I1128 05:07:09.644436 137274321021824 train.py:125] NOTE: Steps:3750/112603 [3.3%]
+Walltime:7h16m (0s eval)
+ETA:8d18h31m
+Total train time:9d1h46m
+I1128 05:12:21.405940 137274321021824 utils.py:1231] [3800] l2_params = 210.00426630323543
+I1128 05:12:21.406181 137274321021824 utils.py:1231] [3800] train/loss = 5.676540195941925
+I1128 05:12:21.406282 137274321021824 utils.py:1231] [3800] l2_grads = 1.187603235244751
+I1128 05:12:21.406361 137274321021824 utils.py:1231] [3800] lr = 0.0003799
+I1128 05:12:21.406441 137274321021824 utils.py:1231] [3800] uptime = 26530.768798728
+I1128 05:12:21.406499 137274321021824 utils.py:1231] [3800] examples_seen = 3891200.0
+I1128 05:12:21.406549 137274321021824 utils.py:1231] [3800] progress = 0.033746880633730894
+I1128 05:12:21.406599 137274321021824 utils.py:1231] [3800] epoch = 3.037230899640718
+I1128 05:12:21.406649 137274321021824 utils.py:1231] [3800] img/sec/core = 164.22764437752
+I1128 05:12:21.406704 137274321021824 utils.py:1231] [3800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.3353403972899995
+I1128 05:12:21.406755 137274321021824 utils.py:1231] [3800] core_hours = 7.3353403972899995
+I1128 05:12:21.406816 137274321021824 train.py:125] NOTE: Steps:3800/112603 [3.4%]
+Walltime:7h22m (0s eval)
+ETA:8d18h8m
+Total train time:9d1h28m
+I1128 05:17:33.172911 137274321021824 utils.py:1231] [3850] l2_params = 210.20630938101363
+I1128 05:17:33.173156 137274321021824 utils.py:1231] [3850] train/loss = 6.417073309421539
+I1128 05:17:33.173297 137274321021824 utils.py:1231] [3850] l2_grads = 1.320041537284851
+I1128 05:17:33.173394 137274321021824 utils.py:1231] [3850] lr = 0.00038490000000000003
+I1128 05:17:33.173473 137274321021824 utils.py:1231] [3850] uptime = 26842.535834168004
+I1128 05:17:33.173538 137274321021824 utils.py:1231] [3850] examples_seen = 3942400.0
+I1128 05:17:33.173609 137274321021824 utils.py:1231] [3850] progress = 0.0341909185368063
+I1128 05:17:33.173674 137274321021824 utils.py:1231] [3850] epoch = 3.077194464109675
+I1128 05:17:33.173741 137274321021824 utils.py:1231] [3850] img/sec/core = 164.2251879764657
+I1128 05:17:33.173815 137274321021824 utils.py:1231] [3850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.421942351578889
+I1128 05:17:33.173879 137274321021824 utils.py:1231] [3850] core_hours = 7.421942351578889
+I1128 05:17:33.173966 137274321021824 train.py:125] NOTE: Steps:3850/112603 [3.4%]
+Walltime:7h27m (0s eval)
+ETA:8d17h45m
+Total train time:9d1h11m
+I1128 05:22:44.956176 137274321021824 utils.py:1231] [3900] l2_params = 210.4250006079954
+I1128 05:22:44.956472 137274321021824 utils.py:1231] [3900] train/loss = 5.516971111297607
+I1128 05:22:44.956652 137274321021824 utils.py:1231] [3900] l2_grads = 1.4729803800582886
+I1128 05:22:44.956750 137274321021824 utils.py:1231] [3900] lr = 0.00038990000000000004
+I1128 05:22:44.956825 137274321021824 utils.py:1231] [3900] uptime = 27154.319182623003
+I1128 05:22:44.956940 137274321021824 utils.py:1231] [3900] examples_seen = 3993600.0
+I1128 05:22:44.957073 137274321021824 utils.py:1231] [3900] progress = 0.03463495643988171
+I1128 05:22:44.957189 137274321021824 utils.py:1231] [3900] epoch = 3.117158028578632
+I1128 05:22:44.957303 137274321021824 utils.py:1231] [3900] img/sec/core = 164.21659544589141
+I1128 05:22:44.957415 137274321021824 utils.py:1231] [3900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.508548837260833
+I1128 05:22:44.957490 137274321021824 utils.py:1231] [3900] core_hours = 7.508548837260833
+I1128 05:22:44.957572 137274321021824 train.py:125] NOTE: Steps:3900/112603 [3.5%]
+Walltime:7h32m (0s eval)
+ETA:8d17h23m
+Total train time:9d0h54m
+I1128 05:27:56.735066 137274321021824 utils.py:1231] [3950] l2_params = 210.60457530162375
+I1128 05:27:56.735345 137274321021824 utils.py:1231] [3950] train/loss = 6.029466390609741
+I1128 05:27:56.735541 137274321021824 utils.py:1231] [3950] l2_grads = 1.5912528038024902
+I1128 05:27:56.735626 137274321021824 utils.py:1231] [3950] lr = 0.0003949
+I1128 05:27:56.735707 137274321021824 utils.py:1231] [3950] uptime = 27466.098067807005
+I1128 05:27:56.735780 137274321021824 utils.py:1231] [3950] examples_seen = 4044800.0
+I1128 05:27:56.735842 137274321021824 utils.py:1231] [3950] progress = 0.035078994342957116
+I1128 05:27:56.735938 137274321021824 utils.py:1231] [3950] epoch = 3.1571215930475884
+I1128 05:27:56.735995 137274321021824 utils.py:1231] [3950] img/sec/core = 164.2189462887568
+I1128 05:27:56.736056 137274321021824 utils.py:1231] [3950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.595154083145278
+I1128 05:27:56.736112 137274321021824 utils.py:1231] [3950] core_hours = 7.595154083145278
+I1128 05:27:56.736175 137274321021824 train.py:125] NOTE: Steps:3950/112603 [3.5%]
+Walltime:7h37m (0s eval)
+ETA:8d17h1m
+Total train time:9d0h37m
+I1128 05:33:08.518368 137274321021824 utils.py:1231] [4000] l2_params = 210.77089856567474
+I1128 05:33:08.518596 137274321021824 utils.py:1231] [4000] train/loss = 5.467583775520325
+I1128 05:33:08.518715 137274321021824 utils.py:1231] [4000] l2_grads = 1.511406421661377
+I1128 05:33:08.518792 137274321021824 utils.py:1231] [4000] lr = 0.00039989999999999996
+I1128 05:33:08.518852 137274321021824 utils.py:1231] [4000] uptime = 27777.881213992005
+I1128 05:33:08.518926 137274321021824 utils.py:1231] [4000] examples_seen = 4096000.0
+I1128 05:33:08.518977 137274321021824 utils.py:1231] [4000] progress = 0.035523032246032524
+I1128 05:33:08.519025 137274321021824 utils.py:1231] [4000] epoch = 3.1970851575165455
+I1128 05:33:08.519076 137274321021824 utils.py:1231] [4000] img/sec/core = 164.21670198176747
+I1128 05:33:08.519131 137274321021824 utils.py:1231] [4000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.6817605126411115
+I1128 05:33:08.519180 137274321021824 utils.py:1231] [4000] core_hours = 7.6817605126411115
+I1128 05:33:08.519240 137274321021824 train.py:125] NOTE: Steps:4000/112603 [3.6%]
+Walltime:7h42m (0s eval)
+ETA:8d16h40m
+Total train time:9d0h21m
+I1128 05:38:20.648277 137274321021824 utils.py:1231] [4050] l2_params = 210.95750023246538
+I1128 05:38:20.648508 137274321021824 utils.py:1231] [4050] train/loss = 5.454846203327179
+I1128 05:38:20.648597 137274321021824 utils.py:1231] [4050] l2_grads = 1.3712505102157593
+I1128 05:38:20.648655 137274321021824 utils.py:1231] [4050] lr = 0.0004049
+I1128 05:38:20.648706 137274321021824 utils.py:1231] [4050] uptime = 28090.011068391002
+I1128 05:38:20.648760 137274321021824 utils.py:1231] [4050] examples_seen = 4147200.0
+I1128 05:38:20.648808 137274321021824 utils.py:1231] [4050] progress = 0.03596707014910793
+I1128 05:38:20.648854 137274321021824 utils.py:1231] [4050] epoch = 3.237048721985502
+I1128 05:38:20.648907 137274321021824 utils.py:1231] [4050] img/sec/core = 164.03429303033175
+I1128 05:38:20.648967 137274321021824 utils.py:1231] [4050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.768463249974166
+I1128 05:38:20.649028 137274321021824 utils.py:1231] [4050] core_hours = 7.768463249974166
+I1128 05:38:20.649085 137274321021824 train.py:125] NOTE: Steps:4050/112603 [3.6%]
+Walltime:7h48m (0s eval)
+ETA:8d16h19m
+Total train time:9d0h5m
+I1128 05:43:32.426218 137274321021824 utils.py:1231] [4100] l2_params = 211.1569469734689
+I1128 05:43:32.426429 137274321021824 utils.py:1231] [4100] train/loss = 5.988946557044983
+I1128 05:43:32.426544 137274321021824 utils.py:1231] [4100] l2_grads = 1.2481768131256104
+I1128 05:43:32.426620 137274321021824 utils.py:1231] [4100] lr = 0.0004099
+I1128 05:43:32.426685 137274321021824 utils.py:1231] [4100] uptime = 28401.789045689
+I1128 05:43:32.426749 137274321021824 utils.py:1231] [4100] examples_seen = 4198400.0
+I1128 05:43:32.426812 137274321021824 utils.py:1231] [4100] progress = 0.03641110805218333
+I1128 05:43:32.426873 137274321021824 utils.py:1231] [4100] epoch = 3.2770122864544593
+I1128 05:43:32.426946 137274321021824 utils.py:1231] [4100] img/sec/core = 164.2194244882881
+I1128 05:43:32.427012 137274321021824 utils.py:1231] [4100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.855068243668056
+I1128 05:43:32.427071 137274321021824 utils.py:1231] [4100] core_hours = 7.855068243668056
+I1128 05:43:32.427144 137274321021824 train.py:125] NOTE: Steps:4100/112603 [3.6%]
+Walltime:7h53m (0s eval)
+ETA:8d15h58m
+Total train time:8d23h50m
+I1128 05:48:43.130007 137274321021824 utils.py:1231] [4150] l2_params = 211.36497975809135
+I1128 05:48:43.130218 137274321021824 utils.py:1231] [4150] train/loss = 6.119887053966522
+I1128 05:48:43.130316 137274321021824 utils.py:1231] [4150] l2_grads = 1.6912795305252075
+I1128 05:48:43.130381 137274321021824 utils.py:1231] [4150] lr = 0.0004149
+I1128 05:48:43.130455 137274321021824 utils.py:1231] [4150] uptime = 28712.492816571
+I1128 05:48:43.130516 137274321021824 utils.py:1231] [4150] examples_seen = 4249600.0
+I1128 05:48:43.130569 137274321021824 utils.py:1231] [4150] progress = 0.03685514595525874
+I1128 05:48:43.130654 137274321021824 utils.py:1231] [4150] epoch = 3.316975850923416
+I1128 05:48:43.130743 137274321021824 utils.py:1231] [4150] img/sec/core = 164.7871857321135
+I1128 05:48:43.130836 137274321021824 utils.py:1231] [4150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 7.941374846690833
+I1128 05:48:43.130924 137274321021824 utils.py:1231] [4150] core_hours = 7.941374846690833
+I1128 05:48:43.131018 137274321021824 train.py:125] NOTE: Steps:4150/112603 [3.7%]
+Walltime:7h58m (0s eval)
+ETA:8d15h38m
+Total train time:8d23h34m
+I1128 05:53:54.788642 137274321021824 utils.py:1231] [4200] l2_params = 211.55783165566166
+I1128 05:53:54.788886 137274321021824 utils.py:1231] [4200] train/loss = 6.071891725063324
+I1128 05:53:54.789053 137274321021824 utils.py:1231] [4200] l2_grads = 1.3146586418151855
+I1128 05:53:54.789142 137274321021824 utils.py:1231] [4200] lr = 0.0004199
+I1128 05:53:54.789221 137274321021824 utils.py:1231] [4200] uptime = 29024.151580767997
+I1128 05:53:54.789288 137274321021824 utils.py:1231] [4200] examples_seen = 4300800.0
+I1128 05:53:54.789357 137274321021824 utils.py:1231] [4200] progress = 0.037299183858334146
+I1128 05:53:54.789425 137274321021824 utils.py:1231] [4200] epoch = 3.3569394153923726
+I1128 05:53:54.789487 137274321021824 utils.py:1231] [4200] img/sec/core = 164.2822403275558
+I1128 05:53:54.789546 137274321021824 utils.py:1231] [4200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.027946725634443
+I1128 05:53:54.789595 137274321021824 utils.py:1231] [4200] core_hours = 8.027946725634443
+I1128 05:53:54.789658 137274321021824 train.py:125] NOTE: Steps:4200/112603 [3.7%]
+Walltime:8h3m (0s eval)
+ETA:8d15h18m
+Total train time:8d23h20m
+I1128 05:59:06.567332 137274321021824 utils.py:1231] [4250] l2_params = 211.80787507510075
+I1128 05:59:06.567536 137274321021824 utils.py:1231] [4250] train/loss = 5.340620994567871
+I1128 05:59:06.567648 137274321021824 utils.py:1231] [4250] l2_grads = 1.6270452737808228
+I1128 05:59:06.567714 137274321021824 utils.py:1231] [4250] lr = 0.00042490000000000003
+I1128 05:59:06.567772 137274321021824 utils.py:1231] [4250] uptime = 29335.930133659
+I1128 05:59:06.567833 137274321021824 utils.py:1231] [4250] examples_seen = 4352000.0
+I1128 05:59:06.567923 137274321021824 utils.py:1231] [4250] progress = 0.037743221761409554
+I1128 05:59:06.567993 137274321021824 utils.py:1231] [4250] epoch = 3.3969029798613297
+I1128 05:59:06.568053 137274321021824 utils.py:1231] [4250] img/sec/core = 164.21912131300274
+I1128 05:59:06.568118 137274321021824 utils.py:1231] [4250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.114551879215277
+I1128 05:59:06.568176 137274321021824 utils.py:1231] [4250] core_hours = 8.114551879215277
+I1128 05:59:06.568241 137274321021824 train.py:125] NOTE: Steps:4250/112603 [3.8%]
+Walltime:8h8m (0s eval)
+ETA:8d14h58m
+Total train time:8d23h5m
+I1128 06:04:18.351568 137274321021824 utils.py:1231] [4300] l2_params = 212.057296686654
+I1128 06:04:18.351763 137274321021824 utils.py:1231] [4300] train/loss = 5.378673493862152
+I1128 06:04:18.351855 137274321021824 utils.py:1231] [4300] l2_grads = 1.3687108755111694
+I1128 06:04:18.351923 137274321021824 utils.py:1231] [4300] lr = 0.0004299
+I1128 06:04:18.351974 137274321021824 utils.py:1231] [4300] uptime = 29647.714336321005
+I1128 06:04:18.352026 137274321021824 utils.py:1231] [4300] examples_seen = 4403200.0
+I1128 06:04:18.352073 137274321021824 utils.py:1231] [4300] progress = 0.03818725966448496
+I1128 06:04:18.352120 137274321021824 utils.py:1231] [4300] epoch = 3.4368665443302864
+I1128 06:04:18.352173 137274321021824 utils.py:1231] [4300] img/sec/core = 164.21614553545595
+I1128 06:04:18.352227 137274321021824 utils.py:1231] [4300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.201158602176944
+I1128 06:04:18.352274 137274321021824 utils.py:1231] [4300] core_hours = 8.201158602176944
+I1128 06:04:18.352332 137274321021824 train.py:125] NOTE: Steps:4300/112603 [3.8%]
+Walltime:8h14m (0s eval)
+ETA:8d14h39m
+Total train time:8d22h51m
+I1128 06:09:30.138820 137274321021824 utils.py:1231] [4350] l2_params = 212.27492568590813
+I1128 06:09:30.139110 137274321021824 utils.py:1231] [4350] train/loss = 6.340286433696747
+I1128 06:09:30.139321 137274321021824 utils.py:1231] [4350] l2_grads = 1.039141058921814
+I1128 06:09:30.139429 137274321021824 utils.py:1231] [4350] lr = 0.0004349
+I1128 06:09:30.139481 137274321021824 utils.py:1231] [4350] uptime = 29959.501843332997
+I1128 06:09:30.139541 137274321021824 utils.py:1231] [4350] examples_seen = 4454400.0
+I1128 06:09:30.139589 137274321021824 utils.py:1231] [4350] progress = 0.03863129756756037
+I1128 06:09:30.139651 137274321021824 utils.py:1231] [4350] epoch = 3.476830108799243
+I1128 06:09:30.139710 137274321021824 utils.py:1231] [4350] img/sec/core = 164.21440515905851
+I1128 06:09:30.139777 137274321021824 utils.py:1231] [4350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.28776624301361
+I1128 06:09:30.139826 137274321021824 utils.py:1231] [4350] core_hours = 8.28776624301361
+I1128 06:09:30.139901 137274321021824 train.py:125] NOTE: Steps:4350/112603 [3.9%]
+Walltime:8h19m (0s eval)
+ETA:8d14h20m
+Total train time:8d22h38m
+I1128 06:14:41.921361 137274321021824 utils.py:1231] [4400] l2_params = 212.50805946898384
+I1128 06:14:41.921604 137274321021824 utils.py:1231] [4400] train/loss = 5.275092720985413
+I1128 06:14:41.921705 137274321021824 utils.py:1231] [4400] l2_grads = 1.392372965812683
+I1128 06:14:41.921762 137274321021824 utils.py:1231] [4400] lr = 0.0004399
+I1128 06:14:41.921820 137274321021824 utils.py:1231] [4400] uptime = 30271.284182983
+I1128 06:14:41.921872 137274321021824 utils.py:1231] [4400] examples_seen = 4505600.0
+I1128 06:14:41.921925 137274321021824 utils.py:1231] [4400] progress = 0.039075335470635776
+I1128 06:14:41.921972 137274321021824 utils.py:1231] [4400] epoch = 3.5167936732682
+I1128 06:14:41.922022 137274321021824 utils.py:1231] [4400] img/sec/core = 164.21712678619187
+I1128 06:14:41.922077 137274321021824 utils.py:1231] [4400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.374372448471943
+I1128 06:14:41.922126 137274321021824 utils.py:1231] [4400] core_hours = 8.374372448471943
+I1128 06:14:41.922184 137274321021824 train.py:125] NOTE: Steps:4400/112603 [3.9%]
+Walltime:8h24m (0s eval)
+ETA:8d14h1m
+Total train time:8d22h24m
+I1128 06:19:53.721743 137274321021824 utils.py:1231] [4450] l2_params = 212.7402662610677
+I1128 06:19:53.721976 137274321021824 utils.py:1231] [4450] train/loss = 5.391070783138275
+I1128 06:19:53.722127 137274321021824 utils.py:1231] [4450] l2_grads = 1.4320721626281738
+I1128 06:19:53.722222 137274321021824 utils.py:1231] [4450] lr = 0.0004449
+I1128 06:19:53.722309 137274321021824 utils.py:1231] [4450] uptime = 30583.084666081006
+I1128 06:19:53.722391 137274321021824 utils.py:1231] [4450] examples_seen = 4556800.0
+I1128 06:19:53.722480 137274321021824 utils.py:1231] [4450] progress = 0.03951937337371118
+I1128 06:19:53.722564 137274321021824 utils.py:1231] [4450] epoch = 3.556757237737157
+I1128 06:19:53.722648 137274321021824 utils.py:1231] [4450] img/sec/core = 164.20757110856252
+I1128 06:19:53.722742 137274321021824 utils.py:1231] [4450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.460983693776946
+I1128 06:19:53.722830 137274321021824 utils.py:1231] [4450] core_hours = 8.460983693776946
+I1128 06:19:53.722945 137274321021824 train.py:125] NOTE: Steps:4450/112603 [4.0%]
+Walltime:8h29m (0s eval)
+ETA:8d13h43m
+Total train time:8d22h11m
+I1128 06:25:05.533653 137274321021824 utils.py:1231] [4500] l2_params = 213.02985243077572
+I1128 06:25:05.533920 137274321021824 utils.py:1231] [4500] train/loss = 5.921860992908478
+I1128 06:25:05.534104 137274321021824 utils.py:1231] [4500] l2_grads = 1.1099419593811035
+I1128 06:25:05.534203 137274321021824 utils.py:1231] [4500] lr = 0.00044990000000000004
+I1128 06:25:05.534303 137274321021824 utils.py:1231] [4500] uptime = 30894.89666226601
+I1128 06:25:05.534379 137274321021824 utils.py:1231] [4500] examples_seen = 4608000.0
+I1128 06:25:05.534438 137274321021824 utils.py:1231] [4500] progress = 0.039963411276786584
+I1128 06:25:05.534501 137274321021824 utils.py:1231] [4500] epoch = 3.5967208022061135
+I1128 06:25:05.534570 137274321021824 utils.py:1231] [4500] img/sec/core = 164.20150804468187
+I1128 06:25:05.534648 137274321021824 utils.py:1231] [4500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.547598137161668
+I1128 06:25:05.534702 137274321021824 utils.py:1231] [4500] core_hours = 8.547598137161668
+I1128 06:25:05.534787 137274321021824 train.py:125] NOTE: Steps:4500/112603 [4.0%]
+Walltime:8h34m (0s eval)
+ETA:8d13h25m
+Total train time:8d21h58m
+I1128 06:30:17.239208 137274321021824 utils.py:1231] [4550] l2_params = 213.24275763520257
+I1128 06:30:17.239424 137274321021824 utils.py:1231] [4550] train/loss = 5.294945240020752
+I1128 06:30:17.239569 137274321021824 utils.py:1231] [4550] l2_grads = 1.6424120664596558
+I1128 06:30:17.239653 137274321021824 utils.py:1231] [4550] lr = 0.00045490000000000005
+I1128 06:30:17.239733 137274321021824 utils.py:1231] [4550] uptime = 31206.602090627
+I1128 06:30:17.239809 137274321021824 utils.py:1231] [4550] examples_seen = 4659200.0
+I1128 06:30:17.239877 137274321021824 utils.py:1231] [4550] progress = 0.04040744917986199
+I1128 06:30:17.239964 137274321021824 utils.py:1231] [4550] epoch = 3.6366843666750706
+I1128 06:30:17.240047 137274321021824 utils.py:1231] [4550] img/sec/core = 164.2576462951571
+I1128 06:30:17.240130 137274321021824 utils.py:1231] [4550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.634182978373055
+I1128 06:30:17.240218 137274321021824 utils.py:1231] [4550] core_hours = 8.634182978373055
+I1128 06:30:17.240314 137274321021824 train.py:125] NOTE: Steps:4550/112603 [4.0%]
+Walltime:8h40m (0s eval)
+ETA:8d13h8m
+Total train time:8d21h46m
+I1128 06:35:28.971472 137274321021824 utils.py:1231] [4600] l2_params = 213.5136421076753
+I1128 06:35:28.971706 137274321021824 utils.py:1231] [4600] train/loss = 5.338135123252869
+I1128 06:35:28.971832 137274321021824 utils.py:1231] [4600] l2_grads = 1.776111125946045
+I1128 06:35:28.971910 137274321021824 utils.py:1231] [4600] lr = 0.0004599
+I1128 06:35:28.971976 137274321021824 utils.py:1231] [4600] uptime = 31518.334336387008
+I1128 06:35:28.972031 137274321021824 utils.py:1231] [4600] examples_seen = 4710400.0
+I1128 06:35:28.972078 137274321021824 utils.py:1231] [4600] progress = 0.0408514870829374
+I1128 06:35:28.972126 137274321021824 utils.py:1231] [4600] epoch = 3.676647931144027
+I1128 06:35:28.972176 137274321021824 utils.py:1231] [4600] img/sec/core = 164.2435156978175
+I1128 06:35:28.972248 137274321021824 utils.py:1231] [4600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.720775268861946
+I1128 06:35:28.972305 137274321021824 utils.py:1231] [4600] core_hours = 8.720775268861946
+I1128 06:35:28.972370 137274321021824 train.py:125] NOTE: Steps:4600/112603 [4.1%]
+Walltime:8h45m (0s eval)
+ETA:8d12h50m
+Total train time:8d21h34m
+I1128 06:40:40.761720 137274321021824 utils.py:1231] [4650] l2_params = 213.7518509557053
+I1128 06:40:40.761956 137274321021824 utils.py:1231] [4650] train/loss = 6.153346300125122
+I1128 06:40:40.762104 137274321021824 utils.py:1231] [4650] l2_grads = 1.3430322408676147
+I1128 06:40:40.762190 137274321021824 utils.py:1231] [4650] lr = 0.00046489999999999997
+I1128 06:40:40.762245 137274321021824 utils.py:1231] [4650] uptime = 31830.124607277
+I1128 06:40:40.762298 137274321021824 utils.py:1231] [4650] examples_seen = 4761600.0
+I1128 06:40:40.762352 137274321021824 utils.py:1231] [4650] progress = 0.041295524986012806
+I1128 06:40:40.762400 137274321021824 utils.py:1231] [4650] epoch = 3.7166114956129843
+I1128 06:40:40.762450 137274321021824 utils.py:1231] [4650] img/sec/core = 164.21294947354082
+I1128 06:40:40.762505 137274321021824 utils.py:1231] [4650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.807383677442498
+I1128 06:40:40.762554 137274321021824 utils.py:1231] [4650] core_hours = 8.807383677442498
+I1128 06:40:40.762617 137274321021824 train.py:125] NOTE: Steps:4650/112603 [4.1%]
+Walltime:8h50m (0s eval)
+ETA:8d12h33m
+Total train time:8d21h22m
+I1128 06:45:52.575864 137274321021824 utils.py:1231] [4700] l2_params = 214.00109331572548
+I1128 06:45:52.576169 137274321021824 utils.py:1231] [4700] train/loss = 6.071167469024658
+I1128 06:45:52.576347 137274321021824 utils.py:1231] [4700] l2_grads = 1.1703828573226929
+I1128 06:45:52.576436 137274321021824 utils.py:1231] [4700] lr = 0.0004699
+I1128 06:45:52.576502 137274321021824 utils.py:1231] [4700] uptime = 32141.938862780997
+I1128 06:45:52.576575 137274321021824 utils.py:1231] [4700] examples_seen = 4812800.0
+I1128 06:45:52.576640 137274321021824 utils.py:1231] [4700] progress = 0.041739562889088214
+I1128 06:45:52.576700 137274321021824 utils.py:1231] [4700] epoch = 3.756575060081941
+I1128 06:45:52.576775 137274321021824 utils.py:1231] [4700] img/sec/core = 164.20031828642257
+I1128 06:45:52.576834 137274321021824 utils.py:1231] [4700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.893998748415832
+I1128 06:45:52.576910 137274321021824 utils.py:1231] [4700] core_hours = 8.893998748415832
+I1128 06:45:52.576995 137274321021824 train.py:125] NOTE: Steps:4700/112603 [4.2%]
+Walltime:8h55m (0s eval)
+ETA:8d12h16m
+Total train time:8d21h10m
+I1128 06:51:04.342252 137274321021824 utils.py:1231] [4750] l2_params = 214.23016084927087
+I1128 06:51:04.342473 137274321021824 utils.py:1231] [4750] train/loss = 5.128497779369354
+I1128 06:51:04.342581 137274321021824 utils.py:1231] [4750] l2_grads = 1.7852028608322144
+I1128 06:51:04.342662 137274321021824 utils.py:1231] [4750] lr = 0.0004749
+I1128 06:51:04.342730 137274321021824 utils.py:1231] [4750] uptime = 32453.705091448006
+I1128 06:51:04.342788 137274321021824 utils.py:1231] [4750] examples_seen = 4864000.0
+I1128 06:51:04.342857 137274321021824 utils.py:1231] [4750] progress = 0.04218360079216362
+I1128 06:51:04.342924 137274321021824 utils.py:1231] [4750] epoch = 3.7965386245508976
+I1128 06:51:04.342980 137274321021824 utils.py:1231] [4750] img/sec/core = 164.22561295016212
+I1128 06:51:04.343041 137274321021824 utils.py:1231] [4750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 8.980600478601112
+I1128 06:51:04.343096 137274321021824 utils.py:1231] [4750] core_hours = 8.980600478601112
+I1128 06:51:04.343157 137274321021824 train.py:125] NOTE: Steps:4750/112603 [4.2%]
+Walltime:9h0m (0s eval)
+ETA:8d11h59m
+Total train time:8d20h58m
+I1128 06:56:16.121146 137274321021824 utils.py:1231] [4800] l2_params = 214.4954645195314
+I1128 06:56:16.121402 137274321021824 utils.py:1231] [4800] train/loss = 5.166594684123993
+I1128 06:56:16.121537 137274321021824 utils.py:1231] [4800] l2_grads = 1.428576946258545
+I1128 06:56:16.121644 137274321021824 utils.py:1231] [4800] lr = 0.0004799
+I1128 06:56:16.121720 137274321021824 utils.py:1231] [4800] uptime = 32765.484078670997
+I1128 06:56:16.121793 137274321021824 utils.py:1231] [4800] examples_seen = 4915200.0
+I1128 06:56:16.121877 137274321021824 utils.py:1231] [4800] progress = 0.04262763869523903
+I1128 06:56:16.121943 137274321021824 utils.py:1231] [4800] epoch = 3.8365021890198547
+I1128 06:56:16.121999 137274321021824 utils.py:1231] [4800] img/sec/core = 164.2188925431995
+I1128 06:56:16.122061 137274321021824 utils.py:1231] [4800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.067205752829722
+I1128 06:56:16.122129 137274321021824 utils.py:1231] [4800] core_hours = 9.067205752829722
+I1128 06:56:16.122206 137274321021824 train.py:125] NOTE: Steps:4800/112603 [4.3%]
+Walltime:9h6m (0s eval)
+ETA:8d11h43m
+Total train time:8d20h47m
+I1128 07:01:27.894151 137274321021824 utils.py:1231] [4850] l2_params = 214.80069246804325
+I1128 07:01:27.894384 137274321021824 utils.py:1231] [4850] train/loss = 5.297779619693756
+I1128 07:01:27.894506 137274321021824 utils.py:1231] [4850] l2_grads = 1.3856258392333984
+I1128 07:01:27.894587 137274321021824 utils.py:1231] [4850] lr = 0.0004849
+I1128 07:01:27.894640 137274321021824 utils.py:1231] [4850] uptime = 33077.257002925006
+I1128 07:01:27.894696 137274321021824 utils.py:1231] [4850] examples_seen = 4966400.0
+I1128 07:01:27.894744 137274321021824 utils.py:1231] [4850] progress = 0.04307167659831443
+I1128 07:01:27.894792 137274321021824 utils.py:1231] [4850] epoch = 3.8764657534888114
+I1128 07:01:27.894841 137274321021824 utils.py:1231] [4850] img/sec/core = 164.22208606635212
+I1128 07:01:27.894900 137274321021824 utils.py:1231] [4850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.153809342900278
+I1128 07:01:27.894950 137274321021824 utils.py:1231] [4850] core_hours = 9.153809342900278
+I1128 07:01:27.895019 137274321021824 train.py:125] NOTE: Steps:4850/112603 [4.3%]
+Walltime:9h11m (0s eval)
+ETA:8d11h27m
+Total train time:8d20h36m
+I1128 07:06:39.681039 137274321021824 utils.py:1231] [4900] l2_params = 215.0843066374724
+I1128 07:06:39.681295 137274321021824 utils.py:1231] [4900] train/loss = 5.398894131183624
+I1128 07:06:39.681428 137274321021824 utils.py:1231] [4900] l2_grads = 1.630181908607483
+I1128 07:06:39.681510 137274321021824 utils.py:1231] [4900] lr = 0.0004899
+I1128 07:06:39.681584 137274321021824 utils.py:1231] [4900] uptime = 33389.043946045
+I1128 07:06:39.681645 137274321021824 utils.py:1231] [4900] examples_seen = 5017600.0
+I1128 07:06:39.681695 137274321021824 utils.py:1231] [4900] progress = 0.043515714501389836
+I1128 07:06:39.681746 137274321021824 utils.py:1231] [4900] epoch = 3.916429317957768
+I1128 07:06:39.681805 137274321021824 utils.py:1231] [4900] img/sec/core = 164.21470215414115
+I1128 07:06:39.681865 137274321021824 utils.py:1231] [4900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.240416827100278
+I1128 07:06:39.681923 137274321021824 utils.py:1231] [4900] core_hours = 9.240416827100278
+I1128 07:06:39.681999 137274321021824 train.py:125] NOTE: Steps:4900/112603 [4.4%]
+Walltime:9h16m (0s eval)
+ETA:8d11h11m
+Total train time:8d20h26m
+I1128 07:11:51.471251 137274321021824 utils.py:1231] [4950] l2_params = 215.36882583238852
+I1128 07:11:51.471473 137274321021824 utils.py:1231] [4950] train/loss = 5.103683412075043
+I1128 07:11:51.471580 137274321021824 utils.py:1231] [4950] l2_grads = 1.6695133447647095
+I1128 07:11:51.471646 137274321021824 utils.py:1231] [4950] lr = 0.0004949
+I1128 07:11:51.471702 137274321021824 utils.py:1231] [4950] uptime = 33700.834063696006
+I1128 07:11:51.471758 137274321021824 utils.py:1231] [4950] examples_seen = 5068800.0
+I1128 07:11:51.471811 137274321021824 utils.py:1231] [4950] progress = 0.043959752404465244
+I1128 07:11:51.471863 137274321021824 utils.py:1231] [4950] epoch = 3.956392882426725
+I1128 07:11:51.471923 137274321021824 utils.py:1231] [4950] img/sec/core = 164.21303018112135
+I1128 07:11:51.471987 137274321021824 utils.py:1231] [4950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.327025193114446
+I1128 07:11:51.472059 137274321021824 utils.py:1231] [4950] core_hours = 9.327025193114446
+I1128 07:11:51.472126 137274321021824 train.py:125] NOTE: Steps:4950/112603 [4.4%]
+Walltime:9h21m (0s eval)
+ETA:8d10h55m
+Total train time:8d20h15m
+I1128 07:17:03.256979 137274321021824 utils.py:1231] [5000] l2_params = 215.66799134723584
+I1128 07:17:03.257267 137274321021824 utils.py:1231] [5000] train/loss = 5.082447826862335
+I1128 07:17:03.257505 137274321021824 utils.py:1231] [5000] l2_grads = 1.4169096946716309
+I1128 07:17:03.257624 137274321021824 utils.py:1231] [5000] lr = 0.0004999000000000001
+I1128 07:17:03.257708 137274321021824 utils.py:1231] [5000] uptime = 34012.62006582301
+I1128 07:17:03.257805 137274321021824 utils.py:1231] [5000] examples_seen = 5120000.0
+I1128 07:17:03.257924 137274321021824 utils.py:1231] [5000] progress = 0.04440379030754065
+I1128 07:17:03.258010 137274321021824 utils.py:1231] [5000] epoch = 3.996356446895682
+I1128 07:17:03.258099 137274321021824 utils.py:1231] [5000] img/sec/core = 164.2151977661404
+I1128 07:17:03.258198 137274321021824 utils.py:1231] [5000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.413632415927502
+I1128 07:17:03.258295 137274321021824 utils.py:1231] [5000] core_hours = 9.413632415927502
+I1128 07:17:03.258410 137274321021824 train.py:125] NOTE: Steps:5000/112603 [4.4%]
+Walltime:9h26m (0s eval)
+ETA:8d10h40m
+Total train time:8d20h5m
+I1128 07:17:03.627286 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:5000/112603 [4.4%]
+Walltime:9h26m (0s eval)
+ETA:8d10h40m
+Total train time:8d20h5m
+I1128 07:18:38.810895 137274321021824 utils.py:1231] [5000] val/acc@1 = 0.1714764030612245
+I1128 07:18:38.811139 137274321021824 utils.py:1231] [5000] val/loss = 4.34275604632436
+I1128 07:18:38.811282 137274321021824 utils.py:1231] [5000] z/secs/eval/val = 95.18373591698764
+I1128 07:18:38.811347 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 95.18373591698764
+I1128 07:23:49.783586 137274321021824 utils.py:1231] [5050] l2_params = 215.957720582196
+I1128 07:23:49.783810 137274321021824 utils.py:1231] [5050] train/loss = 5.235377728939056
+I1128 07:23:49.783901 137274321021824 utils.py:1231] [5050] l2_grads = 1.2735220193862915
+I1128 07:23:49.783963 137274321021824 utils.py:1231] [5050] lr = 0.0005049000000000001
+I1128 07:23:49.784037 137274321021824 utils.py:1231] [5050] uptime = 34419.146394597
+I1128 07:23:49.784100 137274321021824 utils.py:1231] [5050] examples_seen = 5171200.0
+I1128 07:23:49.784149 137274321021824 utils.py:1231] [5050] progress = 0.04484782821061606
+I1128 07:23:49.784195 137274321021824 utils.py:1231] [5050] epoch = 4.036320011364639
+I1128 07:23:49.784243 137274321021824 utils.py:1231] [5050] img/sec/core = 125.94510213990172
+I1128 07:23:49.784296 137274321021824 utils.py:1231] [5050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.5265563961425
+I1128 07:23:49.784343 137274321021824 utils.py:1231] [5050] core_hours = 9.5265563961425
+I1128 07:23:49.784401 137274321021824 train.py:125] NOTE: Steps:5050/112603 [4.5%]
+Walltime:9h33m (0s eval)
+ETA:8d10h58m
+Total train time:8d20h30m
+I1128 07:29:01.542865 137274321021824 utils.py:1231] [5100] l2_params = 216.2385606510307
+I1128 07:29:01.543072 137274321021824 utils.py:1231] [5100] train/loss = 5.078700840473175
+I1128 07:29:01.543172 137274321021824 utils.py:1231] [5100] l2_grads = 1.8423705101013184
+I1128 07:29:01.543239 137274321021824 utils.py:1231] [5100] lr = 0.0005099
+I1128 07:29:01.543297 137274321021824 utils.py:1231] [5100] uptime = 34730.905658956006
+I1128 07:29:01.543356 137274321021824 utils.py:1231] [5100] examples_seen = 5222400.0
+I1128 07:29:01.543411 137274321021824 utils.py:1231] [5100] progress = 0.045291866113691466
+I1128 07:29:01.543464 137274321021824 utils.py:1231] [5100] epoch = 4.076283575833595
+I1128 07:29:01.543519 137274321021824 utils.py:1231] [5100] img/sec/core = 164.22928154282897
+I1128 07:29:01.543586 137274321021824 utils.py:1231] [5100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.613156191797778
+I1128 07:29:01.543639 137274321021824 utils.py:1231] [5100] core_hours = 9.613156191797778
+I1128 07:29:01.543702 137274321021824 train.py:125] NOTE: Steps:5100/112603 [4.5%]
+Walltime:9h38m (0s eval)
+ETA:8d10h42m
+Total train time:8d20h19m
+I1128 07:34:13.298231 137274321021824 utils.py:1231] [5150] l2_params = 216.5699611034548
+I1128 07:34:13.298470 137274321021824 utils.py:1231] [5150] train/loss = 6.125650227069855
+I1128 07:34:13.298600 137274321021824 utils.py:1231] [5150] l2_grads = 1.0313384532928467
+I1128 07:34:13.298690 137274321021824 utils.py:1231] [5150] lr = 0.0005149
+I1128 07:34:13.298760 137274321021824 utils.py:1231] [5150] uptime = 35042.66112206201
+I1128 07:34:13.298839 137274321021824 utils.py:1231] [5150] examples_seen = 5273600.0
+I1128 07:34:13.298905 137274321021824 utils.py:1231] [5150] progress = 0.04573590401676687
+I1128 07:34:13.298961 137274321021824 utils.py:1231] [5150] epoch = 4.116247140302552
+I1128 07:34:13.299015 137274321021824 utils.py:1231] [5150] img/sec/core = 164.23128400027653
+I1128 07:34:13.299077 137274321021824 utils.py:1231] [5150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.699754931549446
+I1128 07:34:13.299131 137274321021824 utils.py:1231] [5150] core_hours = 9.699754931549446
+I1128 07:34:13.299192 137274321021824 train.py:125] NOTE: Steps:5150/112603 [4.6%]
+Walltime:9h44m (0s eval)
+ETA:8d10h27m
+Total train time:8d20h9m
+I1128 07:39:25.052022 137274321021824 utils.py:1231] [5200] l2_params = 216.8762341297081
+I1128 07:39:25.052224 137274321021824 utils.py:1231] [5200] train/loss = 5.339911222457886
+I1128 07:39:25.052322 137274321021824 utils.py:1231] [5200] l2_grads = 1.2280982732772827
+I1128 07:39:25.052398 137274321021824 utils.py:1231] [5200] lr = 0.0005199
+I1128 07:39:25.052494 137274321021824 utils.py:1231] [5200] uptime = 35354.414852713
+I1128 07:39:25.052569 137274321021824 utils.py:1231] [5200] examples_seen = 5324800.0
+I1128 07:39:25.052629 137274321021824 utils.py:1231] [5200] progress = 0.04617994191984228
+I1128 07:39:25.052684 137274321021824 utils.py:1231] [5200] epoch = 4.156210704771509
+I1128 07:39:25.052746 137274321021824 utils.py:1231] [5200] img/sec/core = 164.2321966543473
+I1128 07:39:25.052800 137274321021824 utils.py:1231] [5200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.786353190063611
+I1128 07:39:25.052848 137274321021824 utils.py:1231] [5200] core_hours = 9.786353190063611
+I1128 07:39:25.052928 137274321021824 train.py:125] NOTE: Steps:5200/112603 [4.6%]
+Walltime:9h49m (0s eval)
+ETA:8d10h12m
+Total train time:8d19h59m
+I1128 07:44:36.813964 137274321021824 utils.py:1231] [5250] l2_params = 217.1611917770231
+I1128 07:44:36.814199 137274321021824 utils.py:1231] [5250] train/loss = 5.097334861755371
+I1128 07:44:36.814419 137274321021824 utils.py:1231] [5250] l2_grads = 1.691284418106079
+I1128 07:44:36.814497 137274321021824 utils.py:1231] [5250] lr = 0.0005249
+I1128 07:44:36.814557 137274321021824 utils.py:1231] [5250] uptime = 35666.17691856601
+I1128 07:44:36.814616 137274321021824 utils.py:1231] [5250] examples_seen = 5376000.0
+I1128 07:44:36.814672 137274321021824 utils.py:1231] [5250] progress = 0.04662397982291768
+I1128 07:44:36.814726 137274321021824 utils.py:1231] [5250] epoch = 4.1961742692404655
+I1128 07:44:36.814790 137274321021824 utils.py:1231] [5250] img/sec/core = 164.22780577846302
+I1128 07:44:36.814849 137274321021824 utils.py:1231] [5250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.872953763911669
+I1128 07:44:36.814913 137274321021824 utils.py:1231] [5250] core_hours = 9.872953763911669
+I1128 07:44:36.814985 137274321021824 train.py:125] NOTE: Steps:5250/112603 [4.7%]
+Walltime:9h54m (0s eval)
+ETA:8d9h57m
+Total train time:8d19h50m
+I1128 07:49:48.571016 137274321021824 utils.py:1231] [5300] l2_params = 217.4692455477241
+I1128 07:49:48.571312 137274321021824 utils.py:1231] [5300] train/loss = 5.366478383541107
+I1128 07:49:48.571475 137274321021824 utils.py:1231] [5300] l2_grads = 1.2400307655334473
+I1128 07:49:48.571552 137274321021824 utils.py:1231] [5300] lr = 0.0005299
+I1128 07:49:48.571619 137274321021824 utils.py:1231] [5300] uptime = 35977.93398109301
+I1128 07:49:48.571674 137274321021824 utils.py:1231] [5300] examples_seen = 5427200.0
+I1128 07:49:48.571749 137274321021824 utils.py:1231] [5300] progress = 0.04706801772599309
+I1128 07:49:48.571799 137274321021824 utils.py:1231] [5300] epoch = 4.236137833709423
+I1128 07:49:48.571850 137274321021824 utils.py:1231] [5300] img/sec/core = 164.2304414372825
+I1128 07:49:48.571910 137274321021824 utils.py:1231] [5300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 9.959552947946946
+I1128 07:49:48.571961 137274321021824 utils.py:1231] [5300] core_hours = 9.959552947946946
+I1128 07:49:48.572030 137274321021824 train.py:125] NOTE: Steps:5300/112603 [4.7%]
+Walltime:9h59m (0s eval)
+ETA:8d9h42m
+Total train time:8d19h40m
+I1128 07:54:58.001293 137274321021824 utils.py:1231] [5350] l2_params = 217.73030123328803
+I1128 07:54:58.001510 137274321021824 utils.py:1231] [5350] train/loss = 4.98268449306488
+I1128 07:54:58.001604 137274321021824 utils.py:1231] [5350] l2_grads = 1.3429676294326782
+I1128 07:54:58.001666 137274321021824 utils.py:1231] [5350] lr = 0.0005349
+I1128 07:54:58.001730 137274321021824 utils.py:1231] [5350] uptime = 36287.36408926701
+I1128 07:54:58.001788 137274321021824 utils.py:1231] [5350] examples_seen = 5478400.0
+I1128 07:54:58.001843 137274321021824 utils.py:1231] [5350] progress = 0.047512055629068496
+I1128 07:54:58.001897 137274321021824 utils.py:1231] [5350] epoch = 4.27610139817838
+I1128 07:54:58.001949 137274321021824 utils.py:1231] [5350] img/sec/core = 165.46547555485256
+I1128 07:54:58.002005 137274321021824 utils.py:1231] [5350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.045505755773057
+I1128 07:54:58.002055 137274321021824 utils.py:1231] [5350] core_hours = 10.045505755773057
+I1128 07:54:58.002114 137274321021824 train.py:125] NOTE: Steps:5350/112603 [4.8%]
+Walltime:10h4m (0s eval)
+ETA:8d9h27m
+Total train time:8d19h30m
+I1128 08:00:09.761590 137274321021824 utils.py:1231] [5400] l2_params = 217.99084677410562
+I1128 08:00:09.761811 137274321021824 utils.py:1231] [5400] train/loss = 5.342824220657349
+I1128 08:00:09.761909 137274321021824 utils.py:1231] [5400] l2_grads = 1.4111391305923462
+I1128 08:00:09.761975 137274321021824 utils.py:1231] [5400] lr = 0.0005399000000000001
+I1128 08:00:09.762029 137274321021824 utils.py:1231] [5400] uptime = 36599.12439121101
+I1128 08:00:09.762079 137274321021824 utils.py:1231] [5400] examples_seen = 5529600.0
+I1128 08:00:09.762125 137274321021824 utils.py:1231] [5400] progress = 0.047956093532143904
+I1128 08:00:09.762171 137274321021824 utils.py:1231] [5400] epoch = 4.316064962647336
+I1128 08:00:09.762219 137274321021824 utils.py:1231] [5400] img/sec/core = 164.22873496317345
+I1128 08:00:09.762275 137274321021824 utils.py:1231] [5400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.13210583964639
+I1128 08:00:09.762322 137274321021824 utils.py:1231] [5400] core_hours = 10.13210583964639
+I1128 08:00:09.762388 137274321021824 train.py:125] NOTE: Steps:5400/112603 [4.8%]
+Walltime:10h9m (0s eval)
+ETA:8d9h13m
+Total train time:8d19h21m
+I1128 08:05:21.564261 137274321021824 utils.py:1231] [5450] l2_params = 218.35166518264552
+I1128 08:05:21.564548 137274321021824 utils.py:1231] [5450] train/loss = 4.882747173309326
+I1128 08:05:21.564754 137274321021824 utils.py:1231] [5450] l2_grads = 1.772116780281067
+I1128 08:05:21.564857 137274321021824 utils.py:1231] [5450] lr = 0.0005449000000000001
+I1128 08:05:21.564941 137274321021824 utils.py:1231] [5450] uptime = 36910.92729900101
+I1128 08:05:21.565015 137274321021824 utils.py:1231] [5450] examples_seen = 5580800.0
+I1128 08:05:21.565085 137274321021824 utils.py:1231] [5450] progress = 0.04840013143521931
+I1128 08:05:21.565161 137274321021824 utils.py:1231] [5450] epoch = 4.356028527116293
+I1128 08:05:21.565242 137274321021824 utils.py:1231] [5450] img/sec/core = 164.20629417119952
+I1128 08:05:21.565324 137274321021824 utils.py:1231] [5450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.218717758476945
+I1128 08:05:21.565405 137274321021824 utils.py:1231] [5450] core_hours = 10.218717758476945
+I1128 08:05:21.565508 137274321021824 train.py:125] NOTE: Steps:5450/112603 [4.8%]
+Walltime:10h15m (0s eval)
+ETA:8d8h59m
+Total train time:8d19h12m
+I1128 08:10:33.345701 137274321021824 utils.py:1231] [5500] l2_params = 218.70481541318202
+I1128 08:10:33.345919 137274321021824 utils.py:1231] [5500] train/loss = 5.435912609100342
+I1128 08:10:33.346026 137274321021824 utils.py:1231] [5500] l2_grads = 1.3068479299545288
+I1128 08:10:33.346095 137274321021824 utils.py:1231] [5500] lr = 0.0005499000000000001
+I1128 08:10:33.346184 137274321021824 utils.py:1231] [5500] uptime = 37222.70854553
+I1128 08:10:33.346242 137274321021824 utils.py:1231] [5500] examples_seen = 5632000.0
+I1128 08:10:33.346310 137274321021824 utils.py:1231] [5500] progress = 0.04884416933829472
+I1128 08:10:33.346368 137274321021824 utils.py:1231] [5500] epoch = 4.39599209158525
+I1128 08:10:33.346423 137274321021824 utils.py:1231] [5500] img/sec/core = 164.21770253984585
+I1128 08:10:33.346485 137274321021824 utils.py:1231] [5500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.305323660290554
+I1128 08:10:33.346541 137274321021824 utils.py:1231] [5500] core_hours = 10.305323660290554
+I1128 08:10:33.346602 137274321021824 train.py:125] NOTE: Steps:5500/112603 [4.9%]
+Walltime:10h20m (0s eval)
+ETA:8d8h45m
+Total train time:8d19h3m
+I1128 08:15:45.127648 137274321021824 utils.py:1231] [5550] l2_params = 219.06017769275223
+I1128 08:15:45.127866 137274321021824 utils.py:1231] [5550] train/loss = 4.944257915019989
+I1128 08:15:45.127974 137274321021824 utils.py:1231] [5550] l2_grads = 1.3854482173919678
+I1128 08:15:45.128053 137274321021824 utils.py:1231] [5550] lr = 0.0005549
+I1128 08:15:45.128132 137274321021824 utils.py:1231] [5550] uptime = 37534.490488544005
+I1128 08:15:45.128209 137274321021824 utils.py:1231] [5550] examples_seen = 5683200.0
+I1128 08:15:45.128264 137274321021824 utils.py:1231] [5550] progress = 0.049288207241370126
+I1128 08:15:45.128324 137274321021824 utils.py:1231] [5550] epoch = 4.435955656054206
+I1128 08:15:45.128379 137274321021824 utils.py:1231] [5550] img/sec/core = 164.2173356963769
+I1128 08:15:45.128444 137274321021824 utils.py:1231] [5550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.391929755572223
+I1128 08:15:45.128503 137274321021824 utils.py:1231] [5550] core_hours = 10.391929755572223
+I1128 08:15:45.128582 137274321021824 train.py:125] NOTE: Steps:5550/112603 [4.9%]
+Walltime:10h25m (0s eval)
+ETA:8d8h31m
+Total train time:8d18h54m
+I1128 08:20:56.894094 137274321021824 utils.py:1231] [5600] l2_params = 219.38282532485772
+I1128 08:20:56.894324 137274321021824 utils.py:1231] [5600] train/loss = 5.061885595321655
+I1128 08:20:56.894428 137274321021824 utils.py:1231] [5600] l2_grads = 1.3571323156356812
+I1128 08:20:56.894507 137274321021824 utils.py:1231] [5600] lr = 0.0005599
+I1128 08:20:56.894569 137274321021824 utils.py:1231] [5600] uptime = 37846.256930809
+I1128 08:20:56.894629 137274321021824 utils.py:1231] [5600] examples_seen = 5734400.0
+I1128 08:20:56.894685 137274321021824 utils.py:1231] [5600] progress = 0.04973224514444553
+I1128 08:20:56.894744 137274321021824 utils.py:1231] [5600] epoch = 4.4759192205231635
+I1128 08:20:56.894800 137274321021824 utils.py:1231] [5600] img/sec/core = 164.2255004356133
+I1128 08:20:56.894861 137274321021824 utils.py:1231] [5600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.478531545090277
+I1128 08:20:56.894933 137274321021824 utils.py:1231] [5600] core_hours = 10.478531545090277
+I1128 08:20:56.894999 137274321021824 train.py:125] NOTE: Steps:5600/112603 [5.0%]
+Walltime:10h30m (0s eval)
+ETA:8d8h17m
+Total train time:8d18h46m
+I1128 08:26:08.667027 137274321021824 utils.py:1231] [5650] l2_params = 219.7031855592518
+I1128 08:26:08.667264 137274321021824 utils.py:1231] [5650] train/loss = 5.188402950763702
+I1128 08:26:08.667386 137274321021824 utils.py:1231] [5650] l2_grads = 1.4021399021148682
+I1128 08:26:08.667462 137274321021824 utils.py:1231] [5650] lr = 0.0005649
+I1128 08:26:08.667516 137274321021824 utils.py:1231] [5650] uptime = 38158.029878136
+I1128 08:26:08.667568 137274321021824 utils.py:1231] [5650] examples_seen = 5785600.0
+I1128 08:26:08.667623 137274321021824 utils.py:1231] [5650] progress = 0.050176283047520934
+I1128 08:26:08.667679 137274321021824 utils.py:1231] [5650] epoch = 4.515882784992121
+I1128 08:26:08.667729 137274321021824 utils.py:1231] [5650] img/sec/core = 164.22207391297235
+I1128 08:26:08.667784 137274321021824 utils.py:1231] [5650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.565135141570002
+I1128 08:26:08.667833 137274321021824 utils.py:1231] [5650] core_hours = 10.565135141570002
+I1128 08:26:08.667896 137274321021824 train.py:125] NOTE: Steps:5650/112603 [5.0%]
+Walltime:10h35m (0s eval)
+ETA:8d8h3m
+Total train time:8d18h38m
+I1128 08:31:20.431902 137274321021824 utils.py:1231] [5700] l2_params = 220.07337080303063
+I1128 08:31:20.432171 137274321021824 utils.py:1231] [5700] train/loss = 5.137414991855621
+I1128 08:31:20.432425 137274321021824 utils.py:1231] [5700] l2_grads = 1.5441851615905762
+I1128 08:31:20.432500 137274321021824 utils.py:1231] [5700] lr = 0.0005698999999999999
+I1128 08:31:20.432551 137274321021824 utils.py:1231] [5700] uptime = 38469.794913473
+I1128 08:31:20.432603 137274321021824 utils.py:1231] [5700] examples_seen = 5836800.0
+I1128 08:31:20.432652 137274321021824 utils.py:1231] [5700] progress = 0.05062032095059634
+I1128 08:31:20.432700 137274321021824 utils.py:1231] [5700] epoch = 4.555846349461078
+I1128 08:31:20.432749 137274321021824 utils.py:1231] [5700] img/sec/core = 164.22624154968568
+I1128 08:31:20.432804 137274321021824 utils.py:1231] [5700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.651736540274722
+I1128 08:31:20.432851 137274321021824 utils.py:1231] [5700] core_hours = 10.651736540274722
+I1128 08:31:20.432916 137274321021824 train.py:125] NOTE: Steps:5700/112603 [5.1%]
+Walltime:10h41m (0s eval)
+ETA:8d7h50m
+Total train time:8d18h29m
+I1128 08:36:30.072098 137274321021824 utils.py:1231] [5750] l2_params = 220.40032555961469
+I1128 08:36:30.072315 137274321021824 utils.py:1231] [5750] train/loss = 4.981255829334259
+I1128 08:36:30.072410 137274321021824 utils.py:1231] [5750] l2_grads = 1.8548132181167603
+I1128 08:36:30.072527 137274321021824 utils.py:1231] [5750] lr = 0.0005748999999999999
+I1128 08:36:30.072583 137274321021824 utils.py:1231] [5750] uptime = 38779.43494452801
+I1128 08:36:30.072634 137274321021824 utils.py:1231] [5750] examples_seen = 5888000.0
+I1128 08:36:30.072680 137274321021824 utils.py:1231] [5750] progress = 0.05106435885367175
+I1128 08:36:30.072726 137274321021824 utils.py:1231] [5750] epoch = 4.595809913930034
+I1128 08:36:30.072775 137274321021824 utils.py:1231] [5750] img/sec/core = 165.35329694145426
+I1128 08:36:30.072828 137274321021824 utils.py:1231] [5750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.737747660012223
+I1128 08:36:30.072875 137274321021824 utils.py:1231] [5750] core_hours = 10.737747660012223
+I1128 08:36:30.072943 137274321021824 train.py:125] NOTE: Steps:5750/112603 [5.1%]
+Walltime:10h46m (0s eval)
+ETA:8d7h36m
+Total train time:8d18h21m
+I1128 08:41:41.830082 137274321021824 utils.py:1231] [5800] l2_params = 220.73930753227538
+I1128 08:41:41.830317 137274321021824 utils.py:1231] [5800] train/loss = 5.069477617740631
+I1128 08:41:41.830412 137274321021824 utils.py:1231] [5800] l2_grads = 1.6205769777297974
+I1128 08:41:41.830472 137274321021824 utils.py:1231] [5800] lr = 0.0005799
+I1128 08:41:41.830531 137274321021824 utils.py:1231] [5800] uptime = 39091.19289268801
+I1128 08:41:41.830584 137274321021824 utils.py:1231] [5800] examples_seen = 5939200.0
+I1128 08:41:41.830633 137274321021824 utils.py:1231] [5800] progress = 0.051508396756747156
+I1128 08:41:41.830682 137274321021824 utils.py:1231] [5800] epoch = 4.635773478398991
+I1128 08:41:41.830734 137274321021824 utils.py:1231] [5800] img/sec/core = 164.2299748961772
+I1128 08:41:41.830794 137274321021824 utils.py:1231] [5800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.824347090056667
+I1128 08:41:41.830854 137274321021824 utils.py:1231] [5800] core_hours = 10.824347090056667
+I1128 08:41:41.830926 137274321021824 train.py:125] NOTE: Steps:5800/112603 [5.2%]
+Walltime:10h51m (0s eval)
+ETA:8d7h23m
+Total train time:8d18h13m
+I1128 08:46:50.633038 137274321021824 utils.py:1231] [5850] l2_params = 221.08878181258055
+I1128 08:46:50.633254 137274321021824 utils.py:1231] [5850] train/loss = 4.9202576875686646
+I1128 08:46:50.633351 137274321021824 utils.py:1231] [5850] l2_grads = 1.5966460704803467
+I1128 08:46:50.633407 137274321021824 utils.py:1231] [5850] lr = 0.0005849
+I1128 08:46:50.633457 137274321021824 utils.py:1231] [5850] uptime = 39399.99581936
+I1128 08:46:50.633506 137274321021824 utils.py:1231] [5850] examples_seen = 5990400.0
+I1128 08:46:50.633554 137274321021824 utils.py:1231] [5850] progress = 0.05195243465982256
+I1128 08:46:50.633601 137274321021824 utils.py:1231] [5850] epoch = 4.675737042867948
+I1128 08:46:50.633649 137274321021824 utils.py:1231] [5850] img/sec/core = 165.80153741348258
+I1128 08:46:50.633704 137274321021824 utils.py:1231] [5850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.91012568079889
+I1128 08:46:50.633751 137274321021824 utils.py:1231] [5850] core_hours = 10.91012568079889
+I1128 08:46:50.633808 137274321021824 train.py:125] NOTE: Steps:5850/112603 [5.2%]
+Walltime:10h56m (0s eval)
+ETA:8d7h9m
+Total train time:8d18h4m
+I1128 08:52:02.386903 137274321021824 utils.py:1231] [5900] l2_params = 221.4298805610111
+I1128 08:52:02.387111 137274321021824 utils.py:1231] [5900] train/loss = 4.822545886039734
+I1128 08:52:02.387215 137274321021824 utils.py:1231] [5900] l2_grads = 1.437412977218628
+I1128 08:52:02.387283 137274321021824 utils.py:1231] [5900] lr = 0.0005899
+I1128 08:52:02.387341 137274321021824 utils.py:1231] [5900] uptime = 39711.74970307101
+I1128 08:52:02.387403 137274321021824 utils.py:1231] [5900] examples_seen = 6041600.0
+I1128 08:52:02.387458 137274321021824 utils.py:1231] [5900] progress = 0.05239647256289797
+I1128 08:52:02.387532 137274321021824 utils.py:1231] [5900] epoch = 4.715700607336904
+I1128 08:52:02.387593 137274321021824 utils.py:1231] [5900] img/sec/core = 164.23211602220937
+I1128 08:52:02.387654 137274321021824 utils.py:1231] [5900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 10.996723981829723
+I1128 08:52:02.387707 137274321021824 utils.py:1231] [5900] core_hours = 10.996723981829723
+I1128 08:52:02.387778 137274321021824 train.py:125] NOTE: Steps:5900/112603 [5.2%]
+Walltime:11h1m (0s eval)
+ETA:8d6h56m
+Total train time:8d17h56m
+I1128 08:57:14.154576 137274321021824 utils.py:1231] [5950] l2_params = 221.7708513339344
+I1128 08:57:14.154777 137274321021824 utils.py:1231] [5950] train/loss = 6.329286456108093
+I1128 08:57:14.154871 137274321021824 utils.py:1231] [5950] l2_grads = 1.08024001121521
+I1128 08:57:14.154962 137274321021824 utils.py:1231] [5950] lr = 0.0005949
+I1128 08:57:14.155031 137274321021824 utils.py:1231] [5950] uptime = 40023.51739363601
+I1128 08:57:14.155116 137274321021824 utils.py:1231] [5950] examples_seen = 6092800.0
+I1128 08:57:14.155168 137274321021824 utils.py:1231] [5950] progress = 0.05284051046597338
+I1128 08:57:14.155220 137274321021824 utils.py:1231] [5950] epoch = 4.755664171805861
+I1128 08:57:14.155272 137274321021824 utils.py:1231] [5950] img/sec/core = 164.22484288610175
+I1128 08:57:14.155345 137274321021824 utils.py:1231] [5950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.083326118097778
+I1128 08:57:14.155394 137274321021824 utils.py:1231] [5950] core_hours = 11.083326118097778
+I1128 08:57:14.155454 137274321021824 train.py:125] NOTE: Steps:5950/112603 [5.3%]
+Walltime:11h7m (0s eval)
+ETA:8d6h44m
+Total train time:8d17h49m
+I1128 09:02:25.919705 137274321021824 utils.py:1231] [6000] l2_params = 222.18113231588634
+I1128 09:02:25.919949 137274321021824 utils.py:1231] [6000] train/loss = 5.017584979534149
+I1128 09:02:25.920058 137274321021824 utils.py:1231] [6000] l2_grads = 1.5822887420654297
+I1128 09:02:25.920128 137274321021824 utils.py:1231] [6000] lr = 0.0005999
+I1128 09:02:25.920188 137274321021824 utils.py:1231] [6000] uptime = 40335.282549624
+I1128 09:02:25.920247 137274321021824 utils.py:1231] [6000] examples_seen = 6144000.0
+I1128 09:02:25.920305 137274321021824 utils.py:1231] [6000] progress = 0.05328454836904878
+I1128 09:02:25.920361 137274321021824 utils.py:1231] [6000] epoch = 4.7956277362748185
+I1128 09:02:25.920418 137274321021824 utils.py:1231] [6000] img/sec/core = 164.226177995249
+I1128 09:02:25.920476 137274321021824 utils.py:1231] [6000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.169927550316665
+I1128 09:02:25.920529 137274321021824 utils.py:1231] [6000] core_hours = 11.169927550316665
+I1128 09:02:25.920593 137274321021824 train.py:125] NOTE: Steps:6000/112603 [5.3%]
+Walltime:11h12m (0s eval)
+ETA:8d6h31m
+Total train time:8d17h41m
+I1128 09:07:37.088573 137274321021824 utils.py:1231] [6050] l2_params = 222.5367334714948
+I1128 09:07:37.088813 137274321021824 utils.py:1231] [6050] train/loss = 4.8581371903419495
+I1128 09:07:37.088949 137274321021824 utils.py:1231] [6050] l2_grads = 1.413646936416626
+I1128 09:07:37.089014 137274321021824 utils.py:1231] [6050] lr = 0.0006049
+I1128 09:07:37.089066 137274321021824 utils.py:1231] [6050] uptime = 40646.451427502005
+I1128 09:07:37.089115 137274321021824 utils.py:1231] [6050] examples_seen = 6195200.0
+I1128 09:07:37.089164 137274321021824 utils.py:1231] [6050] progress = 0.053728586272124186
+I1128 09:07:37.089211 137274321021824 utils.py:1231] [6050] epoch = 4.835591300743775
+I1128 09:07:37.089259 137274321021824 utils.py:1231] [6050] img/sec/core = 164.54087680347217
+I1128 09:07:37.089313 137274321021824 utils.py:1231] [6050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.256363349727224
+I1128 09:07:37.089361 137274321021824 utils.py:1231] [6050] core_hours = 11.256363349727224
+I1128 09:07:37.089429 137274321021824 train.py:125] NOTE: Steps:6050/112603 [5.4%]
+Walltime:11h17m (0s eval)
+ETA:8d6h18m
+Total train time:8d17h34m
+I1128 09:12:48.851601 137274321021824 utils.py:1231] [6100] l2_params = 222.83694110006485
+I1128 09:12:48.851932 137274321021824 utils.py:1231] [6100] train/loss = 4.895958304405212
+I1128 09:12:48.852173 137274321021824 utils.py:1231] [6100] l2_grads = 1.2995320558547974
+I1128 09:12:48.852259 137274321021824 utils.py:1231] [6100] lr = 0.0006099
+I1128 09:12:48.852330 137274321021824 utils.py:1231] [6100] uptime = 40958.21468678501
+I1128 09:12:48.852389 137274321021824 utils.py:1231] [6100] examples_seen = 6246400.0
+I1128 09:12:48.852441 137274321021824 utils.py:1231] [6100] progress = 0.054172624175199593
+I1128 09:12:48.852497 137274321021824 utils.py:1231] [6100] epoch = 4.875554865212732
+I1128 09:12:48.852552 137274321021824 utils.py:1231] [6100] img/sec/core = 164.2271771142952
+I1128 09:12:48.852615 137274321021824 utils.py:1231] [6100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.342964255083613
+I1128 09:12:48.852666 137274321021824 utils.py:1231] [6100] core_hours = 11.342964255083613
+I1128 09:12:48.852726 137274321021824 train.py:125] NOTE: Steps:6100/112603 [5.4%]
+Walltime:11h22m (0s eval)
+ETA:8d6h6m
+Total train time:8d17h27m
+I1128 09:18:00.611477 137274321021824 utils.py:1231] [6150] l2_params = 223.21951984178372
+I1128 09:18:00.611697 137274321021824 utils.py:1231] [6150] train/loss = 4.9184330701828
+I1128 09:18:00.611788 137274321021824 utils.py:1231] [6150] l2_grads = 1.3588517904281616
+I1128 09:18:00.611846 137274321021824 utils.py:1231] [6150] lr = 0.0006149
+I1128 09:18:00.611907 137274321021824 utils.py:1231] [6150] uptime = 41269.974268604
+I1128 09:18:00.611957 137274321021824 utils.py:1231] [6150] examples_seen = 6297600.0
+I1128 09:18:00.612002 137274321021824 utils.py:1231] [6150] progress = 0.054616662078275
+I1128 09:18:00.612049 137274321021824 utils.py:1231] [6150] epoch = 4.915518429681689
+I1128 09:18:00.612097 137274321021824 utils.py:1231] [6150] img/sec/core = 164.22911431067905
+I1128 09:18:00.612159 137274321021824 utils.py:1231] [6150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.42956413892222
+I1128 09:18:00.612206 137274321021824 utils.py:1231] [6150] core_hours = 11.42956413892222
+I1128 09:18:00.612262 137274321021824 train.py:125] NOTE: Steps:6150/112603 [5.5%]
+Walltime:11h27m (0s eval)
+ETA:8d5h54m
+Total train time:8d17h20m
+I1128 09:23:12.378784 137274321021824 utils.py:1231] [6200] l2_params = 223.60310288978567
+I1128 09:23:12.379007 137274321021824 utils.py:1231] [6200] train/loss = 4.709064304828644
+I1128 09:23:12.379096 137274321021824 utils.py:1231] [6200] l2_grads = 1.4518834352493286
+I1128 09:23:12.379151 137274321021824 utils.py:1231] [6200] lr = 0.0006199
+I1128 09:23:12.379199 137274321021824 utils.py:1231] [6200] uptime = 41581.741561588
+I1128 09:23:12.379247 137274321021824 utils.py:1231] [6200] examples_seen = 6348800.0
+I1128 09:23:12.379295 137274321021824 utils.py:1231] [6200] progress = 0.05506069998135041
+I1128 09:23:12.379338 137274321021824 utils.py:1231] [6200] epoch = 4.955481994150645
+I1128 09:23:12.379390 137274321021824 utils.py:1231] [6200] img/sec/core = 164.22505231370388
+I1128 09:23:12.379440 137274321021824 utils.py:1231] [6200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.51616616475111
+I1128 09:23:12.379485 137274321021824 utils.py:1231] [6200] core_hours = 11.51616616475111
+I1128 09:23:12.379539 137274321021824 train.py:125] NOTE: Steps:6200/112603 [5.5%]
+Walltime:11h33m (0s eval)
+ETA:8d5h42m
+Total train time:8d17h13m
+I1128 09:28:24.146053 137274321021824 utils.py:1231] [6250] l2_params = 224.02568710946178
+I1128 09:28:24.146320 137274321021824 utils.py:1231] [6250] train/loss = 4.786445915699005
+I1128 09:28:24.146530 137274321021824 utils.py:1231] [6250] l2_grads = 1.4092936515808105
+I1128 09:28:24.146643 137274321021824 utils.py:1231] [6250] lr = 0.0006249000000000001
+I1128 09:28:24.146751 137274321021824 utils.py:1231] [6250] uptime = 41893.50910458401
+I1128 09:28:24.146833 137274321021824 utils.py:1231] [6250] examples_seen = 6400000.0
+I1128 09:28:24.146931 137274321021824 utils.py:1231] [6250] progress = 0.055504737884425816
+I1128 09:28:24.147013 137274321021824 utils.py:1231] [6250] epoch = 4.995445558619602
+I1128 09:28:24.147098 137274321021824 utils.py:1231] [6250] img/sec/core = 164.22492061867936
+I1128 09:28:24.147194 137274321021824 utils.py:1231] [6250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.60276826002778
+I1128 09:28:24.147287 137274321021824 utils.py:1231] [6250] core_hours = 11.60276826002778
+I1128 09:28:24.147395 137274321021824 train.py:125] NOTE: Steps:6250/112603 [5.6%]
+Walltime:11h38m (0s eval)
+ETA:8d5h30m
+Total train time:8d17h6m
+I1128 09:33:35.913871 137274321021824 utils.py:1231] [6300] l2_params = 224.38743607053092
+I1128 09:33:35.914098 137274321021824 utils.py:1231] [6300] train/loss = 4.731373488903046
+I1128 09:33:35.914206 137274321021824 utils.py:1231] [6300] l2_grads = 1.562499761581421
+I1128 09:33:35.914295 137274321021824 utils.py:1231] [6300] lr = 0.0006299000000000001
+I1128 09:33:35.914357 137274321021824 utils.py:1231] [6300] uptime = 42205.276719641
+I1128 09:33:35.914422 137274321021824 utils.py:1231] [6300] examples_seen = 6451200.0
+I1128 09:33:35.914482 137274321021824 utils.py:1231] [6300] progress = 0.05594877578750122
+I1128 09:33:35.914538 137274321021824 utils.py:1231] [6300] epoch = 5.035409123088559
+I1128 09:33:35.914597 137274321021824 utils.py:1231] [6300] img/sec/core = 164.22488266024965
+I1128 09:33:35.914664 137274321021824 utils.py:1231] [6300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.689370375321387
+I1128 09:33:35.914717 137274321021824 utils.py:1231] [6300] core_hours = 11.689370375321387
+I1128 09:33:35.914781 137274321021824 train.py:125] NOTE: Steps:6300/112603 [5.6%]
+Walltime:11h43m (0s eval)
+ETA:8d5h18m
+Total train time:8d16h59m
+I1128 09:38:45.850574 137274321021824 utils.py:1231] [6350] l2_params = 224.82344474329508
+I1128 09:38:45.850805 137274321021824 utils.py:1231] [6350] train/loss = 5.373353719711304
+I1128 09:38:45.850928 137274321021824 utils.py:1231] [6350] l2_grads = 1.222509503364563
+I1128 09:38:45.851002 137274321021824 utils.py:1231] [6350] lr = 0.0006349
+I1128 09:38:45.851060 137274321021824 utils.py:1231] [6350] uptime = 42515.21342215201
+I1128 09:38:45.851132 137274321021824 utils.py:1231] [6350] examples_seen = 6502400.0
+I1128 09:38:45.851192 137274321021824 utils.py:1231] [6350] progress = 0.05639281369057663
+I1128 09:38:45.851250 137274321021824 utils.py:1231] [6350] epoch = 5.075372687557516
+I1128 09:38:45.851313 137274321021824 utils.py:1231] [6350] img/sec/core = 165.19502074195609
+I1128 09:38:45.851389 137274321021824 utils.py:1231] [6350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.775463903796668
+I1128 09:38:45.851447 137274321021824 utils.py:1231] [6350] core_hours = 11.775463903796668
+I1128 09:38:45.851510 137274321021824 train.py:125] NOTE: Steps:6350/112603 [5.6%]
+Walltime:11h48m (0s eval)
+ETA:8d5h5m
+Total train time:8d16h52m
+I1128 09:43:57.611660 137274321021824 utils.py:1231] [6400] l2_params = 225.18368156462395
+I1128 09:43:57.611938 137274321021824 utils.py:1231] [6400] train/loss = 5.489324331283569
+I1128 09:43:57.612120 137274321021824 utils.py:1231] [6400] l2_grads = 1.2599471807479858
+I1128 09:43:57.612194 137274321021824 utils.py:1231] [6400] lr = 0.0006399
+I1128 09:43:57.612268 137274321021824 utils.py:1231] [6400] uptime = 42826.974630325
+I1128 09:43:57.612328 137274321021824 utils.py:1231] [6400] examples_seen = 6553600.0
+I1128 09:43:57.612381 137274321021824 utils.py:1231] [6400] progress = 0.05683685159365203
+I1128 09:43:57.612433 137274321021824 utils.py:1231] [6400] epoch = 5.115336252026473
+I1128 09:43:57.612484 137274321021824 utils.py:1231] [6400] img/sec/core = 164.22825758229115
+I1128 09:43:57.612552 137274321021824 utils.py:1231] [6400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.862064239400276
+I1128 09:43:57.612603 137274321021824 utils.py:1231] [6400] core_hours = 11.862064239400276
+I1128 09:43:57.612677 137274321021824 train.py:125] NOTE: Steps:6400/112603 [5.7%]
+Walltime:11h53m (0s eval)
+ETA:8d4h54m
+Total train time:8d16h46m
+I1128 09:49:09.378492 137274321021824 utils.py:1231] [6450] l2_params = 225.58270062136546
+I1128 09:49:09.378722 137274321021824 utils.py:1231] [6450] train/loss = 4.693307995796204
+I1128 09:49:09.378814 137274321021824 utils.py:1231] [6450] l2_grads = 1.3428409099578857
+I1128 09:49:09.378895 137274321021824 utils.py:1231] [6450] lr = 0.0006449
+I1128 09:49:09.378949 137274321021824 utils.py:1231] [6450] uptime = 43138.741311198
+I1128 09:49:09.378998 137274321021824 utils.py:1231] [6450] examples_seen = 6604800.0
+I1128 09:49:09.379054 137274321021824 utils.py:1231] [6450] progress = 0.05728088949672744
+I1128 09:49:09.379109 137274321021824 utils.py:1231] [6450] epoch = 5.15529981649543
+I1128 09:49:09.379157 137274321021824 utils.py:1231] [6450] img/sec/core = 164.2253747470108
+I1128 09:49:09.379210 137274321021824 utils.py:1231] [6450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 11.948666095198332
+I1128 09:49:09.379257 137274321021824 utils.py:1231] [6450] core_hours = 11.948666095198332
+I1128 09:49:09.379313 137274321021824 train.py:125] NOTE: Steps:6450/112603 [5.7%]
+Walltime:11h58m (0s eval)
+ETA:8d4h42m
+Total train time:8d16h39m
+I1128 09:54:21.149003 137274321021824 utils.py:1231] [6500] l2_params = 226.05896355985703
+I1128 09:54:21.149252 137274321021824 utils.py:1231] [6500] train/loss = 4.709942936897278
+I1128 09:54:21.149375 137274321021824 utils.py:1231] [6500] l2_grads = 1.330288052558899
+I1128 09:54:21.149455 137274321021824 utils.py:1231] [6500] lr = 0.0006499
+I1128 09:54:21.149513 137274321021824 utils.py:1231] [6500] uptime = 43450.51187494601
+I1128 09:54:21.149564 137274321021824 utils.py:1231] [6500] examples_seen = 6656000.0
+I1128 09:54:21.149613 137274321021824 utils.py:1231] [6500] progress = 0.057724927399802846
+I1128 09:54:21.149660 137274321021824 utils.py:1231] [6500] epoch = 5.195263380964386
+I1128 09:54:21.149709 137274321021824 utils.py:1231] [6500] img/sec/core = 164.22332943973052
+I1128 09:54:21.149762 137274321021824 utils.py:1231] [6500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.03526902957278
+I1128 09:54:21.149809 137274321021824 utils.py:1231] [6500] core_hours = 12.03526902957278
+I1128 09:54:21.149868 137274321021824 train.py:125] NOTE: Steps:6500/112603 [5.8%]
+Walltime:12h4m (0s eval)
+ETA:8d4h31m
+Total train time:8d16h33m
+I1128 09:59:32.926621 137274321021824 utils.py:1231] [6550] l2_params = 226.47629411400638
+I1128 09:59:32.926887 137274321021824 utils.py:1231] [6550] train/loss = 5.05803245306015
+I1128 09:59:32.927031 137274321021824 utils.py:1231] [6550] l2_grads = 1.3094801902770996
+I1128 09:59:32.927121 137274321021824 utils.py:1231] [6550] lr = 0.0006549
+I1128 09:59:32.927199 137274321021824 utils.py:1231] [6550] uptime = 43762.28955681801
+I1128 09:59:32.927260 137274321021824 utils.py:1231] [6550] examples_seen = 6707200.0
+I1128 09:59:32.927323 137274321021824 utils.py:1231] [6550] progress = 0.05816896530287825
+I1128 09:59:32.927393 137274321021824 utils.py:1231] [6550] epoch = 5.235226945433343
+I1128 09:59:32.927456 137274321021824 utils.py:1231] [6550] img/sec/core = 164.21958009495978
+I1128 09:59:32.927515 137274321021824 utils.py:1231] [6550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.12187394120389
+I1128 09:59:32.927568 137274321021824 utils.py:1231] [6550] core_hours = 12.12187394120389
+I1128 09:59:32.927629 137274321021824 train.py:125] NOTE: Steps:6550/112603 [5.8%]
+Walltime:12h9m (0s eval)
+ETA:8d4h19m
+Total train time:8d16h27m
+I1128 10:04:44.872497 137274321021824 utils.py:1231] [6600] l2_params = 226.9200767889123
+I1128 10:04:44.872736 137274321021824 utils.py:1231] [6600] train/loss = 5.143229961395264
+I1128 10:04:44.872887 137274321021824 utils.py:1231] [6600] l2_grads = 1.1252105236053467
+I1128 10:04:44.872984 137274321021824 utils.py:1231] [6600] lr = 0.0006599
+I1128 10:04:44.873045 137274321021824 utils.py:1231] [6600] uptime = 44074.235406141
+I1128 10:04:44.873105 137274321021824 utils.py:1231] [6600] examples_seen = 6758400.0
+I1128 10:04:44.873160 137274321021824 utils.py:1231] [6600] progress = 0.05861300320595366
+I1128 10:04:44.873220 137274321021824 utils.py:1231] [6600] epoch = 5.2751905099023
+I1128 10:04:44.873291 137274321021824 utils.py:1231] [6600] img/sec/core = 164.13105066510195
+I1128 10:04:44.873351 137274321021824 utils.py:1231] [6600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.208525566015831
+I1128 10:04:44.873403 137274321021824 utils.py:1231] [6600] core_hours = 12.208525566015831
+I1128 10:04:44.873485 137274321021824 train.py:125] NOTE: Steps:6600/112603 [5.9%]
+Walltime:12h14m (0s eval)
+ETA:8d4h8m
+Total train time:8d16h21m
+I1128 10:09:56.640020 137274321021824 utils.py:1231] [6650] l2_params = 227.3113340097357
+I1128 10:09:56.640293 137274321021824 utils.py:1231] [6650] train/loss = 4.83147519826889
+I1128 10:09:56.640399 137274321021824 utils.py:1231] [6650] l2_grads = 1.6441189050674438
+I1128 10:09:56.640474 137274321021824 utils.py:1231] [6650] lr = 0.0006649000000000001
+I1128 10:09:56.640533 137274321021824 utils.py:1231] [6650] uptime = 44386.002894506004
+I1128 10:09:56.640597 137274321021824 utils.py:1231] [6650] examples_seen = 6809600.0
+I1128 10:09:56.640654 137274321021824 utils.py:1231] [6650] progress = 0.05905704110902907
+I1128 10:09:56.640713 137274321021824 utils.py:1231] [6650] epoch = 5.315154074371256
+I1128 10:09:56.640773 137274321021824 utils.py:1231] [6650] img/sec/core = 164.2249493958053
+I1128 10:09:56.640837 137274321021824 utils.py:1231] [6650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.295127646117221
+I1128 10:09:56.640899 137274321021824 utils.py:1231] [6650] core_hours = 12.295127646117221
+I1128 10:09:56.640967 137274321021824 train.py:125] NOTE: Steps:6650/112603 [5.9%]
+Walltime:12h19m (0s eval)
+ETA:8d3h57m
+Total train time:8d16h15m
+I1128 10:15:05.734821 137274321021824 utils.py:1231] [6700] l2_params = 227.74719240873856
+I1128 10:15:05.735058 137274321021824 utils.py:1231] [6700] train/loss = 6.246649265289307
+I1128 10:15:05.735206 137274321021824 utils.py:1231] [6700] l2_grads = 1.1860175132751465
+I1128 10:15:05.735278 137274321021824 utils.py:1231] [6700] lr = 0.0006699000000000001
+I1128 10:15:05.735339 137274321021824 utils.py:1231] [6700] uptime = 44695.09770069401
+I1128 10:15:05.735397 137274321021824 utils.py:1231] [6700] examples_seen = 6860800.0
+I1128 10:15:05.735452 137274321021824 utils.py:1231] [6700] progress = 0.059501079012104476
+I1128 10:15:05.735507 137274321021824 utils.py:1231] [6700] epoch = 5.3551176388402135
+I1128 10:15:05.735563 137274321021824 utils.py:1231] [6700] img/sec/core = 165.64497032945295
+I1128 10:15:05.735622 137274321021824 utils.py:1231] [6700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.38098731450278
+I1128 10:15:05.735675 137274321021824 utils.py:1231] [6700] core_hours = 12.38098731450278
+I1128 10:15:05.735740 137274321021824 train.py:125] NOTE: Steps:6700/112603 [6.0%]
+Walltime:12h24m (0s eval)
+ETA:8d3h45m
+Total train time:8d16h8m
+I1128 10:20:17.506603 137274321021824 utils.py:1231] [6750] l2_params = 228.19395629208708
+I1128 10:20:17.506806 137274321021824 utils.py:1231] [6750] train/loss = 4.824877858161926
+I1128 10:20:17.506898 137274321021824 utils.py:1231] [6750] l2_grads = 1.3263790607452393
+I1128 10:20:17.506969 137274321021824 utils.py:1231] [6750] lr = 0.0006749000000000001
+I1128 10:20:17.507019 137274321021824 utils.py:1231] [6750] uptime = 45006.86938160901
+I1128 10:20:17.507070 137274321021824 utils.py:1231] [6750] examples_seen = 6912000.0
+I1128 10:20:17.507117 137274321021824 utils.py:1231] [6750] progress = 0.05994511691517988
+I1128 10:20:17.507161 137274321021824 utils.py:1231] [6750] epoch = 5.395081203309171
+I1128 10:20:17.507210 137274321021824 utils.py:1231] [6750] img/sec/core = 164.22274098063073
+I1128 10:20:17.507262 137274321021824 utils.py:1231] [6750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.46759055920139
+I1128 10:20:17.507317 137274321021824 utils.py:1231] [6750] core_hours = 12.46759055920139
+I1128 10:20:17.507381 137274321021824 train.py:125] NOTE: Steps:6750/112603 [6.0%]
+Walltime:12h30m (0s eval)
+ETA:8d3h34m
+Total train time:8d16h2m
+I1128 10:25:29.273638 137274321021824 utils.py:1231] [6800] l2_params = 228.64588295712653
+I1128 10:25:29.273896 137274321021824 utils.py:1231] [6800] train/loss = 4.8585580587387085
+I1128 10:25:29.274008 137274321021824 utils.py:1231] [6800] l2_grads = 1.355700969696045
+I1128 10:25:29.274083 137274321021824 utils.py:1231] [6800] lr = 0.0006799
+I1128 10:25:29.274143 137274321021824 utils.py:1231] [6800] uptime = 45318.63650430301
+I1128 10:25:29.274209 137274321021824 utils.py:1231] [6800] examples_seen = 6963200.0
+I1128 10:25:29.274268 137274321021824 utils.py:1231] [6800] progress = 0.06038915481825528
+I1128 10:25:29.274321 137274321021824 utils.py:1231] [6800] epoch = 5.435044767778128
+I1128 10:25:29.274374 137274321021824 utils.py:1231] [6800] img/sec/core = 164.22514201490344
+I1128 10:25:29.274435 137274321021824 utils.py:1231] [6800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.5541925377275
+I1128 10:25:29.274485 137274321021824 utils.py:1231] [6800] core_hours = 12.5541925377275
+I1128 10:25:29.274558 137274321021824 train.py:125] NOTE: Steps:6800/112603 [6.0%]
+Walltime:12h35m (0s eval)
+ETA:8d3h23m
+Total train time:8d15h56m
+I1128 10:30:41.041980 137274321021824 utils.py:1231] [6850] l2_params = 229.0384933606064
+I1128 10:30:41.042281 137274321021824 utils.py:1231] [6850] train/loss = 5.029408931732178
+I1128 10:30:41.042506 137274321021824 utils.py:1231] [6850] l2_grads = 1.600185513496399
+I1128 10:30:41.042614 137274321021824 utils.py:1231] [6850] lr = 0.0006849
+I1128 10:30:41.042727 137274321021824 utils.py:1231] [6850] uptime = 45630.405085757
+I1128 10:30:41.042806 137274321021824 utils.py:1231] [6850] examples_seen = 7014400.0
+I1128 10:30:41.042892 137274321021824 utils.py:1231] [6850] progress = 0.06083319272133069
+I1128 10:30:41.042972 137274321021824 utils.py:1231] [6850] epoch = 5.475008332247084
+I1128 10:30:41.043046 137274321021824 utils.py:1231] [6850] img/sec/core = 164.22437360820595
+I1128 10:30:41.043144 137274321021824 utils.py:1231] [6850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.640794921464721
+I1128 10:30:41.043229 137274321021824 utils.py:1231] [6850] core_hours = 12.640794921464721
+I1128 10:30:41.043328 137274321021824 train.py:125] NOTE: Steps:6850/112603 [6.1%]
+Walltime:12h40m (0s eval)
+ETA:8d3h12m
+Total train time:8d15h51m
+I1128 10:35:52.809722 137274321021824 utils.py:1231] [6900] l2_params = 229.45129362715065
+I1128 10:35:52.809960 137274321021824 utils.py:1231] [6900] train/loss = 4.740576863288879
+I1128 10:35:52.810067 137274321021824 utils.py:1231] [6900] l2_grads = 1.5763335227966309
+I1128 10:35:52.810129 137274321021824 utils.py:1231] [6900] lr = 0.0006899
+I1128 10:35:52.810183 137274321021824 utils.py:1231] [6900] uptime = 45942.172544983
+I1128 10:35:52.810236 137274321021824 utils.py:1231] [6900] examples_seen = 7065600.0
+I1128 10:35:52.810286 137274321021824 utils.py:1231] [6900] progress = 0.0612772306244061
+I1128 10:35:52.810334 137274321021824 utils.py:1231] [6900] epoch = 5.514971896716041
+I1128 10:35:52.810388 137274321021824 utils.py:1231] [6900] img/sec/core = 164.22496474490944
+I1128 10:35:52.810454 137274321021824 utils.py:1231] [6900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.727396993471944
+I1128 10:35:52.810509 137274321021824 utils.py:1231] [6900] core_hours = 12.727396993471944
+I1128 10:35:52.810572 137274321021824 train.py:125] NOTE: Steps:6900/112603 [6.1%]
+Walltime:12h45m (0s eval)
+ETA:8d3h1m
+Total train time:8d15h45m
+I1128 10:41:04.581219 137274321021824 utils.py:1231] [6950] l2_params = 229.87714943985634
+I1128 10:41:04.581434 137274321021824 utils.py:1231] [6950] train/loss = 5.762225449085236
+I1128 10:41:04.581541 137274321021824 utils.py:1231] [6950] l2_grads = 1.0829293727874756
+I1128 10:41:04.581608 137274321021824 utils.py:1231] [6950] lr = 0.0006948999999999999
+I1128 10:41:04.581667 137274321021824 utils.py:1231] [6950] uptime = 46253.944028704005
+I1128 10:41:04.581728 137274321021824 utils.py:1231] [6950] examples_seen = 7116800.0
+I1128 10:41:04.581782 137274321021824 utils.py:1231] [6950] progress = 0.061721268527481506
+I1128 10:41:04.581835 137274321021824 utils.py:1231] [6950] epoch = 5.554935461184998
+I1128 10:41:04.581894 137274321021824 utils.py:1231] [6950] img/sec/core = 164.22284485073973
+I1128 10:41:04.581975 137274321021824 utils.py:1231] [6950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.814000183394446
+I1128 10:41:04.582023 137274321021824 utils.py:1231] [6950] core_hours = 12.814000183394446
+I1128 10:41:04.582086 137274321021824 train.py:125] NOTE: Steps:6950/112603 [6.2%]
+Walltime:12h50m (0s eval)
+ETA:8d2h51m
+Total train time:8d15h40m
+I1128 10:46:16.523886 137274321021824 utils.py:1231] [7000] l2_params = 230.35046219359995
+I1128 10:46:16.524108 137274321021824 utils.py:1231] [7000] train/loss = 5.325067579746246
+I1128 10:46:16.524207 137274321021824 utils.py:1231] [7000] l2_grads = 1.0705318450927734
+I1128 10:46:16.524299 137274321021824 utils.py:1231] [7000] lr = 0.0006998999999999999
+I1128 10:46:16.524389 137274321021824 utils.py:1231] [7000] uptime = 46565.886745632
+I1128 10:46:16.524462 137274321021824 utils.py:1231] [7000] examples_seen = 7168000.0
+I1128 10:46:16.524535 137274321021824 utils.py:1231] [7000] progress = 0.06216530643055691
+I1128 10:46:16.524598 137274321021824 utils.py:1231] [7000] epoch = 5.594899025653954
+I1128 10:46:16.524665 137274321021824 utils.py:1231] [7000] img/sec/core = 164.1326987987296
+I1128 10:46:16.524757 137274321021824 utils.py:1231] [7000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.900650938096666
+I1128 10:46:16.524824 137274321021824 utils.py:1231] [7000] core_hours = 12.900650938096666
+I1128 10:46:16.524903 137274321021824 train.py:125] NOTE: Steps:7000/112603 [6.2%]
+Walltime:12h56m (0s eval)
+ETA:8d2h40m
+Total train time:8d15h34m
+I1128 10:51:28.638025 137274321021824 utils.py:1231] [7050] l2_params = 230.822408363866
+I1128 10:51:28.638291 137274321021824 utils.py:1231] [7050] train/loss = 6.20952695608139
+I1128 10:51:28.638409 137274321021824 utils.py:1231] [7050] l2_grads = 0.859561562538147
+I1128 10:51:28.638481 137274321021824 utils.py:1231] [7050] lr = 0.0007049
+I1128 10:51:28.638537 137274321021824 utils.py:1231] [7050] uptime = 46878.000896682
+I1128 10:51:28.638601 137274321021824 utils.py:1231] [7050] examples_seen = 7219200.0
+I1128 10:51:28.638655 137274321021824 utils.py:1231] [7050] progress = 0.06260934433363231
+I1128 10:51:28.638706 137274321021824 utils.py:1231] [7050] epoch = 5.6348625901229115
+I1128 10:51:28.638756 137274321021824 utils.py:1231] [7050] img/sec/core = 164.04254606128882
+I1128 10:51:28.638821 137274321021824 utils.py:1231] [7050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 12.987349313388334
+I1128 10:51:28.638889 137274321021824 utils.py:1231] [7050] core_hours = 12.987349313388334
+I1128 10:51:28.638957 137274321021824 train.py:125] NOTE: Steps:7050/112603 [6.3%]
+Walltime:13h1m (0s eval)
+ETA:8d2h30m
+Total train time:8d15h29m
+I1128 10:56:40.383102 137274321021824 utils.py:1231] [7100] l2_params = 231.20745844465515
+I1128 10:56:40.383298 137274321021824 utils.py:1231] [7100] train/loss = 4.597099483013153
+I1128 10:56:40.383385 137274321021824 utils.py:1231] [7100] l2_grads = 1.4155476093292236
+I1128 10:56:40.383456 137274321021824 utils.py:1231] [7100] lr = 0.0007099
+I1128 10:56:40.383504 137274321021824 utils.py:1231] [7100] uptime = 47189.74586675
+I1128 10:56:40.383551 137274321021824 utils.py:1231] [7100] examples_seen = 7270400.0
+I1128 10:56:40.383596 137274321021824 utils.py:1231] [7100] progress = 0.06305338223670773
+I1128 10:56:40.383640 137274321021824 utils.py:1231] [7100] epoch = 5.674826154591869
+I1128 10:56:40.383686 137274321021824 utils.py:1231] [7100] img/sec/core = 164.2368118684698
+I1128 10:56:40.383736 137274321021824 utils.py:1231] [7100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.073945138407222
+I1128 10:56:40.383781 137274321021824 utils.py:1231] [7100] core_hours = 13.073945138407222
+I1128 10:56:40.383835 137274321021824 train.py:125] NOTE: Steps:7100/112603 [6.3%]
+Walltime:13h6m (0s eval)
+ETA:8d2h19m
+Total train time:8d15h24m
+I1128 11:01:52.135619 137274321021824 utils.py:1231] [7150] l2_params = 231.6857010801696
+I1128 11:01:52.135875 137274321021824 utils.py:1231] [7150] train/loss = 4.454483211040497
+I1128 11:01:52.135978 137274321021824 utils.py:1231] [7150] l2_grads = 1.5560232400894165
+I1128 11:01:52.136038 137274321021824 utils.py:1231] [7150] lr = 0.0007149
+I1128 11:01:52.136088 137274321021824 utils.py:1231] [7150] uptime = 47501.498450319006
+I1128 11:01:52.136139 137274321021824 utils.py:1231] [7150] examples_seen = 7321600.0
+I1128 11:01:52.136187 137274321021824 utils.py:1231] [7150] progress = 0.06349742013978313
+I1128 11:01:52.136234 137274321021824 utils.py:1231] [7150] epoch = 5.714789719060825
+I1128 11:01:52.136284 137274321021824 utils.py:1231] [7150] img/sec/core = 164.23280094058126
+I1128 11:01:52.136338 137274321021824 utils.py:1231] [7150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.160543078287501
+I1128 11:01:52.136386 137274321021824 utils.py:1231] [7150] core_hours = 13.160543078287501
+I1128 11:01:52.136443 137274321021824 train.py:125] NOTE: Steps:7150/112603 [6.3%]
+Walltime:13h11m (0s eval)
+ETA:8d2h9m
+Total train time:8d15h19m
+I1128 11:07:03.900625 137274321021824 utils.py:1231] [7200] l2_params = 232.15793164630136
+I1128 11:07:03.900900 137274321021824 utils.py:1231] [7200] train/loss = 4.585776627063751
+I1128 11:07:03.901046 137274321021824 utils.py:1231] [7200] l2_grads = 1.8999218940734863
+I1128 11:07:03.901140 137274321021824 utils.py:1231] [7200] lr = 0.0007199
+I1128 11:07:03.901216 137274321021824 utils.py:1231] [7200] uptime = 47813.263577936
+I1128 11:07:03.901286 137274321021824 utils.py:1231] [7200] examples_seen = 7372800.0
+I1128 11:07:03.901364 137274321021824 utils.py:1231] [7200] progress = 0.06394145804285854
+I1128 11:07:03.901419 137274321021824 utils.py:1231] [7200] epoch = 5.754753283529782
+I1128 11:07:03.901486 137274321021824 utils.py:1231] [7200] img/sec/core = 164.22619294002578
+I1128 11:07:03.901556 137274321021824 utils.py:1231] [7200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.247144502625554
+I1128 11:07:03.901621 137274321021824 utils.py:1231] [7200] core_hours = 13.247144502625554
+I1128 11:07:03.901690 137274321021824 train.py:125] NOTE: Steps:7200/112603 [6.4%]
+Walltime:13h16m (0s eval)
+ETA:8d1h58m
+Total train time:8d15h14m
+I1128 11:12:15.712030 137274321021824 utils.py:1231] [7250] l2_params = 232.55626883588909
+I1128 11:12:15.712301 137274321021824 utils.py:1231] [7250] train/loss = 5.666185021400452
+I1128 11:12:15.712505 137274321021824 utils.py:1231] [7250] l2_grads = 1.0682841539382935
+I1128 11:12:15.712600 137274321021824 utils.py:1231] [7250] lr = 0.0007249
+I1128 11:12:15.712667 137274321021824 utils.py:1231] [7250] uptime = 48125.075025014
+I1128 11:12:15.712730 137274321021824 utils.py:1231] [7250] examples_seen = 7424000.0
+I1128 11:12:15.712803 137274321021824 utils.py:1231] [7250] progress = 0.06438549594593394
+I1128 11:12:15.712860 137274321021824 utils.py:1231] [7250] epoch = 5.794716847998739
+I1128 11:12:15.712943 137274321021824 utils.py:1231] [7250] img/sec/core = 164.20179720724792
+I1128 11:12:15.713023 137274321021824 utils.py:1231] [7250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.333758793480554
+I1128 11:12:15.713117 137274321021824 utils.py:1231] [7250] core_hours = 13.333758793480554
+I1128 11:12:15.713212 137274321021824 train.py:125] NOTE: Steps:7250/112603 [6.4%]
+Walltime:13h22m (0s eval)
+ETA:8d1h48m
+Total train time:8d15h8m
+I1128 11:17:27.578595 137274321021824 utils.py:1231] [7300] l2_params = 233.06617912915013
+I1128 11:17:27.578826 137274321021824 utils.py:1231] [7300] train/loss = 5.082215845584869
+I1128 11:17:27.578932 137274321021824 utils.py:1231] [7300] l2_grads = 1.640141248703003
+I1128 11:17:27.578994 137274321021824 utils.py:1231] [7300] lr = 0.0007299
+I1128 11:17:27.579047 137274321021824 utils.py:1231] [7300] uptime = 48436.941408937
+I1128 11:17:27.579099 137274321021824 utils.py:1231] [7300] examples_seen = 7475200.0
+I1128 11:17:27.579154 137274321021824 utils.py:1231] [7300] progress = 0.06482953384900936
+I1128 11:17:27.579201 137274321021824 utils.py:1231] [7300] epoch = 5.834680412467695
+I1128 11:17:27.579251 137274321021824 utils.py:1231] [7300] img/sec/core = 164.1728722280033
+I1128 11:17:27.579305 137274321021824 utils.py:1231] [7300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.420388344570277
+I1128 11:17:27.579354 137274321021824 utils.py:1231] [7300] core_hours = 13.420388344570277
+I1128 11:17:27.579424 137274321021824 train.py:125] NOTE: Steps:7300/112603 [6.5%]
+Walltime:13h27m (0s eval)
+ETA:8d1h38m
+Total train time:8d15h4m
+I1128 11:22:39.415706 137274321021824 utils.py:1231] [7350] l2_params = 233.55572297024145
+I1128 11:22:39.415992 137274321021824 utils.py:1231] [7350] train/loss = 5.6322057247161865
+I1128 11:22:39.416112 137274321021824 utils.py:1231] [7350] l2_grads = 1.1427346467971802
+I1128 11:22:39.416193 137274321021824 utils.py:1231] [7350] lr = 0.0007349
+I1128 11:22:39.416256 137274321021824 utils.py:1231] [7350] uptime = 48748.778617728996
+I1128 11:22:39.416318 137274321021824 utils.py:1231] [7350] examples_seen = 7526400.0
+I1128 11:22:39.416375 137274321021824 utils.py:1231] [7350] progress = 0.06527357175208476
+I1128 11:22:39.416427 137274321021824 utils.py:1231] [7350] epoch = 5.874643976936652
+I1128 11:22:39.416478 137274321021824 utils.py:1231] [7350] img/sec/core = 164.18823205331856
+I1128 11:22:39.416536 137274321021824 utils.py:1231] [7350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.507009791456943
+I1128 11:22:39.416590 137274321021824 utils.py:1231] [7350] core_hours = 13.507009791456943
+I1128 11:22:39.416652 137274321021824 train.py:125] NOTE: Steps:7350/112603 [6.5%]
+Walltime:13h32m (0s eval)
+ETA:8d1h28m
+Total train time:8d14h59m
+I1128 11:27:51.235516 137274321021824 utils.py:1231] [7400] l2_params = 233.99027801873646
+I1128 11:27:51.235762 137274321021824 utils.py:1231] [7400] train/loss = 5.458814024925232
+I1128 11:27:51.235928 137274321021824 utils.py:1231] [7400] l2_grads = 1.0812976360321045
+I1128 11:27:51.236021 137274321021824 utils.py:1231] [7400] lr = 0.0007399
+I1128 11:27:51.236078 137274321021824 utils.py:1231] [7400] uptime = 49060.598439733
+I1128 11:27:51.236130 137274321021824 utils.py:1231] [7400] examples_seen = 7577600.0
+I1128 11:27:51.236178 137274321021824 utils.py:1231] [7400] progress = 0.06571760965516016
+I1128 11:27:51.236227 137274321021824 utils.py:1231] [7400] epoch = 5.914607541405609
+I1128 11:27:51.236276 137274321021824 utils.py:1231] [7400] img/sec/core = 164.1973870389231
+I1128 11:27:51.236331 137274321021824 utils.py:1231] [7400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.593626408680278
+I1128 11:27:51.236397 137274321021824 utils.py:1231] [7400] core_hours = 13.593626408680278
+I1128 11:27:51.236457 137274321021824 train.py:125] NOTE: Steps:7400/112603 [6.6%]
+Walltime:13h37m (0s eval)
+ETA:8d1h18m
+Total train time:8d14h54m
+I1128 11:33:03.046079 137274321021824 utils.py:1231] [7450] l2_params = 234.56826662991418
+I1128 11:33:03.046349 137274321021824 utils.py:1231] [7450] train/loss = 4.818124175071716
+I1128 11:33:03.046485 137274321021824 utils.py:1231] [7450] l2_grads = 1.301426887512207
+I1128 11:33:03.046575 137274321021824 utils.py:1231] [7450] lr = 0.0007449000000000001
+I1128 11:33:03.046641 137274321021824 utils.py:1231] [7450] uptime = 49372.409002607004
+I1128 11:33:03.046709 137274321021824 utils.py:1231] [7450] examples_seen = 7628800.0
+I1128 11:33:03.046772 137274321021824 utils.py:1231] [7450] progress = 0.06616164755823557
+I1128 11:33:03.046856 137274321021824 utils.py:1231] [7450] epoch = 5.954571105874566
+I1128 11:33:03.046923 137274321021824 utils.py:1231] [7450] img/sec/core = 164.20226283575028
+I1128 11:33:03.046987 137274321021824 utils.py:1231] [7450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.680240453923057
+I1128 11:33:03.047042 137274321021824 utils.py:1231] [7450] core_hours = 13.680240453923057
+I1128 11:33:03.047103 137274321021824 train.py:125] NOTE: Steps:7450/112603 [6.6%]
+Walltime:13h42m (0s eval)
+ETA:8d1h8m
+Total train time:8d14h49m
+I1128 11:38:14.881986 137274321021824 utils.py:1231] [7500] l2_params = 235.12165418337756
+I1128 11:38:14.882197 137274321021824 utils.py:1231] [7500] train/loss = 5.67201954126358
+I1128 11:38:14.882302 137274321021824 utils.py:1231] [7500] l2_grads = 1.4535528421401978
+I1128 11:38:14.882366 137274321021824 utils.py:1231] [7500] lr = 0.0007499000000000001
+I1128 11:38:14.882433 137274321021824 utils.py:1231] [7500] uptime = 49684.24479554601
+I1128 11:38:14.882512 137274321021824 utils.py:1231] [7500] examples_seen = 7680000.0
+I1128 11:38:14.882571 137274321021824 utils.py:1231] [7500] progress = 0.06660568546131097
+I1128 11:38:14.882627 137274321021824 utils.py:1231] [7500] epoch = 5.994534670343523
+I1128 11:38:14.882683 137274321021824 utils.py:1231] [7500] img/sec/core = 164.18897753028264
+I1128 11:38:14.882756 137274321021824 utils.py:1231] [7500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.766861507517225
+I1128 11:38:14.882811 137274321021824 utils.py:1231] [7500] core_hours = 13.766861507517225
+I1128 11:38:14.882874 137274321021824 train.py:125] NOTE: Steps:7500/112603 [6.7%]
+Walltime:13h48m (0s eval)
+ETA:8d0h58m
+Total train time:8d14h44m
+I1128 11:38:14.882992 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:7500/112603 [6.7%]
+Walltime:13h48m (0s eval)
+ETA:8d0h58m
+Total train time:8d14h44m
+I1128 11:39:53.478972 137274321021824 utils.py:1231] [7500] val/acc@1 = 0.2616390306122449
+I1128 11:39:53.487650 137274321021824 utils.py:1231] [7500] val/loss = 3.62663003376552
+I1128 11:39:53.487823 137274321021824 utils.py:1231] [7500] z/secs/eval/val = 98.60475225899427
+I1128 11:39:53.487895 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.60475225899427
+I1128 11:45:05.274846 137274321021824 utils.py:1231] [7550] l2_params = 235.57554548607476
+I1128 11:45:05.275069 137274321021824 utils.py:1231] [7550] train/loss = 5.738597810268402
+I1128 11:45:05.275174 137274321021824 utils.py:1231] [7550] l2_grads = 1.1784006357192993
+I1128 11:45:05.275244 137274321021824 utils.py:1231] [7550] lr = 0.0007549000000000001
+I1128 11:45:05.275306 137274321021824 utils.py:1231] [7550] uptime = 50094.637667854
+I1128 11:45:05.275364 137274321021824 utils.py:1231] [7550] examples_seen = 7731200.0
+I1128 11:45:05.275419 137274321021824 utils.py:1231] [7550] progress = 0.06704972336438639
+I1128 11:45:05.275475 137274321021824 utils.py:1231] [7550] epoch = 6.03449823481248
+I1128 11:45:05.275530 137274321021824 utils.py:1231] [7550] img/sec/core = 124.7585020472185
+I1128 11:45:05.275595 137274321021824 utils.py:1231] [7550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.880859527602777
+I1128 11:45:05.275650 137274321021824 utils.py:1231] [7550] core_hours = 13.880859527602777
+I1128 11:45:05.275715 137274321021824 train.py:125] NOTE: Steps:7550/112603 [6.7%]
+Walltime:13h54m (0s eval)
+ETA:8d1h11m
+Total train time:8d15h4m
+I1128 11:50:17.061783 137274321021824 utils.py:1231] [7600] l2_params = 236.1228909558596
+I1128 11:50:17.062022 137274321021824 utils.py:1231] [7600] train/loss = 4.478292882442474
+I1128 11:50:17.062153 137274321021824 utils.py:1231] [7600] l2_grads = 1.4045454263687134
+I1128 11:50:17.062232 137274321021824 utils.py:1231] [7600] lr = 0.0007599
+I1128 11:50:17.062299 137274321021824 utils.py:1231] [7600] uptime = 50406.424660477
+I1128 11:50:17.062369 137274321021824 utils.py:1231] [7600] examples_seen = 7782400.0
+I1128 11:50:17.062424 137274321021824 utils.py:1231] [7600] progress = 0.06749376126746179
+I1128 11:50:17.062479 137274321021824 utils.py:1231] [7600] epoch = 6.074461799281436
+I1128 11:50:17.062534 137274321021824 utils.py:1231] [7600] img/sec/core = 164.21467608146213
+I1128 11:50:17.062603 137274321021824 utils.py:1231] [7600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 13.967467025553612
+I1128 11:50:17.062657 137274321021824 utils.py:1231] [7600] core_hours = 13.967467025553612
+I1128 11:50:17.062732 137274321021824 train.py:125] NOTE: Steps:7600/112603 [6.7%]
+Walltime:14h0m (0s eval)
+ETA:8d1h1m
+Total train time:8d14h59m
+I1128 11:55:28.844752 137274321021824 utils.py:1231] [7650] l2_params = 236.69055269808055
+I1128 11:55:28.845011 137274321021824 utils.py:1231] [7650] train/loss = 4.594302952289581
+I1128 11:55:28.845121 137274321021824 utils.py:1231] [7650] l2_grads = 1.3482311964035034
+I1128 11:55:28.845179 137274321021824 utils.py:1231] [7650] lr = 0.0007649
+I1128 11:55:28.845249 137274321021824 utils.py:1231] [7650] uptime = 50718.20760688701
+I1128 11:55:28.845297 137274321021824 utils.py:1231] [7650] examples_seen = 7833600.0
+I1128 11:55:28.845343 137274321021824 utils.py:1231] [7650] progress = 0.0679377991705372
+I1128 11:55:28.845387 137274321021824 utils.py:1231] [7650] epoch = 6.114425363750393
+I1128 11:55:28.845432 137274321021824 utils.py:1231] [7650] img/sec/core = 164.21680720365632
+I1128 11:55:28.845484 137274321021824 utils.py:1231] [7650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.05407339955639
+I1128 11:55:28.845530 137274321021824 utils.py:1231] [7650] core_hours = 14.05407339955639
+I1128 11:55:28.845587 137274321021824 train.py:125] NOTE: Steps:7650/112603 [6.8%]
+Walltime:14h5m (0s eval)
+ETA:8d0h51m
+Total train time:8d14h55m
+I1128 12:00:40.626605 137274321021824 utils.py:1231] [7700] l2_params = 237.18744100051958
+I1128 12:00:40.626831 137274321021824 utils.py:1231] [7700] train/loss = 5.043022036552429
+I1128 12:00:40.626932 137274321021824 utils.py:1231] [7700] l2_grads = 1.24905264377594
+I1128 12:00:40.626993 137274321021824 utils.py:1231] [7700] lr = 0.0007699
+I1128 12:00:40.627044 137274321021824 utils.py:1231] [7700] uptime = 51029.98940619601
+I1128 12:00:40.627096 137274321021824 utils.py:1231] [7700] examples_seen = 7884800.0
+I1128 12:00:40.627144 137274321021824 utils.py:1231] [7700] progress = 0.0683818370736126
+I1128 12:00:40.627191 137274321021824 utils.py:1231] [7700] epoch = 6.15438892821935
+I1128 12:00:40.627239 137274321021824 utils.py:1231] [7700] img/sec/core = 164.21741138666317
+I1128 12:00:40.627294 137274321021824 utils.py:1231] [7700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.14067945492
+I1128 12:00:40.627347 137274321021824 utils.py:1231] [7700] core_hours = 14.14067945492
+I1128 12:00:40.627407 137274321021824 train.py:125] NOTE: Steps:7700/112603 [6.8%]
+Walltime:14h10m (0s eval)
+ETA:8d0h41m
+Total train time:8d14h50m
+I1128 12:05:52.419819 137274321021824 utils.py:1231] [7750] l2_params = 237.6261135772045
+I1128 12:05:52.420011 137274321021824 utils.py:1231] [7750] train/loss = 4.781372368335724
+I1128 12:05:52.420103 137274321021824 utils.py:1231] [7750] l2_grads = 1.324341893196106
+I1128 12:05:52.420171 137274321021824 utils.py:1231] [7750] lr = 0.0007749
+I1128 12:05:52.420267 137274321021824 utils.py:1231] [7750] uptime = 51341.782609787
+I1128 12:05:52.420328 137274321021824 utils.py:1231] [7750] examples_seen = 7936000.0
+I1128 12:05:52.420389 137274321021824 utils.py:1231] [7750] progress = 0.068825874976688
+I1128 12:05:52.420447 137274321021824 utils.py:1231] [7750] epoch = 6.1943524926883065
+I1128 12:05:52.420514 137274321021824 utils.py:1231] [7750] img/sec/core = 164.21140490016577
+I1128 12:05:52.420571 137274321021824 utils.py:1231] [7750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.227288678139722
+I1128 12:05:52.420625 137274321021824 utils.py:1231] [7750] core_hours = 14.227288678139722
+I1128 12:05:52.420694 137274321021824 train.py:125] NOTE: Steps:7750/112603 [6.9%]
+Walltime:14h15m (0s eval)
+ETA:8d0h32m
+Total train time:8d14h46m
+I1128 12:11:04.227737 137274321021824 utils.py:1231] [7800] l2_params = 238.12992759844494
+I1128 12:11:04.228030 137274321021824 utils.py:1231] [7800] train/loss = 6.316031575202942
+I1128 12:11:04.228225 137274321021824 utils.py:1231] [7800] l2_grads = 1.117388129234314
+I1128 12:11:04.228299 137274321021824 utils.py:1231] [7800] lr = 0.0007799
+I1128 12:11:04.228363 137274321021824 utils.py:1231] [7800] uptime = 51653.590723891
+I1128 12:11:04.228424 137274321021824 utils.py:1231] [7800] examples_seen = 7987200.0
+I1128 12:11:04.228484 137274321021824 utils.py:1231] [7800] progress = 0.06926991287976342
+I1128 12:11:04.228557 137274321021824 utils.py:1231] [7800] epoch = 6.234316057157264
+I1128 12:11:04.228619 137274321021824 utils.py:1231] [7800] img/sec/core = 164.2035523903092
+I1128 12:11:04.228678 137274321021824 utils.py:1231] [7800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.31390204316861
+I1128 12:11:04.228734 137274321021824 utils.py:1231] [7800] core_hours = 14.31390204316861
+I1128 12:11:04.228801 137274321021824 train.py:125] NOTE: Steps:7800/112603 [6.9%]
+Walltime:14h20m (0s eval)
+ETA:8d0h22m
+Total train time:8d14h41m
+I1128 12:16:16.072135 137274321021824 utils.py:1231] [7850] l2_params = 238.7274904178916
+I1128 12:16:16.072458 137274321021824 utils.py:1231] [7850] train/loss = 4.422773540019989
+I1128 12:16:16.072657 137274321021824 utils.py:1231] [7850] l2_grads = 1.528489589691162
+I1128 12:16:16.072744 137274321021824 utils.py:1231] [7850] lr = 0.0007849
+I1128 12:16:16.072826 137274321021824 utils.py:1231] [7850] uptime = 51965.435179545006
+I1128 12:16:16.072911 137274321021824 utils.py:1231] [7850] examples_seen = 8038400.0
+I1128 12:16:16.072971 137274321021824 utils.py:1231] [7850] progress = 0.06971395078283882
+I1128 12:16:16.073027 137274321021824 utils.py:1231] [7850] epoch = 6.274279621626221
+I1128 12:16:16.073085 137274321021824 utils.py:1231] [7850] img/sec/core = 164.18441653106362
+I1128 12:16:16.073142 137274321021824 utils.py:1231] [7850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.4005255030725
+I1128 12:16:16.073199 137274321021824 utils.py:1231] [7850] core_hours = 14.4005255030725
+I1128 12:16:16.073267 137274321021824 train.py:125] NOTE: Steps:7850/112603 [7.0%]
+Walltime:14h26m (0s eval)
+ETA:8d0h12m
+Total train time:8d14h37m
+I1128 12:21:27.862543 137274321021824 utils.py:1231] [7900] l2_params = 239.22860870968387
+I1128 12:21:27.862754 137274321021824 utils.py:1231] [7900] train/loss = 5.553708791732788
+I1128 12:21:27.862861 137274321021824 utils.py:1231] [7900] l2_grads = 1.22882080078125
+I1128 12:21:27.862931 137274321021824 utils.py:1231] [7900] lr = 0.0007899000000000001
+I1128 12:21:27.862985 137274321021824 utils.py:1231] [7900] uptime = 52277.22534677801
+I1128 12:21:27.863038 137274321021824 utils.py:1231] [7900] examples_seen = 8089600.0
+I1128 12:21:27.863089 137274321021824 utils.py:1231] [7900] progress = 0.07015798868591423
+I1128 12:21:27.863138 137274321021824 utils.py:1231] [7900] epoch = 6.314243186095177
+I1128 12:21:27.863196 137274321021824 utils.py:1231] [7900] img/sec/core = 164.21300406737214
+I1128 12:21:27.863255 137274321021824 utils.py:1231] [7900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.487133882859446
+I1128 12:21:27.863309 137274321021824 utils.py:1231] [7900] core_hours = 14.487133882859446
+I1128 12:21:27.863370 137274321021824 train.py:125] NOTE: Steps:7900/112603 [7.0%]
+Walltime:14h31m (0s eval)
+ETA:8d0h3m
+Total train time:8d14h32m
+I1128 12:26:39.655915 137274321021824 utils.py:1231] [7950] l2_params = 239.72812492650243
+I1128 12:26:39.656131 137274321021824 utils.py:1231] [7950] train/loss = 4.62199068069458
+I1128 12:26:39.656233 137274321021824 utils.py:1231] [7950] l2_grads = 1.4633451700210571
+I1128 12:26:39.656293 137274321021824 utils.py:1231] [7950] lr = 0.0007949000000000001
+I1128 12:26:39.656354 137274321021824 utils.py:1231] [7950] uptime = 52589.01871222601
+I1128 12:26:39.656427 137274321021824 utils.py:1231] [7950] examples_seen = 8140800.0
+I1128 12:26:39.656484 137274321021824 utils.py:1231] [7950] progress = 0.07060202658898963
+I1128 12:26:39.656532 137274321021824 utils.py:1231] [7950] epoch = 6.354206750564134
+I1128 12:26:39.656582 137274321021824 utils.py:1231] [7950] img/sec/core = 164.21131965535344
+I1128 12:26:39.656636 137274321021824 utils.py:1231] [7950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.573743151039446
+I1128 12:26:39.656684 137274321021824 utils.py:1231] [7950] core_hours = 14.573743151039446
+I1128 12:26:39.656743 137274321021824 train.py:125] NOTE: Steps:7950/112603 [7.1%]
+Walltime:14h36m (0s eval)
+ETA:7d23h53m
+Total train time:8d14h28m
+I1128 12:31:51.407239 137274321021824 utils.py:1231] [8000] l2_params = 240.32511142547995
+I1128 12:31:51.407460 137274321021824 utils.py:1231] [8000] train/loss = 4.4282843470573425
+I1128 12:31:51.407574 137274321021824 utils.py:1231] [8000] l2_grads = 1.3582907915115356
+I1128 12:31:51.407645 137274321021824 utils.py:1231] [8000] lr = 0.0007999000000000001
+I1128 12:31:51.407698 137274321021824 utils.py:1231] [8000] uptime = 52900.770061142
+I1128 12:31:51.407768 137274321021824 utils.py:1231] [8000] examples_seen = 8192000.0
+I1128 12:31:51.407813 137274321021824 utils.py:1231] [8000] progress = 0.07104606449206505
+I1128 12:31:51.407856 137274321021824 utils.py:1231] [8000] epoch = 6.394170315033091
+I1128 12:31:51.407904 137274321021824 utils.py:1231] [8000] img/sec/core = 164.233451364463
+I1128 12:31:51.407955 137274321021824 utils.py:1231] [8000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.660340747960555
+I1128 12:31:51.407998 137274321021824 utils.py:1231] [8000] core_hours = 14.660340747960555
+I1128 12:31:51.408050 137274321021824 train.py:125] NOTE: Steps:8000/112603 [7.1%]
+Walltime:14h41m (0s eval)
+ETA:7d23h44m
+Total train time:8d14h24m
+I1128 12:37:03.540453 137274321021824 utils.py:1231] [8050] l2_params = 240.89617052907488
+I1128 12:37:03.540699 137274321021824 utils.py:1231] [8050] train/loss = 4.519928276538849
+I1128 12:37:03.540836 137274321021824 utils.py:1231] [8050] l2_grads = 1.2585433721542358
+I1128 12:37:03.540917 137274321021824 utils.py:1231] [8050] lr = 0.0008049
+I1128 12:37:03.540979 137274321021824 utils.py:1231] [8050] uptime = 53212.903340453006
+I1128 12:37:03.541039 137274321021824 utils.py:1231] [8050] examples_seen = 8243200.0
+I1128 12:37:03.541095 137274321021824 utils.py:1231] [8050] progress = 0.07149010239514045
+I1128 12:37:03.541151 137274321021824 utils.py:1231] [8050] epoch = 6.434133879502048
+I1128 12:37:03.541207 137274321021824 utils.py:1231] [8050] img/sec/core = 164.03249314849594
+I1128 12:37:03.541269 137274321021824 utils.py:1231] [8050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.747044436658056
+I1128 12:37:03.541325 137274321021824 utils.py:1231] [8050] core_hours = 14.747044436658056
+I1128 12:37:03.541391 137274321021824 train.py:125] NOTE: Steps:8050/112603 [7.1%]
+Walltime:14h46m (0s eval)
+ETA:7d23h34m
+Total train time:8d14h19m
+I1128 12:42:15.322513 137274321021824 utils.py:1231] [8100] l2_params = 241.43542456196198
+I1128 12:42:15.322728 137274321021824 utils.py:1231] [8100] train/loss = 4.459672749042511
+I1128 12:42:15.322828 137274321021824 utils.py:1231] [8100] l2_grads = 1.4346057176589966
+I1128 12:42:15.322902 137274321021824 utils.py:1231] [8100] lr = 0.0008099
+I1128 12:42:15.322970 137274321021824 utils.py:1231] [8100] uptime = 53524.685331315006
+I1128 12:42:15.323040 137274321021824 utils.py:1231] [8100] examples_seen = 8294400.0
+I1128 12:42:15.323095 137274321021824 utils.py:1231] [8100] progress = 0.07193414029821586
+I1128 12:42:15.323148 137274321021824 utils.py:1231] [8100] epoch = 6.474097443971004
+I1128 12:42:15.323203 137274321021824 utils.py:1231] [8100] img/sec/core = 164.21731049456903
+I1128 12:42:15.323264 137274321021824 utils.py:1231] [8100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.833650545230833
+I1128 12:42:15.323321 137274321021824 utils.py:1231] [8100] core_hours = 14.833650545230833
+I1128 12:42:15.323381 137274321021824 train.py:125] NOTE: Steps:8100/112603 [7.2%]
+Walltime:14h52m (0s eval)
+ETA:7d23h25m
+Total train time:8d14h15m
+I1128 12:47:27.107865 137274321021824 utils.py:1231] [8150] l2_params = 242.04183160787068
+I1128 12:47:27.108105 137274321021824 utils.py:1231] [8150] train/loss = 5.069432258605957
+I1128 12:47:27.108220 137274321021824 utils.py:1231] [8150] l2_grads = 1.2143360376358032
+I1128 12:47:27.108292 137274321021824 utils.py:1231] [8150] lr = 0.0008149
+I1128 12:47:27.108353 137274321021824 utils.py:1231] [8150] uptime = 53836.470712943
+I1128 12:47:27.108434 137274321021824 utils.py:1231] [8150] examples_seen = 8345600.0
+I1128 12:47:27.108514 137274321021824 utils.py:1231] [8150] progress = 0.07237817820129126
+I1128 12:47:27.108607 137274321021824 utils.py:1231] [8150] epoch = 6.5140610084399615
+I1128 12:47:27.108681 137274321021824 utils.py:1231] [8150] img/sec/core = 164.21552457866264
+I1128 12:47:27.108774 137274321021824 utils.py:1231] [8150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 14.920257595683056
+I1128 12:47:27.108840 137274321021824 utils.py:1231] [8150] core_hours = 14.920257595683056
+I1128 12:47:27.108924 137274321021824 train.py:125] NOTE: Steps:8150/112603 [7.2%]
+Walltime:14h57m (0s eval)
+ETA:7d23h16m
+Total train time:8d14h11m
+I1128 12:52:38.837847 137274321021824 utils.py:1231] [8200] l2_params = 242.635399060014
+I1128 12:52:38.838080 137274321021824 utils.py:1231] [8200] train/loss = 4.384156882762909
+I1128 12:52:38.838204 137274321021824 utils.py:1231] [8200] l2_grads = 1.3072513341903687
+I1128 12:52:38.838280 137274321021824 utils.py:1231] [8200] lr = 0.0008198999999999999
+I1128 12:52:38.838341 137274321021824 utils.py:1231] [8200] uptime = 54148.200700548
+I1128 12:52:38.838400 137274321021824 utils.py:1231] [8200] examples_seen = 8396800.0
+I1128 12:52:38.838447 137274321021824 utils.py:1231] [8200] progress = 0.07282221610436666
+I1128 12:52:38.838493 137274321021824 utils.py:1231] [8200] epoch = 6.554024572908919
+I1128 12:52:38.838541 137274321021824 utils.py:1231] [8200] img/sec/core = 164.2447054688801
+I1128 12:52:38.838596 137274321021824 utils.py:1231] [8200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.006849258906666
+I1128 12:52:38.838643 137274321021824 utils.py:1231] [8200] core_hours = 15.006849258906666
+I1128 12:52:38.838700 137274321021824 train.py:125] NOTE: Steps:8200/112603 [7.3%]
+Walltime:15h2m (0s eval)
+ETA:7d23h6m
+Total train time:8d14h7m
+I1128 12:57:50.615458 137274321021824 utils.py:1231] [8250] l2_params = 243.14711443187454
+I1128 12:57:50.615673 137274321021824 utils.py:1231] [8250] train/loss = 4.35672265291214
+I1128 12:57:50.615801 137274321021824 utils.py:1231] [8250] l2_grads = 1.2934139966964722
+I1128 12:57:50.615877 137274321021824 utils.py:1231] [8250] lr = 0.0008248999999999999
+I1128 12:57:50.615936 137274321021824 utils.py:1231] [8250] uptime = 54459.97829869701
+I1128 12:57:50.615985 137274321021824 utils.py:1231] [8250] examples_seen = 8448000.0
+I1128 12:57:50.616030 137274321021824 utils.py:1231] [8250] progress = 0.07326625400744208
+I1128 12:57:50.616076 137274321021824 utils.py:1231] [8250] epoch = 6.593988137377875
+I1128 12:57:50.616123 137274321021824 utils.py:1231] [8250] img/sec/core = 164.21962419355503
+I1128 12:57:50.616176 137274321021824 utils.py:1231] [8250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.09345414728139
+I1128 12:57:50.616223 137274321021824 utils.py:1231] [8250] core_hours = 15.09345414728139
+I1128 12:57:50.616280 137274321021824 train.py:125] NOTE: Steps:8250/112603 [7.3%]
+Walltime:15h7m (0s eval)
+ETA:7d22h57m
+Total train time:8d14h3m
+I1128 13:03:02.385979 137274321021824 utils.py:1231] [8300] l2_params = 243.79450502742998
+I1128 13:03:02.386230 137274321021824 utils.py:1231] [8300] train/loss = 4.937082290649414
+I1128 13:03:02.386354 137274321021824 utils.py:1231] [8300] l2_grads = 1.3648029565811157
+I1128 13:03:02.386445 137274321021824 utils.py:1231] [8300] lr = 0.0008299
+I1128 13:03:02.386507 137274321021824 utils.py:1231] [8300] uptime = 54771.74886830601
+I1128 13:03:02.386569 137274321021824 utils.py:1231] [8300] examples_seen = 8499200.0
+I1128 13:03:02.386632 137274321021824 utils.py:1231] [8300] progress = 0.07371029191051748
+I1128 13:03:02.386703 137274321021824 utils.py:1231] [8300] epoch = 6.633951701846832
+I1128 13:03:02.386767 137274321021824 utils.py:1231] [8300] img/sec/core = 164.22332635248947
+I1128 13:03:02.386836 137274321021824 utils.py:1231] [8300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.18005708328389
+I1128 13:03:02.386908 137274321021824 utils.py:1231] [8300] core_hours = 15.18005708328389
+I1128 13:03:02.386975 137274321021824 train.py:125] NOTE: Steps:8300/112603 [7.4%]
+Walltime:15h12m (0s eval)
+ETA:7d22h48m
+Total train time:8d13h59m
+I1128 13:08:14.171370 137274321021824 utils.py:1231] [8350] l2_params = 244.37279402948792
+I1128 13:08:14.171619 137274321021824 utils.py:1231] [8350] train/loss = 4.348965227603912
+I1128 13:08:14.171765 137274321021824 utils.py:1231] [8350] l2_grads = 1.4410054683685303
+I1128 13:08:14.171861 137274321021824 utils.py:1231] [8350] lr = 0.0008349
+I1128 13:08:14.171948 137274321021824 utils.py:1231] [8350] uptime = 55083.53430520701
+I1128 13:08:14.172034 137274321021824 utils.py:1231] [8350] examples_seen = 8550400.0
+I1128 13:08:14.172096 137274321021824 utils.py:1231] [8350] progress = 0.07415432981359289
+I1128 13:08:14.172158 137274321021824 utils.py:1231] [8350] epoch = 6.673915266315789
+I1128 13:08:14.172215 137274321021824 utils.py:1231] [8350] img/sec/core = 164.21549546670138
+I1128 13:08:14.172274 137274321021824 utils.py:1231] [8350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.266664149089724
+I1128 13:08:14.172324 137274321021824 utils.py:1231] [8350] core_hours = 15.266664149089724
+I1128 13:08:14.172388 137274321021824 train.py:125] NOTE: Steps:8350/112603 [7.4%]
+Walltime:15h18m (0s eval)
+ETA:7d22h39m
+Total train time:8d13h55m
+I1128 13:13:25.969437 137274321021824 utils.py:1231] [8400] l2_params = 244.94153871572976
+I1128 13:13:25.969658 137274321021824 utils.py:1231] [8400] train/loss = 4.23950058221817
+I1128 13:13:25.969766 137274321021824 utils.py:1231] [8400] l2_grads = 1.3639397621154785
+I1128 13:13:25.969853 137274321021824 utils.py:1231] [8400] lr = 0.0008399
+I1128 13:13:25.969936 137274321021824 utils.py:1231] [8400] uptime = 55395.332297076005
+I1128 13:13:25.969995 137274321021824 utils.py:1231] [8400] examples_seen = 8601600.0
+I1128 13:13:25.970046 137274321021824 utils.py:1231] [8400] progress = 0.07459836771666829
+I1128 13:13:25.970100 137274321021824 utils.py:1231] [8400] epoch = 6.713878830784745
+I1128 13:13:25.970154 137274321021824 utils.py:1231] [8400] img/sec/core = 164.2088831075987
+I1128 13:13:25.970210 137274321021824 utils.py:1231] [8400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.353274702386667
+I1128 13:13:25.970275 137274321021824 utils.py:1231] [8400] core_hours = 15.353274702386667
+I1128 13:13:25.970333 137274321021824 train.py:125] NOTE: Steps:8400/112603 [7.5%]
+Walltime:15h23m (0s eval)
+ETA:7d22h30m
+Total train time:8d13h51m
+I1128 13:18:37.754398 137274321021824 utils.py:1231] [8450] l2_params = 245.52187416278593
+I1128 13:18:37.754600 137274321021824 utils.py:1231] [8450] train/loss = 4.953055262565613
+I1128 13:18:37.754693 137274321021824 utils.py:1231] [8450] l2_grads = 1.1825166940689087
+I1128 13:18:37.754765 137274321021824 utils.py:1231] [8450] lr = 0.0008449
+I1128 13:18:37.754817 137274321021824 utils.py:1231] [8450] uptime = 55707.117179868
+I1128 13:18:37.754870 137274321021824 utils.py:1231] [8450] examples_seen = 8652800.0
+I1128 13:18:37.754925 137274321021824 utils.py:1231] [8450] progress = 0.07504240561974371
+I1128 13:18:37.754973 137274321021824 utils.py:1231] [8450] epoch = 6.753842395253702
+I1128 13:18:37.755024 137274321021824 utils.py:1231] [8450] img/sec/core = 164.21578731306764
+I1128 13:18:37.755079 137274321021824 utils.py:1231] [8450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.439881614273332
+I1128 13:18:37.755128 137274321021824 utils.py:1231] [8450] core_hours = 15.439881614273332
+I1128 13:18:37.755188 137274321021824 train.py:125] NOTE: Steps:8450/112603 [7.5%]
+Walltime:15h28m (0s eval)
+ETA:7d22h21m
+Total train time:8d13h47m
+I1128 13:23:49.533127 137274321021824 utils.py:1231] [8500] l2_params = 246.12696412507225
+I1128 13:23:49.533344 137274321021824 utils.py:1231] [8500] train/loss = 4.249797165393829
+I1128 13:23:49.533448 137274321021824 utils.py:1231] [8500] l2_grads = 1.379814863204956
+I1128 13:23:49.533525 137274321021824 utils.py:1231] [8500] lr = 0.0008499
+I1128 13:23:49.533577 137274321021824 utils.py:1231] [8500] uptime = 56018.895939794005
+I1128 13:23:49.533632 137274321021824 utils.py:1231] [8500] examples_seen = 8704000.0
+I1128 13:23:49.533683 137274321021824 utils.py:1231] [8500] progress = 0.07548644352281911
+I1128 13:23:49.533729 137274321021824 utils.py:1231] [8500] epoch = 6.793805959722659
+I1128 13:23:49.533777 137274321021824 utils.py:1231] [8500] img/sec/core = 164.219012264182
+I1128 13:23:49.533832 137274321021824 utils.py:1231] [8500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.52648682536389
+I1128 13:23:49.533886 137274321021824 utils.py:1231] [8500] core_hours = 15.52648682536389
+I1128 13:23:49.533945 137274321021824 train.py:125] NOTE: Steps:8500/112603 [7.5%]
+Walltime:15h33m (0s eval)
+ETA:7d22h12m
+Total train time:8d13h44m
+I1128 13:29:01.301313 137274321021824 utils.py:1231] [8550] l2_params = 246.71156166472082
+I1128 13:29:01.301540 137274321021824 utils.py:1231] [8550] train/loss = 4.62103134393692
+I1128 13:29:01.301632 137274321021824 utils.py:1231] [8550] l2_grads = 1.2153469324111938
+I1128 13:29:01.301689 137274321021824 utils.py:1231] [8550] lr = 0.0008549
+I1128 13:29:01.301750 137274321021824 utils.py:1231] [8550] uptime = 56330.664111189006
+I1128 13:29:01.301803 137274321021824 utils.py:1231] [8550] examples_seen = 8755200.0
+I1128 13:29:01.301849 137274321021824 utils.py:1231] [8550] progress = 0.07593048142589451
+I1128 13:29:01.301898 137274321021824 utils.py:1231] [8550] epoch = 6.833769524191616
+I1128 13:29:01.301947 137274321021824 utils.py:1231] [8550] img/sec/core = 164.22458960742082
+I1128 13:29:01.302000 137274321021824 utils.py:1231] [8550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.613089095195836
+I1128 13:29:01.302047 137274321021824 utils.py:1231] [8550] core_hours = 15.613089095195836
+I1128 13:29:01.302102 137274321021824 train.py:125] NOTE: Steps:8550/112603 [7.6%]
+Walltime:15h38m (0s eval)
+ETA:7d22h3m
+Total train time:8d13h40m
+I1128 13:34:13.083134 137274321021824 utils.py:1231] [8600] l2_params = 247.23210040806487
+I1128 13:34:13.083328 137274321021824 utils.py:1231] [8600] train/loss = 4.350654602050781
+I1128 13:34:13.083427 137274321021824 utils.py:1231] [8600] l2_grads = 1.3586883544921875
+I1128 13:34:13.083493 137274321021824 utils.py:1231] [8600] lr = 0.0008599
+I1128 13:34:13.083558 137274321021824 utils.py:1231] [8600] uptime = 56642.44592006001
+I1128 13:34:13.083613 137274321021824 utils.py:1231] [8600] examples_seen = 8806400.0
+I1128 13:34:13.083673 137274321021824 utils.py:1231] [8600] progress = 0.07637451932896992
+I1128 13:34:13.083731 137274321021824 utils.py:1231] [8600] epoch = 6.873733088660573
+I1128 13:34:13.083784 137274321021824 utils.py:1231] [8600] img/sec/core = 164.21740635029573
+I1128 13:34:13.083860 137274321021824 utils.py:1231] [8600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.699695153215556
+I1128 13:34:13.083938 137274321021824 utils.py:1231] [8600] core_hours = 15.699695153215556
+I1128 13:34:13.084002 137274321021824 train.py:125] NOTE: Steps:8600/112603 [7.6%]
+Walltime:15h44m (0s eval)
+ETA:7d21h54m
+Total train time:8d13h36m
+I1128 13:39:25.059250 137274321021824 utils.py:1231] [8650] l2_params = 247.72980084823428
+I1128 13:39:25.059487 137274321021824 utils.py:1231] [8650] train/loss = 4.382549703121185
+I1128 13:39:25.059602 137274321021824 utils.py:1231] [8650] l2_grads = 1.5505417585372925
+I1128 13:39:25.059685 137274321021824 utils.py:1231] [8650] lr = 0.0008649
+I1128 13:39:25.059746 137274321021824 utils.py:1231] [8650] uptime = 56954.422108168
+I1128 13:39:25.059807 137274321021824 utils.py:1231] [8650] examples_seen = 8857600.0
+I1128 13:39:25.059864 137274321021824 utils.py:1231] [8650] progress = 0.07681855723204532
+I1128 13:39:25.059927 137274321021824 utils.py:1231] [8650] epoch = 6.91369665312953
+I1128 13:39:25.059985 137274321021824 utils.py:1231] [8650] img/sec/core = 164.11508939354422
+I1128 13:39:25.060048 137274321021824 utils.py:1231] [8650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.786355205467776
+I1128 13:39:25.060104 137274321021824 utils.py:1231] [8650] core_hours = 15.786355205467776
+I1128 13:39:25.060191 137274321021824 train.py:125] NOTE: Steps:8650/112603 [7.7%]
+Walltime:15h49m (0s eval)
+ETA:7d21h45m
+Total train time:8d13h32m
+I1128 13:44:36.841424 137274321021824 utils.py:1231] [8700] l2_params = 248.3427749879619
+I1128 13:44:36.841640 137274321021824 utils.py:1231] [8700] train/loss = 4.329276144504547
+I1128 13:44:36.841738 137274321021824 utils.py:1231] [8700] l2_grads = 1.4208059310913086
+I1128 13:44:36.841811 137274321021824 utils.py:1231] [8700] lr = 0.0008699000000000001
+I1128 13:44:36.841867 137274321021824 utils.py:1231] [8700] uptime = 57266.204229362
+I1128 13:44:36.841928 137274321021824 utils.py:1231] [8700] examples_seen = 8908800.0
+I1128 13:44:36.841980 137274321021824 utils.py:1231] [8700] progress = 0.07726259513512074
+I1128 13:44:36.842031 137274321021824 utils.py:1231] [8700] epoch = 6.953660217598486
+I1128 13:44:36.842083 137274321021824 utils.py:1231] [8700] img/sec/core = 164.21724184800647
+I1128 13:44:36.842140 137274321021824 utils.py:1231] [8700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.872961350243889
+I1128 13:44:36.842191 137274321021824 utils.py:1231] [8700] core_hours = 15.872961350243889
+I1128 13:44:36.842252 137274321021824 train.py:125] NOTE: Steps:8700/112603 [7.7%]
+Walltime:15h54m (0s eval)
+ETA:7d21h36m
+Total train time:8d13h29m
+I1128 13:49:48.631160 137274321021824 utils.py:1231] [8750] l2_params = 248.9908462948028
+I1128 13:49:48.631469 137274321021824 utils.py:1231] [8750] train/loss = 4.45309579372406
+I1128 13:49:48.631615 137274321021824 utils.py:1231] [8750] l2_grads = 1.2565128803253174
+I1128 13:49:48.631693 137274321021824 utils.py:1231] [8750] lr = 0.0008749000000000001
+I1128 13:49:48.631745 137274321021824 utils.py:1231] [8750] uptime = 57577.994106984
+I1128 13:49:48.631797 137274321021824 utils.py:1231] [8750] examples_seen = 8960000.0
+I1128 13:49:48.631845 137274321021824 utils.py:1231] [8750] progress = 0.07770663303819614
+I1128 13:49:48.631905 137274321021824 utils.py:1231] [8750] epoch = 6.993623782067443
+I1128 13:49:48.631957 137274321021824 utils.py:1231] [8750] img/sec/core = 164.21315659924446
+I1128 13:49:48.632011 137274321021824 utils.py:1231] [8750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 15.959569649583331
+I1128 13:49:48.632061 137274321021824 utils.py:1231] [8750] core_hours = 15.959569649583331
+I1128 13:49:48.632122 137274321021824 train.py:125] NOTE: Steps:8750/112603 [7.8%]
+Walltime:15h59m (0s eval)
+ETA:7d21h27m
+Total train time:8d13h25m
+I1128 13:55:00.408375 137274321021824 utils.py:1231] [8800] l2_params = 249.66619165124246
+I1128 13:55:00.408586 137274321021824 utils.py:1231] [8800] train/loss = 5.018545508384705
+I1128 13:55:00.408671 137274321021824 utils.py:1231] [8800] l2_grads = 1.1477620601654053
+I1128 13:55:00.408743 137274321021824 utils.py:1231] [8800] lr = 0.0008799000000000001
+I1128 13:55:00.408797 137274321021824 utils.py:1231] [8800] uptime = 57889.771159839
+I1128 13:55:00.408866 137274321021824 utils.py:1231] [8800] examples_seen = 9011200.0
+I1128 13:55:00.408914 137274321021824 utils.py:1231] [8800] progress = 0.07815067094127155
+I1128 13:55:00.408958 137274321021824 utils.py:1231] [8800] epoch = 7.0335873465364
+I1128 13:55:00.409005 137274321021824 utils.py:1231] [8800] img/sec/core = 164.2199114115414
+I1128 13:55:00.409055 137274321021824 utils.py:1231] [8800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.0461743864875
+I1128 13:55:00.409099 137274321021824 utils.py:1231] [8800] core_hours = 16.0461743864875
+I1128 13:55:00.409154 137274321021824 train.py:125] NOTE: Steps:8800/112603 [7.8%]
+Walltime:16h4m (0s eval)
+ETA:7d21h19m
+Total train time:8d13h22m
+I1128 14:00:12.197672 137274321021824 utils.py:1231] [8850] l2_params = 250.38565653898792
+I1128 14:00:12.197926 137274321021824 utils.py:1231] [8850] train/loss = 4.399335265159607
+I1128 14:00:12.198064 137274321021824 utils.py:1231] [8850] l2_grads = 1.3066332340240479
+I1128 14:00:12.198159 137274321021824 utils.py:1231] [8850] lr = 0.0008849
+I1128 14:00:12.198233 137274321021824 utils.py:1231] [8850] uptime = 58201.560591983
+I1128 14:00:12.198306 137274321021824 utils.py:1231] [8850] examples_seen = 9062400.0
+I1128 14:00:12.198374 137274321021824 utils.py:1231] [8850] progress = 0.07859470884434695
+I1128 14:00:12.198444 137274321021824 utils.py:1231] [8850] epoch = 7.0735509110053565
+I1128 14:00:12.198521 137274321021824 utils.py:1231] [8850] img/sec/core = 164.21339122345202
+I1128 14:00:12.198597 137274321021824 utils.py:1231] [8850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.132782562083055
+I1128 14:00:12.198677 137274321021824 utils.py:1231] [8850] core_hours = 16.132782562083055
+I1128 14:00:12.198772 137274321021824 train.py:125] NOTE: Steps:8850/112603 [7.9%]
+Walltime:16h10m (0s eval)
+ETA:7d21h10m
+Total train time:8d13h18m
+I1128 14:05:23.873236 137274321021824 utils.py:1231] [8900] l2_params = 250.9880192985211
+I1128 14:05:23.873454 137274321021824 utils.py:1231] [8900] train/loss = 4.993966341018677
+I1128 14:05:23.873569 137274321021824 utils.py:1231] [8900] l2_grads = 1.2213099002838135
+I1128 14:05:23.873655 137274321021824 utils.py:1231] [8900] lr = 0.0008899
+I1128 14:05:23.873754 137274321021824 utils.py:1231] [8900] uptime = 58513.236107123
+I1128 14:05:23.873835 137274321021824 utils.py:1231] [8900] examples_seen = 9113600.0
+I1128 14:05:23.873920 137274321021824 utils.py:1231] [8900] progress = 0.07903874674742235
+I1128 14:05:23.873993 137274321021824 utils.py:1231] [8900] epoch = 7.113514475474314
+I1128 14:05:23.874087 137274321021824 utils.py:1231] [8900] img/sec/core = 164.273411008888
+I1128 14:05:23.874170 137274321021824 utils.py:1231] [8900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.219359094066387
+I1128 14:05:23.874280 137274321021824 utils.py:1231] [8900] core_hours = 16.219359094066387
+I1128 14:05:23.874403 137274321021824 train.py:125] NOTE: Steps:8900/112603 [7.9%]
+Walltime:16h15m (0s eval)
+ETA:7d21h1m
+Total train time:8d13h15m
+I1128 14:10:35.659017 137274321021824 utils.py:1231] [8950] l2_params = 251.60696241103594
+I1128 14:10:35.659247 137274321021824 utils.py:1231] [8950] train/loss = 4.08944696187973
+I1128 14:10:35.659389 137274321021824 utils.py:1231] [8950] l2_grads = 1.315635323524475
+I1128 14:10:35.659490 137274321021824 utils.py:1231] [8950] lr = 0.0008949
+I1128 14:10:35.659614 137274321021824 utils.py:1231] [8950] uptime = 58825.02195542901
+I1128 14:10:35.659715 137274321021824 utils.py:1231] [8950] examples_seen = 9164800.0
+I1128 14:10:35.659817 137274321021824 utils.py:1231] [8950] progress = 0.07948278465049777
+I1128 14:10:35.659911 137274321021824 utils.py:1231] [8950] epoch = 7.153478039943271
+I1128 14:10:35.660014 137274321021824 utils.py:1231] [8950] img/sec/core = 164.215278782468
+I1128 14:10:35.660102 137274321021824 utils.py:1231] [8950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.30596627415139
+I1128 14:10:35.660193 137274321021824 utils.py:1231] [8950] core_hours = 16.30596627415139
+I1128 14:10:35.660299 137274321021824 train.py:125] NOTE: Steps:8950/112603 [7.9%]
+Walltime:16h20m (0s eval)
+ETA:7d20h53m
+Total train time:8d13h11m
+I1128 14:15:47.466009 137274321021824 utils.py:1231] [9000] l2_params = 252.2076239745895
+I1128 14:15:47.466264 137274321021824 utils.py:1231] [9000] train/loss = 5.597573459148407
+I1128 14:15:47.466434 137274321021824 utils.py:1231] [9000] l2_grads = 0.9173852801322937
+I1128 14:15:47.466501 137274321021824 utils.py:1231] [9000] lr = 0.0008999
+I1128 14:15:47.466560 137274321021824 utils.py:1231] [9000] uptime = 59136.828922449
+I1128 14:15:47.466621 137274321021824 utils.py:1231] [9000] examples_seen = 9216000.0
+I1128 14:15:47.466675 137274321021824 utils.py:1231] [9000] progress = 0.07992682255357317
+I1128 14:15:47.466730 137274321021824 utils.py:1231] [9000] epoch = 7.193441604412227
+I1128 14:15:47.466787 137274321021824 utils.py:1231] [9000] img/sec/core = 164.20415646683537
+I1128 14:15:47.466859 137274321021824 utils.py:1231] [9000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.39257932054583
+I1128 14:15:47.466922 137274321021824 utils.py:1231] [9000] core_hours = 16.39257932054583
+I1128 14:15:47.466994 137274321021824 train.py:125] NOTE: Steps:9000/112603 [8.0%]
+Walltime:16h25m (0s eval)
+ETA:7d20h44m
+Total train time:8d13h8m
+I1128 14:20:59.600853 137274321021824 utils.py:1231] [9050] l2_params = 252.86196940289213
+I1128 14:20:59.601128 137274321021824 utils.py:1231] [9050] train/loss = 4.21183380484581
+I1128 14:20:59.601259 137274321021824 utils.py:1231] [9050] l2_grads = 1.3639013767242432
+I1128 14:20:59.601342 137274321021824 utils.py:1231] [9050] lr = 0.0009049
+I1128 14:20:59.601408 137274321021824 utils.py:1231] [9050] uptime = 59448.963770548
+I1128 14:20:59.601471 137274321021824 utils.py:1231] [9050] examples_seen = 9267200.0
+I1128 14:20:59.601537 137274321021824 utils.py:1231] [9050] progress = 0.08037086045664858
+I1128 14:20:59.601588 137274321021824 utils.py:1231] [9050] epoch = 7.233405168881184
+I1128 14:20:59.601644 137274321021824 utils.py:1231] [9050] img/sec/core = 164.03166872210682
+I1128 14:20:59.601702 137274321021824 utils.py:1231] [9050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.479283445017778
+I1128 14:20:59.601764 137274321021824 utils.py:1231] [9050] core_hours = 16.479283445017778
+I1128 14:20:59.601828 137274321021824 train.py:125] NOTE: Steps:9050/112603 [8.0%]
+Walltime:16h30m (0s eval)
+ETA:7d20h36m
+Total train time:8d13h5m
+I1128 14:25:59.342367 137274321021824 utils.py:1231] [9100] l2_params = 253.5119045837741
+I1128 14:25:59.342656 137274321021824 utils.py:1231] [9100] train/loss = 4.108033686876297
+I1128 14:25:59.342870 137274321021824 utils.py:1231] [9100] l2_grads = 1.492827296257019
+I1128 14:25:59.343016 137274321021824 utils.py:1231] [9100] lr = 0.0009099
+I1128 14:25:59.343097 137274321021824 utils.py:1231] [9100] uptime = 59748.705455899006
+I1128 14:25:59.343169 137274321021824 utils.py:1231] [9100] examples_seen = 9318400.0
+I1128 14:25:59.343240 137274321021824 utils.py:1231] [9100] progress = 0.08081489835972398
+I1128 14:25:59.343314 137274321021824 utils.py:1231] [9100] epoch = 7.273368733350141
+I1128 14:25:59.343388 137274321021824 utils.py:1231] [9100] img/sec/core = 170.81374564249572
+I1128 14:25:59.343466 137274321021824 utils.py:1231] [9100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.562545024281945
+I1128 14:25:59.343544 137274321021824 utils.py:1231] [9100] core_hours = 16.562545024281945
+I1128 14:25:59.343636 137274321021824 train.py:125] NOTE: Steps:9100/112603 [8.1%]
+Walltime:16h35m (0s eval)
+ETA:7d20h25m
+Total train time:8d12h59m
+I1128 14:31:04.336511 137274321021824 utils.py:1231] [9150] l2_params = 254.1272856419687
+I1128 14:31:04.336713 137274321021824 utils.py:1231] [9150] train/loss = 4.411386668682098
+I1128 14:31:04.336813 137274321021824 utils.py:1231] [9150] l2_grads = 1.1079319715499878
+I1128 14:31:04.336890 137274321021824 utils.py:1231] [9150] lr = 0.0009149000000000001
+I1128 14:31:04.336949 137274321021824 utils.py:1231] [9150] uptime = 60053.699310647
+I1128 14:31:04.337006 137274321021824 utils.py:1231] [9150] examples_seen = 9369600.0
+I1128 14:31:04.337060 137274321021824 utils.py:1231] [9150] progress = 0.0812589362627994
+I1128 14:31:04.337112 137274321021824 utils.py:1231] [9150] epoch = 7.313332297819097
+I1128 14:31:04.337165 137274321021824 utils.py:1231] [9150] img/sec/core = 167.87223481044944
+I1128 14:31:04.337239 137274321021824 utils.py:1231] [9150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.647265539489723
+I1128 14:31:04.337288 137274321021824 utils.py:1231] [9150] core_hours = 16.647265539489723
+I1128 14:31:04.337345 137274321021824 train.py:125] NOTE: Steps:9150/112603 [8.1%]
+Walltime:16h40m (0s eval)
+ETA:7d20h15m
+Total train time:8d12h54m
+I1128 14:36:09.402441 137274321021824 utils.py:1231] [9200] l2_params = 254.84832989157567
+I1128 14:36:09.402691 137274321021824 utils.py:1231] [9200] train/loss = 4.278309851884842
+I1128 14:36:09.402802 137274321021824 utils.py:1231] [9200] l2_grads = 1.4060827493667603
+I1128 14:36:09.402876 137274321021824 utils.py:1231] [9200] lr = 0.0009199000000000001
+I1128 14:36:09.402944 137274321021824 utils.py:1231] [9200] uptime = 60358.765301366
+I1128 14:36:09.403008 137274321021824 utils.py:1231] [9200] examples_seen = 9420800.0
+I1128 14:36:09.403056 137274321021824 utils.py:1231] [9200] progress = 0.0817029741658748
+I1128 14:36:09.403104 137274321021824 utils.py:1231] [9200] epoch = 7.353295862288054
+I1128 14:36:09.403154 137274321021824 utils.py:1231] [9200] img/sec/core = 167.832539705028
+I1128 14:36:09.403208 137274321021824 utils.py:1231] [9200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.732006092467223
+I1128 14:36:09.403256 137274321021824 utils.py:1231] [9200] core_hours = 16.732006092467223
+I1128 14:36:09.403314 137274321021824 train.py:125] NOTE: Steps:9200/112603 [8.2%]
+Walltime:16h45m (0s eval)
+ETA:7d20h5m
+Total train time:8d12h50m
+I1128 14:41:14.624931 137274321021824 utils.py:1231] [9250] l2_params = 255.49189074363488
+I1128 14:41:14.625199 137274321021824 utils.py:1231] [9250] train/loss = 4.661392867565155
+I1128 14:41:14.625340 137274321021824 utils.py:1231] [9250] l2_grads = 1.3262114524841309
+I1128 14:41:14.625571 137274321021824 utils.py:1231] [9250] lr = 0.0009249000000000001
+I1128 14:41:14.625645 137274321021824 utils.py:1231] [9250] uptime = 60663.988004716
+I1128 14:41:14.625705 137274321021824 utils.py:1231] [9250] examples_seen = 9472000.0
+I1128 14:41:14.625772 137274321021824 utils.py:1231] [9250] progress = 0.08214701206895021
+I1128 14:41:14.625828 137274321021824 utils.py:1231] [9250] epoch = 7.3932594267570115
+I1128 14:41:14.625892 137274321021824 utils.py:1231] [9250] img/sec/core = 167.7463682683186
+I1128 14:41:14.625958 137274321021824 utils.py:1231] [9250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.81679017673111
+I1128 14:41:14.626008 137274321021824 utils.py:1231] [9250] core_hours = 16.81679017673111
+I1128 14:41:14.626069 137274321021824 train.py:125] NOTE: Steps:9250/112603 [8.2%]
+Walltime:16h51m (0s eval)
+ETA:7d19h56m
+Total train time:8d12h45m
+I1128 14:46:19.712679 137274321021824 utils.py:1231] [9300] l2_params = 256.02942844438036
+I1128 14:46:19.712875 137274321021824 utils.py:1231] [9300] train/loss = 4.143375784158707
+I1128 14:46:19.712966 137274321021824 utils.py:1231] [9300] l2_grads = 1.3037763833999634
+I1128 14:46:19.713019 137274321021824 utils.py:1231] [9300] lr = 0.0009299
+I1128 14:46:19.713066 137274321021824 utils.py:1231] [9300] uptime = 60969.07542850201
+I1128 14:46:19.713113 137274321021824 utils.py:1231] [9300] examples_seen = 9523200.0
+I1128 14:46:19.713157 137274321021824 utils.py:1231] [9300] progress = 0.08259104997202561
+I1128 14:46:19.713200 137274321021824 utils.py:1231] [9300] epoch = 7.433222991225969
+I1128 14:46:19.713246 137274321021824 utils.py:1231] [9300] img/sec/core = 167.82074909751668
+I1128 14:46:19.713295 137274321021824 utils.py:1231] [9300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.901536683338335
+I1128 14:46:19.713340 137274321021824 utils.py:1231] [9300] core_hours = 16.901536683338335
+I1128 14:46:19.713394 137274321021824 train.py:125] NOTE: Steps:9300/112603 [8.3%]
+Walltime:16h56m (0s eval)
+ETA:7d19h46m
+Total train time:8d12h41m
+I1128 14:51:30.874625 137274321021824 utils.py:1231] [9350] l2_params = 256.67787976386705
+I1128 14:51:30.874903 137274321021824 utils.py:1231] [9350] train/loss = 4.536966264247894
+I1128 14:51:30.875107 137274321021824 utils.py:1231] [9350] l2_grads = 1.2821263074874878
+I1128 14:51:30.875211 137274321021824 utils.py:1231] [9350] lr = 0.0009349
+I1128 14:51:30.875288 137274321021824 utils.py:1231] [9350] uptime = 61280.23764657701
+I1128 14:51:30.875374 137274321021824 utils.py:1231] [9350] examples_seen = 9574400.0
+I1128 14:51:30.875442 137274321021824 utils.py:1231] [9350] progress = 0.08303508787510101
+I1128 14:51:30.875512 137274321021824 utils.py:1231] [9350] epoch = 7.473186555694925
+I1128 14:51:30.875588 137274321021824 utils.py:1231] [9350] img/sec/core = 164.5443984708306
+I1128 14:51:30.875666 137274321021824 utils.py:1231] [9350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 16.987970632803613
+I1128 14:51:30.875749 137274321021824 utils.py:1231] [9350] core_hours = 16.987970632803613
+I1128 14:51:30.875852 137274321021824 train.py:125] NOTE: Steps:9350/112603 [8.3%]
+Walltime:17h1m (0s eval)
+ETA:7d19h38m
+Total train time:8d12h37m
+I1128 14:56:33.470207 137274321021824 utils.py:1231] [9400] l2_params = 257.37624591296094
+I1128 14:56:33.470441 137274321021824 utils.py:1231] [9400] train/loss = 4.122705310583115
+I1128 14:56:33.470544 137274321021824 utils.py:1231] [9400] l2_grads = 1.2959975004196167
+I1128 14:56:33.470622 137274321021824 utils.py:1231] [9400] lr = 0.0009399
+I1128 14:56:33.470700 137274321021824 utils.py:1231] [9400] uptime = 61582.83304792701
+I1128 14:56:33.470758 137274321021824 utils.py:1231] [9400] examples_seen = 9625600.0
+I1128 14:56:33.470812 137274321021824 utils.py:1231] [9400] progress = 0.08347912577817643
+I1128 14:56:33.470869 137274321021824 utils.py:1231] [9400] epoch = 7.513150120163882
+I1128 14:56:33.470933 137274321021824 utils.py:1231] [9400] img/sec/core = 169.20283577204395
+I1128 14:56:33.470995 137274321021824 utils.py:1231] [9400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.07202491095639
+I1128 14:56:33.471043 137274321021824 utils.py:1231] [9400] core_hours = 17.07202491095639
+I1128 14:56:33.471119 137274321021824 train.py:125] NOTE: Steps:9400/112603 [8.3%]
+Walltime:17h6m (0s eval)
+ETA:7d19h28m
+Total train time:8d12h33m
+I1128 15:01:45.262349 137274321021824 utils.py:1231] [9450] l2_params = 257.9865209839257
+I1128 15:01:45.262592 137274321021824 utils.py:1231] [9450] train/loss = 4.479300260543823
+I1128 15:01:45.262715 137274321021824 utils.py:1231] [9450] l2_grads = 1.1822104454040527
+I1128 15:01:45.262800 137274321021824 utils.py:1231] [9450] lr = 0.0009448999999999999
+I1128 15:01:45.262859 137274321021824 utils.py:1231] [9450] uptime = 61894.625220256
+I1128 15:01:45.262926 137274321021824 utils.py:1231] [9450] examples_seen = 9676800.0
+I1128 15:01:45.262987 137274321021824 utils.py:1231] [9450] progress = 0.08392316368125183
+I1128 15:01:45.263041 137274321021824 utils.py:1231] [9450] epoch = 7.553113684632839
+I1128 15:01:45.263097 137274321021824 utils.py:1231] [9450] img/sec/core = 164.21194803433278
+I1128 15:01:45.263159 137274321021824 utils.py:1231] [9450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.158633847714444
+I1128 15:01:45.263213 137274321021824 utils.py:1231] [9450] core_hours = 17.158633847714444
+I1128 15:01:45.263276 137274321021824 train.py:125] NOTE: Steps:9450/112603 [8.4%]
+Walltime:17h11m (0s eval)
+ETA:7d19h20m
+Total train time:8d12h29m
+I1128 15:06:49.482069 137274321021824 utils.py:1231] [9500] l2_params = 258.6118786551022
+I1128 15:06:49.482351 137274321021824 utils.py:1231] [9500] train/loss = 4.149742513895035
+I1128 15:06:49.482475 137274321021824 utils.py:1231] [9500] l2_grads = 1.2904138565063477
+I1128 15:06:49.482564 137274321021824 utils.py:1231] [9500] lr = 0.0009498999999999999
+I1128 15:06:49.482643 137274321021824 utils.py:1231] [9500] uptime = 62198.84500461101
+I1128 15:06:49.482700 137274321021824 utils.py:1231] [9500] examples_seen = 9728000.0
+I1128 15:06:49.482753 137274321021824 utils.py:1231] [9500] progress = 0.08436720158432724
+I1128 15:06:49.482800 137274321021824 utils.py:1231] [9500] epoch = 7.593077249101795
+I1128 15:06:49.482850 137274321021824 utils.py:1231] [9500] img/sec/core = 168.299376414823
+I1128 15:06:49.482912 137274321021824 utils.py:1231] [9500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.243139343368615
+I1128 15:06:49.482961 137274321021824 utils.py:1231] [9500] core_hours = 17.243139343368615
+I1128 15:06:49.483019 137274321021824 train.py:125] NOTE: Steps:9500/112603 [8.4%]
+Walltime:17h16m (0s eval)
+ETA:7d19h10m
+Total train time:8d12h25m
+I1128 15:11:57.948071 137274321021824 utils.py:1231] [9550] l2_params = 259.26421413228564
+I1128 15:11:57.948317 137274321021824 utils.py:1231] [9550] train/loss = 5.985541582107544
+I1128 15:11:57.948432 137274321021824 utils.py:1231] [9550] l2_grads = 0.866272509098053
+I1128 15:11:57.948504 137274321021824 utils.py:1231] [9550] lr = 0.0009549
+I1128 15:11:57.948564 137274321021824 utils.py:1231] [9550] uptime = 62507.31092656801
+I1128 15:11:57.948621 137274321021824 utils.py:1231] [9550] examples_seen = 9779200.0
+I1128 15:11:57.948677 137274321021824 utils.py:1231] [9550] progress = 0.08481123948740264
+I1128 15:11:57.948728 137274321021824 utils.py:1231] [9550] epoch = 7.633040813570752
+I1128 15:11:57.948777 137274321021824 utils.py:1231] [9550] img/sec/core = 165.98267865432766
+I1128 15:11:57.948831 137274321021824 utils.py:1231] [9550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.32882432169
+I1128 15:11:57.948890 137274321021824 utils.py:1231] [9550] core_hours = 17.32882432169
+I1128 15:11:57.948948 137274321021824 train.py:125] NOTE: Steps:9550/112603 [8.5%]
+Walltime:17h21m (0s eval)
+ETA:7d19h1m
+Total train time:8d12h21m
+I1128 15:17:07.809551 137274321021824 utils.py:1231] [9600] l2_params = 259.94698659138055
+I1128 15:17:07.809747 137274321021824 utils.py:1231] [9600] train/loss = 5.280961751937866
+I1128 15:17:07.809846 137274321021824 utils.py:1231] [9600] l2_grads = 1.0321191549301147
+I1128 15:17:07.809920 137274321021824 utils.py:1231] [9600] lr = 0.0009599
+I1128 15:17:07.809981 137274321021824 utils.py:1231] [9600] uptime = 62817.172342178004
+I1128 15:17:07.810036 137274321021824 utils.py:1231] [9600] examples_seen = 9830400.0
+I1128 15:17:07.810088 137274321021824 utils.py:1231] [9600] progress = 0.08525527739047806
+I1128 15:17:07.810140 137274321021824 utils.py:1231] [9600] epoch = 7.6730043780397095
+I1128 15:17:07.810196 137274321021824 utils.py:1231] [9600] img/sec/core = 165.23515810836778
+I1128 15:17:07.810253 137274321021824 utils.py:1231] [9600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.41489693713722
+I1128 15:17:07.810306 137274321021824 utils.py:1231] [9600] core_hours = 17.41489693713722
+I1128 15:17:07.810367 137274321021824 train.py:125] NOTE: Steps:9600/112603 [8.5%]
+Walltime:17h26m (0s eval)
+ETA:7d18h53m
+Total train time:8d12h18m
+I1128 15:22:16.943738 137274321021824 utils.py:1231] [9650] l2_params = 260.6564831759815
+I1128 15:22:16.944028 137274321021824 utils.py:1231] [9650] train/loss = 4.287293255329132
+I1128 15:22:16.944183 137274321021824 utils.py:1231] [9650] l2_grads = 1.216469168663025
+I1128 15:22:16.944261 137274321021824 utils.py:1231] [9650] lr = 0.0009649
+I1128 15:22:16.944321 137274321021824 utils.py:1231] [9650] uptime = 63126.306683284
+I1128 15:22:16.944380 137274321021824 utils.py:1231] [9650] examples_seen = 9881600.0
+I1128 15:22:16.944435 137274321021824 utils.py:1231] [9650] progress = 0.08569931529355346
+I1128 15:22:16.944491 137274321021824 utils.py:1231] [9650] epoch = 7.712967942508666
+I1128 15:22:16.944551 137274321021824 utils.py:1231] [9650] img/sec/core = 165.62378614042422
+I1128 15:22:16.944615 137274321021824 utils.py:1231] [9650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.500767587444443
+I1128 15:22:16.944669 137274321021824 utils.py:1231] [9650] core_hours = 17.500767587444443
+I1128 15:22:16.944736 137274321021824 train.py:125] NOTE: Steps:9650/112603 [8.6%]
+Walltime:17h32m (0s eval)
+ETA:7d18h44m
+Total train time:8d12h15m
+I1128 15:27:25.191515 137274321021824 utils.py:1231] [9700] l2_params = 261.3283666553435
+I1128 15:27:25.191736 137274321021824 utils.py:1231] [9700] train/loss = 4.20892995595932
+I1128 15:27:25.191843 137274321021824 utils.py:1231] [9700] l2_grads = 1.3062185049057007
+I1128 15:27:25.191919 137274321021824 utils.py:1231] [9700] lr = 0.0009699
+I1128 15:27:25.191977 137274321021824 utils.py:1231] [9700] uptime = 63434.554338499
+I1128 15:27:25.192057 137274321021824 utils.py:1231] [9700] examples_seen = 9932800.0
+I1128 15:27:25.192112 137274321021824 utils.py:1231] [9700] progress = 0.08614335319662886
+I1128 15:27:25.192165 137274321021824 utils.py:1231] [9700] epoch = 7.752931506977623
+I1128 15:27:25.192220 137274321021824 utils.py:1231] [9700] img/sec/core = 166.10020914607713
+I1128 15:27:25.192279 137274321021824 utils.py:1231] [9700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.586391936115277
+I1128 15:27:25.192331 137274321021824 utils.py:1231] [9700] core_hours = 17.586391936115277
+I1128 15:27:25.192392 137274321021824 train.py:125] NOTE: Steps:9700/112603 [8.6%]
+Walltime:17h37m (0s eval)
+ETA:7d18h36m
+Total train time:8d12h11m
+I1128 15:32:34.039235 137274321021824 utils.py:1231] [9750] l2_params = 261.99080672036035
+I1128 15:32:34.039442 137274321021824 utils.py:1231] [9750] train/loss = 5.192637801170349
+I1128 15:32:34.039549 137274321021824 utils.py:1231] [9750] l2_grads = 1.125931739807129
+I1128 15:32:34.039625 137274321021824 utils.py:1231] [9750] lr = 0.0009749
+I1128 15:32:34.039693 137274321021824 utils.py:1231] [9750] uptime = 63743.40205427501
+I1128 15:32:34.039771 137274321021824 utils.py:1231] [9750] examples_seen = 9984000.0
+I1128 15:32:34.039847 137274321021824 utils.py:1231] [9750] progress = 0.08658739109970427
+I1128 15:32:34.039936 137274321021824 utils.py:1231] [9750] epoch = 7.79289507144658
+I1128 15:32:34.040007 137274321021824 utils.py:1231] [9750] img/sec/core = 165.77749287008854
+I1128 15:32:34.040072 137274321021824 utils.py:1231] [9750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.672182968275276
+I1128 15:32:34.040140 137274321021824 utils.py:1231] [9750] core_hours = 17.672182968275276
+I1128 15:32:34.040235 137274321021824 train.py:125] NOTE: Steps:9750/112603 [8.7%]
+Walltime:17h42m (0s eval)
+ETA:7d18h27m
+Total train time:8d12h8m
+I1128 15:37:43.969491 137274321021824 utils.py:1231] [9800] l2_params = 262.6311000177278
+I1128 15:37:43.969709 137274321021824 utils.py:1231] [9800] train/loss = 4.837673246860504
+I1128 15:37:43.969810 137274321021824 utils.py:1231] [9800] l2_grads = 1.1331664323806763
+I1128 15:37:43.969909 137274321021824 utils.py:1231] [9800] lr = 0.0009799
+I1128 15:37:43.970011 137274321021824 utils.py:1231] [9800] uptime = 64053.332372328005
+I1128 15:37:43.970074 137274321021824 utils.py:1231] [9800] examples_seen = 10035200.0
+I1128 15:37:43.970129 137274321021824 utils.py:1231] [9800] progress = 0.08703142900277967
+I1128 15:37:43.970183 137274321021824 utils.py:1231] [9800] epoch = 7.832858635915536
+I1128 15:37:43.970237 137274321021824 utils.py:1231] [9800] img/sec/core = 165.1984237025982
+I1128 15:37:43.970298 137274321021824 utils.py:1231] [9800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.75827472329
+I1128 15:37:43.970353 137274321021824 utils.py:1231] [9800] core_hours = 17.75827472329
+I1128 15:37:43.970417 137274321021824 train.py:125] NOTE: Steps:9800/112603 [8.7%]
+Walltime:17h47m (0s eval)
+ETA:7d18h19m
+Total train time:8d12h5m
+I1128 15:42:52.701163 137274321021824 utils.py:1231] [9850] l2_params = 263.4085852183666
+I1128 15:42:52.701437 137274321021824 utils.py:1231] [9850] train/loss = 4.0066128969192505
+I1128 15:42:52.701553 137274321021824 utils.py:1231] [9850] l2_grads = 1.3070433139801025
+I1128 15:42:52.701625 137274321021824 utils.py:1231] [9850] lr = 0.0009849
+I1128 15:42:52.701689 137274321021824 utils.py:1231] [9850] uptime = 64362.064050941
+I1128 15:42:52.701743 137274321021824 utils.py:1231] [9850] examples_seen = 10086400.0
+I1128 15:42:52.701797 137274321021824 utils.py:1231] [9850] progress = 0.08747546690585509
+I1128 15:42:52.701846 137274321021824 utils.py:1231] [9850] epoch = 7.872822200384493
+I1128 15:42:52.701904 137274321021824 utils.py:1231] [9850] img/sec/core = 165.83980053495154
+I1128 15:42:52.701961 137274321021824 utils.py:1231] [9850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.844033522904724
+I1128 15:42:52.702010 137274321021824 utils.py:1231] [9850] core_hours = 17.844033522904724
+I1128 15:42:52.702069 137274321021824 train.py:125] NOTE: Steps:9850/112603 [8.7%]
+Walltime:17h52m (0s eval)
+ETA:7d18h10m
+Total train time:8d12h1m
+I1128 15:48:01.310495 137274321021824 utils.py:1231] [9900] l2_params = 264.1255934661947
+I1128 15:48:01.310777 137274321021824 utils.py:1231] [9900] train/loss = 6.076690137386322
+I1128 15:48:01.310893 137274321021824 utils.py:1231] [9900] l2_grads = 1.0135551691055298
+I1128 15:48:01.310972 137274321021824 utils.py:1231] [9900] lr = 0.0009899
+I1128 15:48:01.311042 137274321021824 utils.py:1231] [9900] uptime = 64670.673402275
+I1128 15:48:01.311098 137274321021824 utils.py:1231] [9900] examples_seen = 10137600.0
+I1128 15:48:01.311144 137274321021824 utils.py:1231] [9900] progress = 0.08791950480893049
+I1128 15:48:01.311188 137274321021824 utils.py:1231] [9900] epoch = 7.91278576485345
+I1128 15:48:01.311237 137274321021824 utils.py:1231] [9900] img/sec/core = 165.90553649356926
+I1128 15:48:01.311288 137274321021824 utils.py:1231] [9900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 17.929758342719722
+I1128 15:48:01.311334 137274321021824 utils.py:1231] [9900] core_hours = 17.929758342719722
+I1128 15:48:01.311390 137274321021824 train.py:125] NOTE: Steps:9900/112603 [8.8%]
+Walltime:17h57m (0s eval)
+ETA:7d18h2m
+Total train time:8d11h58m
+I1128 15:53:13.079901 137274321021824 utils.py:1231] [9950] l2_params = 264.78434211806706
+I1128 15:53:13.080187 137274321021824 utils.py:1231] [9950] train/loss = 4.649487376213074
+I1128 15:53:13.080319 137274321021824 utils.py:1231] [9950] l2_grads = 1.1809005737304688
+I1128 15:53:13.080378 137274321021824 utils.py:1231] [9950] lr = 0.0009949
+I1128 15:53:13.080438 137274321021824 utils.py:1231] [9950] uptime = 64982.44279985801
+I1128 15:53:13.080492 137274321021824 utils.py:1231] [9950] examples_seen = 10188800.0
+I1128 15:53:13.080538 137274321021824 utils.py:1231] [9950] progress = 0.0883635427120059
+I1128 15:53:13.080584 137274321021824 utils.py:1231] [9950] epoch = 7.9527493293224065
+I1128 15:53:13.080633 137274321021824 utils.py:1231] [9950] img/sec/core = 164.22394371265608
+I1128 15:53:13.080686 137274321021824 utils.py:1231] [9950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.016360953159445
+I1128 15:53:13.080734 137274321021824 utils.py:1231] [9950] core_hours = 18.016360953159445
+I1128 15:53:13.080791 137274321021824 train.py:125] NOTE: Steps:9950/112603 [8.8%]
+Walltime:18h3m (0s eval)
+ETA:7d17h54m
+Total train time:8d11h55m
+I1128 15:58:16.863949 137274321021824 utils.py:1231] [10000] l2_params = 265.52379849083553
+I1128 15:58:16.864197 137274321021824 utils.py:1231] [10000] train/loss = 6.063216686248779
+I1128 15:58:16.864333 137274321021824 utils.py:1231] [10000] l2_grads = 1.1149541139602661
+I1128 15:58:16.864439 137274321021824 utils.py:1231] [10000] lr = 0.0009999
+I1128 15:58:16.864507 137274321021824 utils.py:1231] [10000] uptime = 65286.226868080004
+I1128 15:58:16.864580 137274321021824 utils.py:1231] [10000] examples_seen = 10240000.0
+I1128 15:58:16.864646 137274321021824 utils.py:1231] [10000] progress = 0.0888075806150813
+I1128 15:58:16.864709 137274321021824 utils.py:1231] [10000] epoch = 7.992712893791364
+I1128 15:58:16.864776 137274321021824 utils.py:1231] [10000] img/sec/core = 168.54076745915575
+I1128 15:58:16.864850 137274321021824 utils.py:1231] [10000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.100745416554446
+I1128 15:58:16.864922 137274321021824 utils.py:1231] [10000] core_hours = 18.100745416554446
+I1128 15:58:16.864997 137274321021824 train.py:125] NOTE: Steps:10000/112603 [8.9%]
+Walltime:18h8m (0s eval)
+ETA:7d17h45m
+Total train time:8d11h51m
+/home/jason-chou/.pyenv/versions/3.11.10/lib/python3.11/site-packages/torch/optim/lr_scheduler.py:240: UserWarning: The epoch parameter in `scheduler.step()` was not necessary and is being deprecated where possible. Please use `scheduler.step()` to step the scheduler. During the deprecation, if epoch is different from None, the closed form is used instead of the new chainable form, where available. Please open an issue if you are unable to replicate your use case: https://github.com/pytorch/pytorch/issues/new/choose.
+  warnings.warn(EPOCH_DEPRECATION_WARNING, UserWarning)
+I1128 15:58:17.213204 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:10000/112603 [8.9%]
+Walltime:18h8m (0s eval)
+ETA:7d17h45m
+Total train time:8d11h51m
+I1128 15:59:48.150008 137274321021824 utils.py:1231] [10000] val/acc@1 = 0.3540138711734694
+I1128 15:59:48.150225 137274321021824 utils.py:1231] [10000] val/loss = 3.045865480693019
+I1128 15:59:48.150377 137274321021824 utils.py:1231] [10000] z/secs/eval/val = 90.93693383398931
+I1128 15:59:48.150455 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 90.93693383398931
+I1128 16:04:58.075328 137274321021824 utils.py:1231] [10050] l2_params = 266.1164889761684
+I1128 16:04:58.075582 137274321021824 utils.py:1231] [10050] train/loss = 3.93902388215065
+I1128 16:04:58.075833 137274321021824 utils.py:1231] [10050] l2_grads = 1.4104194641113281
+I1128 16:04:58.075927 137274321021824 utils.py:1231] [10050] lr = 0.0009999994372549106
+I1128 16:04:58.075982 137274321021824 utils.py:1231] [10050] uptime = 65687.43834442101
+I1128 16:04:58.076031 137274321021824 utils.py:1231] [10050] examples_seen = 10291200.0
+I1128 16:04:58.076077 137274321021824 utils.py:1231] [10050] progress = 0.08925161851815672
+I1128 16:04:58.076122 137274321021824 utils.py:1231] [10050] epoch = 8.03267645826032
+I1128 16:04:58.076169 137274321021824 utils.py:1231] [10050] img/sec/core = 127.61349816544876
+I1128 16:04:58.076220 137274321021824 utils.py:1231] [10050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.212193048871388
+I1128 16:04:58.076267 137274321021824 utils.py:1231] [10050] core_hours = 18.212193048871388
+I1128 16:04:58.076325 137274321021824 train.py:125] NOTE: Steps:10050/112603 [8.9%]
+Walltime:18h14m (0s eval)
+ETA:7d17h52m
+Total train time:8d12h5m
+I1128 16:10:09.869188 137274321021824 utils.py:1231] [10100] l2_params = 266.9257894902607
+I1128 16:10:09.869447 137274321021824 utils.py:1231] [10100] train/loss = 4.025555610656738
+I1128 16:10:09.869570 137274321021824 utils.py:1231] [10100] l2_grads = 1.2195003032684326
+I1128 16:10:09.869660 137274321021824 utils.py:1231] [10100] lr = 0.000999997702848216
+I1128 16:10:09.869721 137274321021824 utils.py:1231] [10100] uptime = 65999.232083365
+I1128 16:10:09.869785 137274321021824 utils.py:1231] [10100] examples_seen = 10342400.0
+I1128 16:10:09.869837 137274321021824 utils.py:1231] [10100] progress = 0.08969565642123212
+I1128 16:10:09.869899 137274321021824 utils.py:1231] [10100] epoch = 8.072640022729278
+I1128 16:10:09.869956 137274321021824 utils.py:1231] [10100] img/sec/core = 164.21112294752464
+I1128 16:10:09.870024 137274321021824 utils.py:1231] [10100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.298802420800275
+I1128 16:10:09.870083 137274321021824 utils.py:1231] [10100] core_hours = 18.298802420800275
+I1128 16:10:09.870151 137274321021824 train.py:125] NOTE: Steps:10100/112603 [9.0%]
+Walltime:18h19m (0s eval)
+ETA:7d17h44m
+Total train time:8d12h3m
+I1128 16:15:21.654437 137274321021824 utils.py:1231] [10150] l2_params = 267.62421670339825
+I1128 16:15:21.654685 137274321021824 utils.py:1231] [10150] train/loss = 5.220943033695221
+I1128 16:15:21.654803 137274321021824 utils.py:1231] [10150] l2_grads = 1.0572046041488647
+I1128 16:15:21.654874 137274321021824 utils.py:1231] [10150] lr = 0.0009999947965496018
+I1128 16:15:21.654940 137274321021824 utils.py:1231] [10150] uptime = 66311.017301649
+I1128 16:15:21.654995 137274321021824 utils.py:1231] [10150] examples_seen = 10393600.0
+I1128 16:15:21.655045 137274321021824 utils.py:1231] [10150] progress = 0.09013969432430752
+I1128 16:15:21.655093 137274321021824 utils.py:1231] [10150] epoch = 8.112603587198235
+I1128 16:15:21.655144 137274321021824 utils.py:1231] [10150] img/sec/core = 164.21561061102906
+I1128 16:15:21.655201 137274321021824 utils.py:1231] [10150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.385409425879164
+I1128 16:15:21.655251 137274321021824 utils.py:1231] [10150] core_hours = 18.385409425879164
+I1128 16:15:21.655311 137274321021824 train.py:125] NOTE: Steps:10150/112603 [9.0%]
+Walltime:18h25m (0s eval)
+ETA:7d17h37m
+Total train time:8d12h0m
+I1128 16:20:33.394328 137274321021824 utils.py:1231] [10200] l2_params = 268.28338784655756
+I1128 16:20:33.394620 137274321021824 utils.py:1231] [10200] train/loss = 4.013728559017181
+I1128 16:20:33.394856 137274321021824 utils.py:1231] [10200] l2_grads = 1.215476393699646
+I1128 16:20:33.394958 137274321021824 utils.py:1231] [10200] lr = 0.0009999907183658799
+I1128 16:20:33.395043 137274321021824 utils.py:1231] [10200] uptime = 66622.757402198
+I1128 16:20:33.395105 137274321021824 utils.py:1231] [10200] examples_seen = 10444800.0
+I1128 16:20:33.395169 137274321021824 utils.py:1231] [10200] progress = 0.09058373222738293
+I1128 16:20:33.395232 137274321021824 utils.py:1231] [10200] epoch = 8.15256715166719
+I1128 16:20:33.395297 137274321021824 utils.py:1231] [10200] img/sec/core = 164.23937732050933
+I1128 16:20:33.395387 137274321021824 utils.py:1231] [10200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.472003898253885
+I1128 16:20:33.395441 137274321021824 utils.py:1231] [10200] core_hours = 18.472003898253885
+I1128 16:20:33.395505 137274321021824 train.py:125] NOTE: Steps:10200/112603 [9.1%]
+Walltime:18h30m (0s eval)
+ETA:7d17h29m
+Total train time:8d11h57m
+I1128 16:25:45.276439 137274321021824 utils.py:1231] [10250] l2_params = 268.89741108624
+I1128 16:25:45.276645 137274321021824 utils.py:1231] [10250] train/loss = 4.8295504450798035
+I1128 16:25:45.276754 137274321021824 utils.py:1231] [10250] l2_grads = 1.1639620065689087
+I1128 16:25:45.276847 137274321021824 utils.py:1231] [10250] lr = 0.0009999854683066078
+I1128 16:25:45.276920 137274321021824 utils.py:1231] [10250] uptime = 66934.63927741701
+I1128 16:25:45.276975 137274321021824 utils.py:1231] [10250] examples_seen = 10496000.0
+I1128 16:25:45.277031 137274321021824 utils.py:1231] [10250] progress = 0.09102777013045833
+I1128 16:25:45.277084 137274321021824 utils.py:1231] [10250] epoch = 8.192530716136147
+I1128 16:25:45.277151 137274321021824 utils.py:1231] [10250] img/sec/core = 164.16471769655047
+I1128 16:25:45.277220 137274321021824 utils.py:1231] [10250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.55863775248139
+I1128 16:25:45.277271 137274321021824 utils.py:1231] [10250] core_hours = 18.55863775248139
+I1128 16:25:45.277337 137274321021824 train.py:125] NOTE: Steps:10250/112603 [9.1%]
+Walltime:18h35m (0s eval)
+ETA:7d17h21m
+Total train time:8d11h55m
+I1128 16:30:57.063241 137274321021824 utils.py:1231] [10300] l2_params = 269.58129094361453
+I1128 16:30:57.063484 137274321021824 utils.py:1231] [10300] train/loss = 3.989045202732086
+I1128 16:30:57.063638 137274321021824 utils.py:1231] [10300] l2_grads = 1.2968025207519531
+I1128 16:30:57.063746 137274321021824 utils.py:1231] [10300] lr = 0.0009999790463840923
+I1128 16:30:57.063833 137274321021824 utils.py:1231] [10300] uptime = 67246.426190513
+I1128 16:30:57.063924 137274321021824 utils.py:1231] [10300] examples_seen = 10547200.0
+I1128 16:30:57.064019 137274321021824 utils.py:1231] [10300] progress = 0.09147180803353375
+I1128 16:30:57.064105 137274321021824 utils.py:1231] [10300] epoch = 8.232494280605104
+I1128 16:30:57.064189 137274321021824 utils.py:1231] [10300] img/sec/core = 164.21471796745158
+I1128 16:30:57.064280 137274321021824 utils.py:1231] [10300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.645245228341384
+I1128 16:30:57.064388 137274321021824 utils.py:1231] [10300] core_hours = 18.645245228341384
+I1128 16:30:57.064508 137274321021824 train.py:125] NOTE: Steps:10300/112603 [9.1%]
+Walltime:18h40m (0s eval)
+ETA:7d17h13m
+Total train time:8d11h52m
+I1128 16:36:08.844642 137274321021824 utils.py:1231] [10350] l2_params = 270.2761709090197
+I1128 16:36:08.844846 137274321021824 utils.py:1231] [10350] train/loss = 4.360057473182678
+I1128 16:36:08.844948 137274321021824 utils.py:1231] [10350] l2_grads = 1.1211391687393188
+I1128 16:36:08.845042 137274321021824 utils.py:1231] [10350] lr = 0.0009999714526133829
+I1128 16:36:08.845104 137274321021824 utils.py:1231] [10350] uptime = 67558.207463189
+I1128 16:36:08.845159 137274321021824 utils.py:1231] [10350] examples_seen = 10598400.0
+I1128 16:36:08.845220 137274321021824 utils.py:1231] [10350] progress = 0.09191584593660915
+I1128 16:36:08.845269 137274321021824 utils.py:1231] [10350] epoch = 8.272457845074062
+I1128 16:36:08.845332 137274321021824 utils.py:1231] [10350] img/sec/core = 164.2176887680043
+I1128 16:36:08.845405 137274321021824 utils.py:1231] [10350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.731851137418055
+I1128 16:36:08.845459 137274321021824 utils.py:1231] [10350] core_hours = 18.731851137418055
+I1128 16:36:08.845521 137274321021824 train.py:125] NOTE: Steps:10350/112603 [9.2%]
+Walltime:18h45m (0s eval)
+ETA:7d17h5m
+Total train time:8d11h49m
+I1128 16:41:20.628909 137274321021824 utils.py:1231] [10400] l2_params = 270.9590100281787
+I1128 16:41:20.629146 137274321021824 utils.py:1231] [10400] train/loss = 4.165138244628906
+I1128 16:41:20.629244 137274321021824 utils.py:1231] [10400] l2_grads = 1.221635341644287
+I1128 16:41:20.629308 137274321021824 utils.py:1231] [10400] lr = 0.0009999626870122785
+I1128 16:41:20.629362 137274321021824 utils.py:1231] [10400] uptime = 67869.991723595
+I1128 16:41:20.629416 137274321021824 utils.py:1231] [10400] examples_seen = 10649600.0
+I1128 16:41:20.629467 137274321021824 utils.py:1231] [10400] progress = 0.09235988383968456
+I1128 16:41:20.629517 137274321021824 utils.py:1231] [10400] epoch = 8.312421409543019
+I1128 16:41:20.629574 137274321021824 utils.py:1231] [10400] img/sec/core = 164.21611512181036
+I1128 16:41:20.629632 137274321021824 utils.py:1231] [10400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.818457876419718
+I1128 16:41:20.629684 137274321021824 utils.py:1231] [10400] core_hours = 18.818457876419718
+I1128 16:41:20.629745 137274321021824 train.py:125] NOTE: Steps:10400/112603 [9.2%]
+Walltime:18h51m (0s eval)
+ETA:7d16h58m
+Total train time:8d11h47m
+I1128 16:46:32.359424 137274321021824 utils.py:1231] [10450] l2_params = 271.6014054487639
+I1128 16:46:32.359605 137274321021824 utils.py:1231] [10450] train/loss = 3.9790751338005066
+I1128 16:46:32.359692 137274321021824 utils.py:1231] [10450] l2_grads = 1.1833181381225586
+I1128 16:46:32.359755 137274321021824 utils.py:1231] [10450] lr = 0.0009999527496013254
+I1128 16:46:32.359807 137274321021824 utils.py:1231] [10450] uptime = 68181.722169977
+I1128 16:46:32.359876 137274321021824 utils.py:1231] [10450] examples_seen = 10700800.0
+I1128 16:46:32.359932 137274321021824 utils.py:1231] [10450] progress = 0.09280392174275996
+I1128 16:46:32.359987 137274321021824 utils.py:1231] [10450] epoch = 8.352384974011976
+I1128 16:46:32.360037 137274321021824 utils.py:1231] [10450] img/sec/core = 164.24446374820252
+I1128 16:46:32.360091 137274321021824 utils.py:1231] [10450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.90504966708139
+I1128 16:46:32.360151 137274321021824 utils.py:1231] [10450] core_hours = 18.90504966708139
+I1128 16:46:32.360217 137274321021824 train.py:125] NOTE: Steps:10450/112603 [9.3%]
+Walltime:18h56m (0s eval)
+ETA:7d16h50m
+Total train time:8d11h44m
+I1128 16:51:44.147095 137274321021824 utils.py:1231] [10500] l2_params = 272.19037183178347
+I1128 16:51:44.147323 137274321021824 utils.py:1231] [10500] train/loss = 4.123651444911957
+I1128 16:51:44.147428 137274321021824 utils.py:1231] [10500] l2_grads = 1.6101856231689453
+I1128 16:51:44.147499 137274321021824 utils.py:1231] [10500] lr = 0.0009999416404038136
+I1128 16:51:44.147571 137274321021824 utils.py:1231] [10500] uptime = 68493.509932858
+I1128 16:51:44.147640 137274321021824 utils.py:1231] [10500] examples_seen = 10752000.0
+I1128 16:51:44.147696 137274321021824 utils.py:1231] [10500] progress = 0.09324795964583536
+I1128 16:51:44.147750 137274321021824 utils.py:1231] [10500] epoch = 8.392348538480931
+I1128 16:51:44.147806 137274321021824 utils.py:1231] [10500] img/sec/core = 164.2142703963092
+I1128 16:51:44.147871 137274321021824 utils.py:1231] [10500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 18.991657378992773
+I1128 16:51:44.147939 137274321021824 utils.py:1231] [10500] core_hours = 18.991657378992773
+I1128 16:51:44.148000 137274321021824 train.py:125] NOTE: Steps:10500/112603 [9.3%]
+Walltime:19h1m (0s eval)
+ETA:7d16h42m
+Total train time:8d11h42m
+I1128 16:56:55.937081 137274321021824 utils.py:1231] [10550] l2_params = 272.762221278444
+I1128 16:56:55.937282 137274321021824 utils.py:1231] [10550] train/loss = 4.337571799755096
+I1128 16:56:55.937397 137274321021824 utils.py:1231] [10550] l2_grads = 1.2656716108322144
+I1128 16:56:55.937491 137274321021824 utils.py:1231] [10550] lr = 0.0009999293594457805
+I1128 16:56:55.937551 137274321021824 utils.py:1231] [10550] uptime = 68805.29991353801
+I1128 16:56:55.937616 137274321021824 utils.py:1231] [10550] examples_seen = 10803200.0
+I1128 16:56:55.937683 137274321021824 utils.py:1231] [10550] progress = 0.09369199754891078
+I1128 16:56:55.937734 137274321021824 utils.py:1231] [10550] epoch = 8.432312102949888
+I1128 16:56:55.937790 137274321021824 utils.py:1231] [10550] img/sec/core = 164.21310232077715
+I1128 16:56:55.937846 137274321021824 utils.py:1231] [10550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.078265706959442
+I1128 16:56:55.937930 137274321021824 utils.py:1231] [10550] core_hours = 19.078265706959442
+I1128 16:56:55.937995 137274321021824 train.py:125] NOTE: Steps:10550/112603 [9.4%]
+Walltime:19h6m (0s eval)
+ETA:7d16h35m
+Total train time:8d11h39m
+I1128 17:02:07.735659 137274321021824 utils.py:1231] [10600] l2_params = 273.4309513648186
+I1128 17:02:07.735961 137274321021824 utils.py:1231] [10600] train/loss = 4.044397801160812
+I1128 17:02:07.736130 137274321021824 utils.py:1231] [10600] l2_grads = 1.2497975826263428
+I1128 17:02:07.736211 137274321021824 utils.py:1231] [10600] lr = 0.0009999159067560098
+I1128 17:02:07.736268 137274321021824 utils.py:1231] [10600] uptime = 69117.098629556
+I1128 17:02:07.736325 137274321021824 utils.py:1231] [10600] examples_seen = 10854400.0
+I1128 17:02:07.736375 137274321021824 utils.py:1231] [10600] progress = 0.09413603545198618
+I1128 17:02:07.736423 137274321021824 utils.py:1231] [10600] epoch = 8.472275667418845
+I1128 17:02:07.736475 137274321021824 utils.py:1231] [10600] img/sec/core = 164.20850173432316
+I1128 17:02:07.736536 137274321021824 utils.py:1231] [10600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.164876461408884
+I1128 17:02:07.736587 137274321021824 utils.py:1231] [10600] core_hours = 19.164876461408884
+I1128 17:02:07.736649 137274321021824 train.py:125] NOTE: Steps:10600/112603 [9.4%]
+Walltime:19h11m (0s eval)
+ETA:7d16h27m
+Total train time:8d11h37m
+I1128 17:07:19.561025 137274321021824 utils.py:1231] [10650] l2_params = 273.9627632874717
+I1128 17:07:19.561314 137274321021824 utils.py:1231] [10650] train/loss = 4.132404148578644
+I1128 17:07:19.561472 137274321021824 utils.py:1231] [10650] l2_grads = 1.4053300619125366
+I1128 17:07:19.561558 137274321021824 utils.py:1231] [10650] lr = 0.0009999012823660334
+I1128 17:07:19.561619 137274321021824 utils.py:1231] [10650] uptime = 69428.923980511
+I1128 17:07:19.561675 137274321021824 utils.py:1231] [10650] examples_seen = 10905600.0
+I1128 17:07:19.561723 137274321021824 utils.py:1231] [10650] progress = 0.09458007335506159
+I1128 17:07:19.561773 137274321021824 utils.py:1231] [10650] epoch = 8.512239231887802
+I1128 17:07:19.561823 137274321021824 utils.py:1231] [10650] img/sec/core = 164.1944756678527
+I1128 17:07:19.561897 137274321021824 utils.py:1231] [10650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.25149461445194
+I1128 17:07:19.561949 137274321021824 utils.py:1231] [10650] core_hours = 19.25149461445194
+I1128 17:07:19.562007 137274321021824 train.py:125] NOTE: Steps:10650/112603 [9.5%]
+Walltime:19h17m (0s eval)
+ETA:7d16h19m
+Total train time:8d11h35m
+I1128 17:12:31.340088 137274321021824 utils.py:1231] [10700] l2_params = 274.53555272391077
+I1128 17:12:31.340310 137274321021824 utils.py:1231] [10700] train/loss = 6.0131784081459045
+I1128 17:12:31.340409 137274321021824 utils.py:1231] [10700] l2_grads = 0.9972632527351379
+I1128 17:12:31.340470 137274321021824 utils.py:1231] [10700] lr = 0.0009998854863101274
+I1128 17:12:31.340526 137274321021824 utils.py:1231] [10700] uptime = 69740.702887194
+I1128 17:12:31.340585 137274321021824 utils.py:1231] [10700] examples_seen = 10956800.0
+I1128 17:12:31.340639 137274321021824 utils.py:1231] [10700] progress = 0.09502411125813699
+I1128 17:12:31.340689 137274321021824 utils.py:1231] [10700] epoch = 8.55220279635676
+I1128 17:12:31.340740 137274321021824 utils.py:1231] [10700] img/sec/core = 164.21893496488923
+I1128 17:12:31.340798 137274321021824 utils.py:1231] [10700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.33809986630833
+I1128 17:12:31.340849 137274321021824 utils.py:1231] [10700] core_hours = 19.33809986630833
+I1128 17:12:31.340915 137274321021824 train.py:125] NOTE: Steps:10700/112603 [9.5%]
+Walltime:19h22m (0s eval)
+ETA:7d16h12m
+Total train time:8d11h32m
+I1128 17:17:43.129086 137274321021824 utils.py:1231] [10750] l2_params = 275.1855060702854
+I1128 17:17:43.129304 137274321021824 utils.py:1231] [10750] train/loss = 4.113736987113953
+I1128 17:17:43.129397 137274321021824 utils.py:1231] [10750] l2_grads = 1.1702749729156494
+I1128 17:17:43.129457 137274321021824 utils.py:1231] [10750] lr = 0.0009998685186253136
+I1128 17:17:43.129507 137274321021824 utils.py:1231] [10750] uptime = 70052.491868704
+I1128 17:17:43.129566 137274321021824 utils.py:1231] [10750] examples_seen = 11008000.0
+I1128 17:17:43.129612 137274321021824 utils.py:1231] [10750] progress = 0.0954681491612124
+I1128 17:17:43.129658 137274321021824 utils.py:1231] [10750] epoch = 8.592166360825717
+I1128 17:17:43.129705 137274321021824 utils.py:1231] [10750] img/sec/core = 164.21362856389734
+I1128 17:17:43.129759 137274321021824 utils.py:1231] [10750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.424707916727776
+I1128 17:17:43.129806 137274321021824 utils.py:1231] [10750] core_hours = 19.424707916727776
+I1128 17:17:43.129862 137274321021824 train.py:125] NOTE: Steps:10750/112603 [9.5%]
+Walltime:19h27m (0s eval)
+ETA:7d16h4m
+Total train time:8d11h30m
+I1128 17:22:52.803107 137274321021824 utils.py:1231] [10800] l2_params = 275.73970128132964
+I1128 17:22:52.803384 137274321021824 utils.py:1231] [10800] train/loss = 3.9318089485168457
+I1128 17:22:52.803503 137274321021824 utils.py:1231] [10800] l2_grads = 1.2117267847061157
+I1128 17:22:52.803583 137274321021824 utils.py:1231] [10800] lr = 0.000999850379351362
+I1128 17:22:52.803637 137274321021824 utils.py:1231] [10800] uptime = 70362.165998998
+I1128 17:22:52.803701 137274321021824 utils.py:1231] [10800] examples_seen = 11059200.0
+I1128 17:22:52.803750 137274321021824 utils.py:1231] [10800] progress = 0.09591218706428781
+I1128 17:22:52.803797 137274321021824 utils.py:1231] [10800] epoch = 8.632129925294672
+I1128 17:22:52.803846 137274321021824 utils.py:1231] [10800] img/sec/core = 165.33508934502487
+I1128 17:22:52.803906 137274321021824 utils.py:1231] [10800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.51072850847611
+I1128 17:22:52.803956 137274321021824 utils.py:1231] [10800] core_hours = 19.51072850847611
+I1128 17:22:52.804015 137274321021824 train.py:125] NOTE: Steps:10800/112603 [9.6%]
+Walltime:19h32m (0s eval)
+ETA:7d15h56m
+Total train time:8d11h27m
+I1128 17:28:03.061999 137274321021824 utils.py:1231] [10850] l2_params = 276.31102464851944
+I1128 17:28:03.062208 137274321021824 utils.py:1231] [10850] train/loss = 3.966874659061432
+I1128 17:28:03.062314 137274321021824 utils.py:1231] [10850] l2_grads = 1.2472760677337646
+I1128 17:28:03.062389 137274321021824 utils.py:1231] [10850] lr = 0.000999831068530787
+I1128 17:28:03.062460 137274321021824 utils.py:1231] [10850] uptime = 70672.424821198
+I1128 17:28:03.062524 137274321021824 utils.py:1231] [10850] examples_seen = 11110400.0
+I1128 17:28:03.062585 137274321021824 utils.py:1231] [10850] progress = 0.09635622496736321
+I1128 17:28:03.062645 137274321021824 utils.py:1231] [10850] epoch = 8.672093489763629
+I1128 17:28:03.062710 137274321021824 utils.py:1231] [10850] img/sec/core = 165.02351049020126
+I1128 17:28:03.062775 137274321021824 utils.py:1231] [10850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.596911514642777
+I1128 17:28:03.062834 137274321021824 utils.py:1231] [10850] core_hours = 19.596911514642777
+I1128 17:28:03.062915 137274321021824 train.py:125] NOTE: Steps:10850/112603 [9.6%]
+Walltime:19h37m (0s eval)
+ETA:7d15h49m
+Total train time:8d11h25m
+I1128 17:33:13.028096 137274321021824 utils.py:1231] [10900] l2_params = 276.9518099893522
+I1128 17:33:13.028319 137274321021824 utils.py:1231] [10900] train/loss = 4.2965075969696045
+I1128 17:33:13.028412 137274321021824 utils.py:1231] [10900] l2_grads = 1.2322537899017334
+I1128 17:33:13.028471 137274321021824 utils.py:1231] [10900] lr = 0.0009998105862088493
+I1128 17:33:13.028520 137274321021824 utils.py:1231] [10900] uptime = 70982.390882708
+I1128 17:33:13.028572 137274321021824 utils.py:1231] [10900] examples_seen = 11161600.0
+I1128 17:33:13.028619 137274321021824 utils.py:1231] [10900] progress = 0.09680026287043862
+I1128 17:33:13.028664 137274321021824 utils.py:1231] [10900] epoch = 8.712057054232586
+I1128 17:33:13.028711 137274321021824 utils.py:1231] [10900] img/sec/core = 165.1793739952662
+I1128 17:33:13.028765 137274321021824 utils.py:1231] [10900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.683013198395553
+I1128 17:33:13.028812 137274321021824 utils.py:1231] [10900] core_hours = 19.683013198395553
+I1128 17:33:13.028872 137274321021824 train.py:125] NOTE: Steps:10900/112603 [9.7%]
+Walltime:19h43m (0s eval)
+ETA:7d15h41m
+Total train time:8d11h22m
+I1128 17:38:22.389519 137274321021824 utils.py:1231] [10950] l2_params = 277.5045733654653
+I1128 17:38:22.389756 137274321021824 utils.py:1231] [10950] train/loss = 5.391323208808899
+I1128 17:38:22.389878 137274321021824 utils.py:1231] [10950] l2_grads = 1.2302606105804443
+I1128 17:38:22.389974 137274321021824 utils.py:1231] [10950] lr = 0.0009997889324335543
+I1128 17:38:22.390023 137274321021824 utils.py:1231] [10950] uptime = 71291.75238527701
+I1128 17:38:22.390087 137274321021824 utils.py:1231] [10950] examples_seen = 11212800.0
+I1128 17:38:22.390146 137274321021824 utils.py:1231] [10950] progress = 0.09724430077351402
+I1128 17:38:22.390192 137274321021824 utils.py:1231] [10950] epoch = 8.752020618701543
+I1128 17:38:22.390257 137274321021824 utils.py:1231] [10950] img/sec/core = 165.50217003351574
+I1128 17:38:22.390309 137274321021824 utils.py:1231] [10950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.768946949109168
+I1128 17:38:22.390356 137274321021824 utils.py:1231] [10950] core_hours = 19.768946949109168
+I1128 17:38:22.390412 137274321021824 train.py:125] NOTE: Steps:10950/112603 [9.7%]
+Walltime:19h48m (0s eval)
+ETA:7d15h33m
+Total train time:8d11h19m
+I1128 17:43:34.171437 137274321021824 utils.py:1231] [11000] l2_params = 278.0069409480174
+I1128 17:43:34.171653 137274321021824 utils.py:1231] [11000] train/loss = 6.038419842720032
+I1128 17:43:34.171753 137274321021824 utils.py:1231] [11000] l2_grads = 0.933613121509552
+I1128 17:43:34.171852 137274321021824 utils.py:1231] [11000] lr = 0.0009997661072556545
+I1128 17:43:34.171922 137274321021824 utils.py:1231] [11000] uptime = 71603.534283466
+I1128 17:43:34.171986 137274321021824 utils.py:1231] [11000] examples_seen = 11264000.0
+I1128 17:43:34.172041 137274321021824 utils.py:1231] [11000] progress = 0.09768833867658944
+I1128 17:43:34.172096 137274321021824 utils.py:1231] [11000] epoch = 8.7919841831705
+I1128 17:43:34.172152 137274321021824 utils.py:1231] [11000] img/sec/core = 164.21735930597575
+I1128 17:43:34.172212 137274321021824 utils.py:1231] [11000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.85555303193944
+I1128 17:43:34.172267 137274321021824 utils.py:1231] [11000] core_hours = 19.85555303193944
+I1128 17:43:34.172331 137274321021824 train.py:125] NOTE: Steps:11000/112603 [9.8%]
+Walltime:19h53m (0s eval)
+ETA:7d15h25m
+Total train time:8d11h17m
+I1128 17:48:44.047290 137274321021824 utils.py:1231] [11050] l2_params = 278.5573671940378
+I1128 17:48:44.047503 137274321021824 utils.py:1231] [11050] train/loss = 5.0834513902664185
+I1128 17:48:44.047605 137274321021824 utils.py:1231] [11050] l2_grads = 1.0803614854812622
+I1128 17:48:44.047700 137274321021824 utils.py:1231] [11050] lr = 0.000999742110728648
+I1128 17:48:44.047759 137274321021824 utils.py:1231] [11050] uptime = 71913.410120817
+I1128 17:48:44.047817 137274321021824 utils.py:1231] [11050] examples_seen = 11315200.0
+I1128 17:48:44.047886 137274321021824 utils.py:1231] [11050] progress = 0.09813237657966484
+I1128 17:48:44.047951 137274321021824 utils.py:1231] [11050] epoch = 8.831947747639457
+I1128 17:48:44.048005 137274321021824 utils.py:1231] [11050] img/sec/core = 165.22746800036595
+I1128 17:48:44.048065 137274321021824 utils.py:1231] [11050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 19.94162965342583
+I1128 17:48:44.048119 137274321021824 utils.py:1231] [11050] core_hours = 19.94162965342583
+I1128 17:48:44.048182 137274321021824 train.py:125] NOTE: Steps:11050/112603 [9.8%]
+Walltime:19h58m (0s eval)
+ETA:7d15h18m
+Total train time:8d11h14m
+I1128 17:53:54.559378 137274321021824 utils.py:1231] [11100] l2_params = 279.0722083118141
+I1128 17:53:54.559619 137274321021824 utils.py:1231] [11100] train/loss = 6.049144506454468
+I1128 17:53:54.559743 137274321021824 utils.py:1231] [11100] l2_grads = 0.8779373168945312
+I1128 17:53:54.559831 137274321021824 utils.py:1231] [11100] lr = 0.0009997169429087768
+I1128 17:53:54.559901 137274321021824 utils.py:1231] [11100] uptime = 72223.922262442
+I1128 17:53:54.559965 137274321021824 utils.py:1231] [11100] examples_seen = 11366400.0
+I1128 17:53:54.560022 137274321021824 utils.py:1231] [11100] progress = 0.09857641448274025
+I1128 17:53:54.560075 137274321021824 utils.py:1231] [11100] epoch = 8.871911312108413
+I1128 17:53:54.560144 137274321021824 utils.py:1231] [11100] img/sec/core = 164.88888238654664
+I1128 17:53:54.560202 137274321021824 utils.py:1231] [11100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.027883026099445
+I1128 17:53:54.560255 137274321021824 utils.py:1231] [11100] core_hours = 20.027883026099445
+I1128 17:53:54.560317 137274321021824 train.py:125] NOTE: Steps:11100/112603 [9.9%]
+Walltime:20h3m (0s eval)
+ETA:7d15h10m
+Total train time:8d11h12m
+I1128 17:59:03.494434 137274321021824 utils.py:1231] [11150] l2_params = 279.69462301241157
+I1128 17:59:03.494678 137274321021824 utils.py:1231] [11150] train/loss = 3.9997617304325104
+I1128 17:59:03.494778 137274321021824 utils.py:1231] [11150] l2_grads = 1.2958202362060547
+I1128 17:59:03.494848 137274321021824 utils.py:1231] [11150] lr = 0.0009996906038550288
+I1128 17:59:03.494925 137274321021824 utils.py:1231] [11150] uptime = 72532.857286825
+I1128 17:59:03.494986 137274321021824 utils.py:1231] [11150] examples_seen = 11417600.0
+I1128 17:59:03.495050 137274321021824 utils.py:1231] [11150] progress = 0.09902045238581565
+I1128 17:59:03.495109 137274321021824 utils.py:1231] [11150] epoch = 8.91187487657737
+I1128 17:59:03.495164 137274321021824 utils.py:1231] [11150] img/sec/core = 165.73064223538978
+I1128 17:59:03.495235 137274321021824 utils.py:1231] [11150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.113698310650275
+I1128 17:59:03.495289 137274321021824 utils.py:1231] [11150] core_hours = 20.113698310650275
+I1128 17:59:03.495353 137274321021824 train.py:125] NOTE: Steps:11150/112603 [9.9%]
+Walltime:20h8m (0s eval)
+ETA:7d15h2m
+Total train time:8d11h9m
+I1128 18:04:15.286941 137274321021824 utils.py:1231] [11200] l2_params = 280.2168282191371
+I1128 18:04:15.287231 137274321021824 utils.py:1231] [11200] train/loss = 6.079107165336609
+I1128 18:04:15.287372 137274321021824 utils.py:1231] [11200] l2_grads = 0.9052096605300903
+I1128 18:04:15.287440 137274321021824 utils.py:1231] [11200] lr = 0.0009996630936291385
+I1128 18:04:15.287497 137274321021824 utils.py:1231] [11200] uptime = 72844.649858419
+I1128 18:04:15.287554 137274321021824 utils.py:1231] [11200] examples_seen = 11468800.0
+I1128 18:04:15.287608 137274321021824 utils.py:1231] [11200] progress = 0.09946449028889107
+I1128 18:04:15.287661 137274321021824 utils.py:1231] [11200] epoch = 8.951838441046327
+I1128 18:04:15.287721 137274321021824 utils.py:1231] [11200] img/sec/core = 164.21173775323663
+I1128 18:04:15.287783 137274321021824 utils.py:1231] [11200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.200307358315275
+I1128 18:04:15.287835 137274321021824 utils.py:1231] [11200] core_hours = 20.200307358315275
+I1128 18:04:15.287903 137274321021824 train.py:125] NOTE: Steps:11200/112603 [9.9%]
+Walltime:20h14m (0s eval)
+ETA:7d14h55m
+Total train time:8d11h7m
+I1128 18:09:24.083657 137274321021824 utils.py:1231] [11250] l2_params = 280.7236061437889
+I1128 18:09:24.083850 137274321021824 utils.py:1231] [11250] train/loss = 5.736836016178131
+I1128 18:09:24.083969 137274321021824 utils.py:1231] [11250] l2_grads = 0.9439842104911804
+I1128 18:09:24.084033 137274321021824 utils.py:1231] [11250] lr = 0.0009996344122955843
+I1128 18:09:24.084095 137274321021824 utils.py:1231] [11250] uptime = 73153.44645670701
+I1128 18:09:24.084161 137274321021824 utils.py:1231] [11250] examples_seen = 11520000.0
+I1128 18:09:24.084213 137274321021824 utils.py:1231] [11250] progress = 0.09990852819196647
+I1128 18:09:24.084278 137274321021824 utils.py:1231] [11250] epoch = 8.991802005515284
+I1128 18:09:24.084334 137274321021824 utils.py:1231] [11250] img/sec/core = 165.80493529998714
+I1128 18:09:24.084394 137274321021824 utils.py:1231] [11250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.286084191173057
+I1128 18:09:24.084443 137274321021824 utils.py:1231] [11250] core_hours = 20.286084191173057
+I1128 18:09:24.084501 137274321021824 train.py:125] NOTE: Steps:11250/112603 [10.0%]
+Walltime:20h19m (0s eval)
+ETA:7d14h47m
+Total train time:8d11h4m
+I1128 18:14:35.851145 137274321021824 utils.py:1231] [11300] l2_params = 281.33474886812803
+I1128 18:14:35.851387 137274321021824 utils.py:1231] [11300] train/loss = 5.275265753269196
+I1128 18:14:35.851563 137274321021824 utils.py:1231] [11300] l2_grads = 0.8990209102630615
+I1128 18:14:35.851683 137274321021824 utils.py:1231] [11300] lr = 0.0009996045599215883
+I1128 18:14:35.851785 137274321021824 utils.py:1231] [11300] uptime = 73465.214141034
+I1128 18:14:35.851886 137274321021824 utils.py:1231] [11300] examples_seen = 11571200.0
+I1128 18:14:35.852017 137274321021824 utils.py:1231] [11300] progress = 0.10035256609504187
+I1128 18:14:35.852127 137274321021824 utils.py:1231] [11300] epoch = 9.031765569984241
+I1128 18:14:35.852247 137274321021824 utils.py:1231] [11300] img/sec/core = 164.2248461719995
+I1128 18:14:35.852344 137274321021824 utils.py:1231] [11300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.37268632570833
+I1128 18:14:35.852437 137274321021824 utils.py:1231] [11300] core_hours = 20.37268632570833
+I1128 18:14:35.852549 137274321021824 train.py:125] NOTE: Steps:11300/112603 [10.0%]
+Walltime:20h24m (0s eval)
+ETA:7d14h40m
+Total train time:8d11h2m
+I1128 18:19:44.931897 137274321021824 utils.py:1231] [11350] l2_params = 281.86558133648373
+I1128 18:19:44.932164 137274321021824 utils.py:1231] [11350] train/loss = 4.268116235733032
+I1128 18:19:44.932294 137274321021824 utils.py:1231] [11350] l2_grads = 1.0767079591751099
+I1128 18:19:44.932380 137274321021824 utils.py:1231] [11350] lr = 0.0009995735365771188
+I1128 18:19:44.932454 137274321021824 utils.py:1231] [11350] uptime = 73774.29481584301
+I1128 18:19:44.932519 137274321021824 utils.py:1231] [11350] examples_seen = 11622400.0
+I1128 18:19:44.932585 137274321021824 utils.py:1231] [11350] progress = 0.10079660399811728
+I1128 18:19:44.932643 137274321021824 utils.py:1231] [11350] epoch = 9.071729134453198
+I1128 18:19:44.932700 137274321021824 utils.py:1231] [11350] img/sec/core = 165.6525437303314
+I1128 18:19:44.932760 137274321021824 utils.py:1231] [11350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.45854206871083
+I1128 18:19:44.932813 137274321021824 utils.py:1231] [11350] core_hours = 20.45854206871083
+I1128 18:19:44.932890 137274321021824 train.py:125] NOTE: Steps:11350/112603 [10.1%]
+Walltime:20h29m (0s eval)
+ETA:7d14h32m
+Total train time:8d11h0m
+I1128 18:24:54.254965 137274321021824 utils.py:1231] [11400] l2_params = 282.44080116038856
+I1128 18:24:54.255167 137274321021824 utils.py:1231] [11400] train/loss = 3.760423332452774
+I1128 18:24:54.255271 137274321021824 utils.py:1231] [11400] l2_grads = 1.2746119499206543
+I1128 18:24:54.255339 137274321021824 utils.py:1231] [11400] lr = 0.0009995413423348884
+I1128 18:24:54.255396 137274321021824 utils.py:1231] [11400] uptime = 74083.617757436
+I1128 18:24:54.255454 137274321021824 utils.py:1231] [11400] examples_seen = 11673600.0
+I1128 18:24:54.255508 137274321021824 utils.py:1231] [11400] progress = 0.10124064190119268
+I1128 18:24:54.255562 137274321021824 utils.py:1231] [11400] epoch = 9.111692698922155
+I1128 18:24:54.255616 137274321021824 utils.py:1231] [11400] img/sec/core = 165.522801950359
+I1128 18:24:54.255676 137274321021824 utils.py:1231] [11400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.54446510804222
+I1128 18:24:54.255729 137274321021824 utils.py:1231] [11400] core_hours = 20.54446510804222
+I1128 18:24:54.255793 137274321021824 train.py:125] NOTE: Steps:11400/112603 [10.1%]
+Walltime:20h34m (0s eval)
+ETA:7d14h24m
+Total train time:8d10h57m
+I1128 18:30:02.462551 137274321021824 utils.py:1231] [11450] l2_params = 283.0241695811256
+I1128 18:30:02.462785 137274321021824 utils.py:1231] [11450] train/loss = 3.7115346789360046
+I1128 18:30:02.462887 137274321021824 utils.py:1231] [11450] l2_grads = 1.216691255569458
+I1128 18:30:02.462957 137274321021824 utils.py:1231] [11450] lr = 0.0009995079772703531
+I1128 18:30:02.463017 137274321021824 utils.py:1231] [11450] uptime = 74391.825378614
+I1128 18:30:02.463078 137274321021824 utils.py:1231] [11450] examples_seen = 11724800.0
+I1128 18:30:02.463134 137274321021824 utils.py:1231] [11450] progress = 0.1016846798042681
+I1128 18:30:02.463190 137274321021824 utils.py:1231] [11450] epoch = 9.15165626339111
+I1128 18:30:02.463246 137274321021824 utils.py:1231] [11450] img/sec/core = 166.1217844137328
+I1128 18:30:02.463306 137274321021824 utils.py:1231] [11450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.630078336147218
+I1128 18:30:02.463360 137274321021824 utils.py:1231] [11450] core_hours = 20.630078336147218
+I1128 18:30:02.463426 137274321021824 train.py:125] NOTE: Steps:11450/112603 [10.2%]
+Walltime:20h39m (0s eval)
+ETA:7d14h17m
+Total train time:8d10h55m
+I1128 18:35:14.227230 137274321021824 utils.py:1231] [11500] l2_params = 283.5000316078052
+I1128 18:35:14.227502 137274321021824 utils.py:1231] [11500] train/loss = 3.8724918961524963
+I1128 18:35:14.227677 137274321021824 utils.py:1231] [11500] l2_grads = 1.334179401397705
+I1128 18:35:14.227778 137274321021824 utils.py:1231] [11500] lr = 0.000999473441461715
+I1128 18:35:14.227863 137274321021824 utils.py:1231] [11500] uptime = 74703.590219385
+I1128 18:35:14.227940 137274321021824 utils.py:1231] [11500] examples_seen = 11776000.0
+I1128 18:35:14.228004 137274321021824 utils.py:1231] [11500] progress = 0.1021287177073435
+I1128 18:35:14.228067 137274321021824 utils.py:1231] [11500] epoch = 9.191619827860068
+I1128 18:35:14.228127 137274321021824 utils.py:1231] [11500] img/sec/core = 164.22634403988673
+I1128 18:35:14.228193 137274321021824 utils.py:1231] [11500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.716679680805832
+I1128 18:35:14.228253 137274321021824 utils.py:1231] [11500] core_hours = 20.716679680805832
+I1128 18:35:14.228327 137274321021824 train.py:125] NOTE: Steps:11500/112603 [10.2%]
+Walltime:20h45m (0s eval)
+ETA:7d14h9m
+Total train time:8d10h53m
+I1128 18:40:25.997552 137274321021824 utils.py:1231] [11550] l2_params = 283.9685918001637
+I1128 18:40:25.997842 137274321021824 utils.py:1231] [11550] train/loss = 3.8658855259418488
+I1128 18:40:25.998098 137274321021824 utils.py:1231] [11550] l2_grads = 1.342137098312378
+I1128 18:40:25.998200 137274321021824 utils.py:1231] [11550] lr = 0.0009994377349899174
+I1128 18:40:25.998270 137274321021824 utils.py:1231] [11550] uptime = 75015.360631277
+I1128 18:40:25.998331 137274321021824 utils.py:1231] [11550] examples_seen = 11827200.0
+I1128 18:40:25.998389 137274321021824 utils.py:1231] [11550] progress = 0.10257275561041891
+I1128 18:40:25.998445 137274321021824 utils.py:1231] [11550] epoch = 9.231583392329025
+I1128 18:40:25.998507 137274321021824 utils.py:1231] [11550] img/sec/core = 164.22340942904216
+I1128 18:40:25.998570 137274321021824 utils.py:1231] [11550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.803282572998054
+I1128 18:40:25.998626 137274321021824 utils.py:1231] [11550] core_hours = 20.803282572998054
+I1128 18:40:25.998692 137274321021824 train.py:125] NOTE: Steps:11550/112603 [10.3%]
+Walltime:20h50m (0s eval)
+ETA:7d14h2m
+Total train time:8d10h51m
+I1128 18:45:37.761101 137274321021824 utils.py:1231] [11600] l2_params = 284.53545932306133
+I1128 18:45:37.761300 137274321021824 utils.py:1231] [11600] train/loss = 3.911993622779846
+I1128 18:45:37.761410 137274321021824 utils.py:1231] [11600] l2_grads = 1.3738867044448853
+I1128 18:45:37.761493 137274321021824 utils.py:1231] [11600] lr = 0.0009994008579386508
+I1128 18:45:37.761555 137274321021824 utils.py:1231] [11600] uptime = 75327.123916808
+I1128 18:45:37.761615 137274321021824 utils.py:1231] [11600] examples_seen = 11878400.0
+I1128 18:45:37.761718 137274321021824 utils.py:1231] [11600] progress = 0.10301679351349431
+I1128 18:45:37.761807 137274321021824 utils.py:1231] [11600] epoch = 9.271546956797982
+I1128 18:45:37.761877 137274321021824 utils.py:1231] [11600] img/sec/core = 164.22716328766595
+I1128 18:45:37.761950 137274321021824 utils.py:1231] [11600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.889883485645555
+I1128 18:45:37.762019 137274321021824 utils.py:1231] [11600] core_hours = 20.889883485645555
+I1128 18:45:37.762087 137274321021824 train.py:125] NOTE: Steps:11600/112603 [10.3%]
+Walltime:20h55m (0s eval)
+ETA:7d13h55m
+Total train time:8d10h48m
+I1128 18:50:49.533493 137274321021824 utils.py:1231] [11650] l2_params = 285.08001887219535
+I1128 18:50:49.533717 137274321021824 utils.py:1231] [11650] train/loss = 5.2059630155563354
+I1128 18:50:49.533811 137274321021824 utils.py:1231] [11650] l2_grads = 0.9523011445999146
+I1128 18:50:49.533884 137274321021824 utils.py:1231] [11650] lr = 0.000999362810394347
+I1128 18:50:49.533939 137274321021824 utils.py:1231] [11650] uptime = 75638.89630040301
+I1128 18:50:49.533990 137274321021824 utils.py:1231] [11650] examples_seen = 11929600.0
+I1128 18:50:49.534038 137274321021824 utils.py:1231] [11650] progress = 0.10346083141656971
+I1128 18:50:49.534084 137274321021824 utils.py:1231] [11650] epoch = 9.31151052126694
+I1128 18:50:49.534133 137274321021824 utils.py:1231] [11650] img/sec/core = 164.22237085151758
+I1128 18:50:49.534192 137274321021824 utils.py:1231] [11650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 20.976486925533052
+I1128 18:50:49.534240 137274321021824 utils.py:1231] [11650] core_hours = 20.976486925533052
+I1128 18:50:49.534299 137274321021824 train.py:125] NOTE: Steps:11650/112603 [10.3%]
+Walltime:21h0m (0s eval)
+ETA:7d13h48m
+Total train time:8d10h46m
+I1128 18:56:01.301044 137274321021824 utils.py:1231] [11700] l2_params = 285.6367432992711
+I1128 18:56:01.301319 137274321021824 utils.py:1231] [11700] train/loss = 3.9150369465351105
+I1128 18:56:01.301483 137274321021824 utils.py:1231] [11700] l2_grads = 1.3338162899017334
+I1128 18:56:01.301544 137274321021824 utils.py:1231] [11700] lr = 0.0009993235924461805
+I1128 18:56:01.301610 137274321021824 utils.py:1231] [11700] uptime = 75950.663971937
+I1128 18:56:01.301673 137274321021824 utils.py:1231] [11700] examples_seen = 11980800.0
+I1128 18:56:01.301722 137274321021824 utils.py:1231] [11700] progress = 0.10390486931964513
+I1128 18:56:01.301801 137274321021824 utils.py:1231] [11700] epoch = 9.351474085735896
+I1128 18:56:01.301857 137274321021824 utils.py:1231] [11700] img/sec/core = 164.2248529107605
+I1128 18:56:01.301918 137274321021824 utils.py:1231] [11700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.06308905651472
+I1128 18:56:01.301969 137274321021824 utils.py:1231] [11700] core_hours = 21.06308905651472
+I1128 18:56:01.302028 137274321021824 train.py:125] NOTE: Steps:11700/112603 [10.4%]
+Walltime:21h5m (0s eval)
+ETA:7d13h40m
+Total train time:8d10h44m
+I1128 19:01:13.102929 137274321021824 utils.py:1231] [11750] l2_params = 286.1334637017765
+I1128 19:01:13.103192 137274321021824 utils.py:1231] [11750] train/loss = 3.902525544166565
+I1128 19:01:13.103338 137274321021824 utils.py:1231] [11750] l2_grads = 1.2395896911621094
+I1128 19:01:13.103426 137274321021824 utils.py:1231] [11750] lr = 0.0009992832041860712
+I1128 19:01:13.103505 137274321021824 utils.py:1231] [11750] uptime = 76262.46586033801
+I1128 19:01:13.103600 137274321021824 utils.py:1231] [11750] examples_seen = 12032000.0
+I1128 19:01:13.103677 137274321021824 utils.py:1231] [11750] progress = 0.10434890722272053
+I1128 19:01:13.103755 137274321021824 utils.py:1231] [11750] epoch = 9.391437650204852
+I1128 19:01:13.103824 137274321021824 utils.py:1231] [11750] img/sec/core = 164.20683101877688
+I1128 19:01:13.103929 137274321021824 utils.py:1231] [11750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.149700692181668
+I1128 19:01:13.103996 137274321021824 utils.py:1231] [11750] core_hours = 21.149700692181668
+I1128 19:01:13.104059 137274321021824 train.py:125] NOTE: Steps:11750/112603 [10.4%]
+Walltime:21h11m (0s eval)
+ETA:7d13h33m
+Total train time:8d10h43m
+I1128 19:06:25.019716 137274321021824 utils.py:1231] [11800] l2_params = 286.5315783865219
+I1128 19:06:25.020019 137274321021824 utils.py:1231] [11800] train/loss = 3.777171105146408
+I1128 19:06:25.020194 137274321021824 utils.py:1231] [11800] l2_grads = 1.3001203536987305
+I1128 19:06:25.020289 137274321021824 utils.py:1231] [11800] lr = 0.0009992416457086796
+I1128 19:06:25.020358 137274321021824 utils.py:1231] [11800] uptime = 76574.382715807
+I1128 19:06:25.020430 137274321021824 utils.py:1231] [11800] examples_seen = 12083200.0
+I1128 19:06:25.020490 137274321021824 utils.py:1231] [11800] progress = 0.10479294512579594
+I1128 19:06:25.020553 137274321021824 utils.py:1231] [11800] epoch = 9.431401214673809
+I1128 19:06:25.020613 137274321021824 utils.py:1231] [11800] img/sec/core = 164.14630726837737
+I1128 19:06:25.020688 137274321021824 utils.py:1231] [11800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.236344263145277
+I1128 19:06:25.020753 137274321021824 utils.py:1231] [11800] core_hours = 21.236344263145277
+I1128 19:06:25.020818 137274321021824 train.py:125] NOTE: Steps:11800/112603 [10.5%]
+Walltime:21h16m (0s eval)
+ETA:7d13h26m
+Total train time:8d10h41m
+I1128 19:11:36.840713 137274321021824 utils.py:1231] [11850] l2_params = 287.01393128166916
+I1128 19:11:36.841010 137274321021824 utils.py:1231] [11850] train/loss = 4.292244136333466
+I1128 19:11:36.841245 137274321021824 utils.py:1231] [11850] l2_grads = 1.1017050743103027
+I1128 19:11:36.841360 137274321021824 utils.py:1231] [11850] lr = 0.000999198917111411
+I1128 19:11:36.841452 137274321021824 utils.py:1231] [11850] uptime = 76886.203800419
+I1128 19:11:36.841544 137274321021824 utils.py:1231] [11850] examples_seen = 12134400.0
+I1128 19:11:36.841609 137274321021824 utils.py:1231] [11850] progress = 0.10523698302887134
+I1128 19:11:36.841686 137274321021824 utils.py:1231] [11850] epoch = 9.471364779142766
+I1128 19:11:36.841753 137274321021824 utils.py:1231] [11850] img/sec/core = 164.19672218031636
+I1128 19:11:36.841815 137274321021824 utils.py:1231] [11850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.322961231093053
+I1128 19:11:36.841900 137274321021824 utils.py:1231] [11850] core_hours = 21.322961231093053
+I1128 19:11:36.841983 137274321021824 train.py:125] NOTE: Steps:11850/112603 [10.5%]
+Walltime:21h21m (0s eval)
+ETA:7d13h19m
+Total train time:8d10h39m
+I1128 19:16:48.729559 137274321021824 utils.py:1231] [11900] l2_params = 287.544027376577
+I1128 19:16:48.729759 137274321021824 utils.py:1231] [11900] train/loss = 5.645751953125
+I1128 19:16:48.729859 137274321021824 utils.py:1231] [11900] l2_grads = 0.8808901906013489
+I1128 19:16:48.729938 137274321021824 utils.py:1231] [11900] lr = 0.0009991550184944118
+I1128 19:16:48.730009 137274321021824 utils.py:1231] [11900] uptime = 77198.092370271
+I1128 19:16:48.730066 137274321021824 utils.py:1231] [11900] examples_seen = 12185600.0
+I1128 19:16:48.730120 137274321021824 utils.py:1231] [11900] progress = 0.10568102093194676
+I1128 19:16:48.730170 137274321021824 utils.py:1231] [11900] epoch = 9.511328343611723
+I1128 19:16:48.730232 137274321021824 utils.py:1231] [11900] img/sec/core = 164.16119392992002
+I1128 19:16:48.730293 137274321021824 utils.py:1231] [11900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.409596944940834
+I1128 19:16:48.730348 137274321021824 utils.py:1231] [11900] core_hours = 21.409596944940834
+I1128 19:16:48.730413 137274321021824 train.py:125] NOTE: Steps:11900/112603 [10.6%]
+Walltime:21h26m (0s eval)
+ETA:7d13h12m
+Total train time:8d10h37m
+I1128 19:22:00.517789 137274321021824 utils.py:1231] [11950] l2_params = 287.8756417664289
+I1128 19:22:00.518037 137274321021824 utils.py:1231] [11950] train/loss = 3.8933759331703186
+I1128 19:22:00.518149 137274321021824 utils.py:1231] [11950] l2_grads = 1.2516862154006958
+I1128 19:22:00.518213 137274321021824 utils.py:1231] [11950] lr = 0.0009991099499605718
+I1128 19:22:00.518266 137274321021824 utils.py:1231] [11950] uptime = 77509.880627708
+I1128 19:22:00.518320 137274321021824 utils.py:1231] [11950] examples_seen = 12236800.0
+I1128 19:22:00.518369 137274321021824 utils.py:1231] [11950] progress = 0.10612505883502216
+I1128 19:22:00.518418 137274321021824 utils.py:1231] [11950] epoch = 9.55129190808068
+I1128 19:22:00.518469 137274321021824 utils.py:1231] [11950] img/sec/core = 164.21400992096454
+I1128 19:22:00.518526 137274321021824 utils.py:1231] [11950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.496204794228888
+I1128 19:22:00.518590 137274321021824 utils.py:1231] [11950] core_hours = 21.496204794228888
+I1128 19:22:00.518650 137274321021824 train.py:125] NOTE: Steps:11950/112603 [10.6%]
+Walltime:21h31m (0s eval)
+ETA:7d13h5m
+Total train time:8d10h35m
+I1128 19:27:12.297349 137274321021824 utils.py:1231] [12000] l2_params = 288.33974774134316
+I1128 19:27:12.297578 137274321021824 utils.py:1231] [12000] train/loss = 5.771905899047852
+I1128 19:27:12.297668 137274321021824 utils.py:1231] [12000] l2_grads = 0.9230403304100037
+I1128 19:27:12.297727 137274321021824 utils.py:1231] [12000] lr = 0.0009990637116155227
+I1128 19:27:12.297775 137274321021824 utils.py:1231] [12000] uptime = 77821.660137641
+I1128 19:27:12.297826 137274321021824 utils.py:1231] [12000] examples_seen = 12288000.0
+I1128 19:27:12.297871 137274321021824 utils.py:1231] [12000] progress = 0.10656909673809756
+I1128 19:27:12.297923 137274321021824 utils.py:1231] [12000] epoch = 9.591255472549637
+I1128 19:27:12.297971 137274321021824 utils.py:1231] [12000] img/sec/core = 164.21861722408667
+I1128 19:27:12.298021 137274321021824 utils.py:1231] [12000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.58281021365472
+I1128 19:27:12.298068 137274321021824 utils.py:1231] [12000] core_hours = 21.58281021365472
+I1128 19:27:12.298125 137274321021824 train.py:125] NOTE: Steps:12000/112603 [10.7%]
+Walltime:21h37m (0s eval)
+ETA:7d12h58m
+Total train time:8d10h33m
+I1128 19:32:23.948856 137274321021824 utils.py:1231] [12050] l2_params = 288.78759555880083
+I1128 19:32:23.949075 137274321021824 utils.py:1231] [12050] train/loss = 3.8915819823741913
+I1128 19:32:23.949186 137274321021824 utils.py:1231] [12050] l2_grads = 1.081240177154541
+I1128 19:32:23.949263 137274321021824 utils.py:1231] [12050] lr = 0.000999016303567638
+I1128 19:32:23.949329 137274321021824 utils.py:1231] [12050] uptime = 78133.311689147
+I1128 19:32:23.949392 137274321021824 utils.py:1231] [12050] examples_seen = 12339200.0
+I1128 19:32:23.949456 137274321021824 utils.py:1231] [12050] progress = 0.10701313464117297
+I1128 19:32:23.949538 137274321021824 utils.py:1231] [12050] epoch = 9.631219037018592
+I1128 19:32:23.949625 137274321021824 utils.py:1231] [12050] img/sec/core = 164.28604238478812
+I1128 19:32:23.949693 137274321021824 utils.py:1231] [12050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.66938008907305
+I1128 19:32:23.949755 137274321021824 utils.py:1231] [12050] core_hours = 21.66938008907305
+I1128 19:32:23.949842 137274321021824 train.py:125] NOTE: Steps:12050/112603 [10.7%]
+Walltime:21h42m (0s eval)
+ETA:7d12h51m
+Total train time:8d10h31m
+I1128 19:37:35.728557 137274321021824 utils.py:1231] [12100] l2_params = 289.2632692877876
+I1128 19:37:35.728841 137274321021824 utils.py:1231] [12100] train/loss = 5.972821354866028
+I1128 19:37:35.729027 137274321021824 utils.py:1231] [12100] l2_grads = 0.8075236678123474
+I1128 19:37:35.729100 137274321021824 utils.py:1231] [12100] lr = 0.0009989677259280318
+I1128 19:37:35.729158 137274321021824 utils.py:1231] [12100] uptime = 78445.09151961701
+I1128 19:37:35.729217 137274321021824 utils.py:1231] [12100] examples_seen = 12390400.0
+I1128 19:37:35.729273 137274321021824 utils.py:1231] [12100] progress = 0.10745717254424837
+I1128 19:37:35.729327 137274321021824 utils.py:1231] [12100] epoch = 9.67118260148755
+I1128 19:37:35.729387 137274321021824 utils.py:1231] [12100] img/sec/core = 164.21844839294252
+I1128 19:37:35.729449 137274321021824 utils.py:1231] [12100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.755985597536945
+I1128 19:37:35.729504 137274321021824 utils.py:1231] [12100] core_hours = 21.755985597536945
+I1128 19:37:35.729571 137274321021824 train.py:125] NOTE: Steps:12100/112603 [10.7%]
+Walltime:21h47m (0s eval)
+ETA:7d12h44m
+Total train time:8d10h29m
+I1128 19:42:47.496935 137274321021824 utils.py:1231] [12150] l2_params = 289.6937599067483
+I1128 19:42:47.497168 137274321021824 utils.py:1231] [12150] train/loss = 3.8554507195949554
+I1128 19:42:47.497304 137274321021824 utils.py:1231] [12150] l2_grads = 1.376664161682129
+I1128 19:42:47.497389 137274321021824 utils.py:1231] [12150] lr = 0.0009989179788105597
+I1128 19:42:47.497460 137274321021824 utils.py:1231] [12150] uptime = 78756.85982146801
+I1128 19:42:47.497539 137274321021824 utils.py:1231] [12150] examples_seen = 12441600.0
+I1128 19:42:47.497597 137274321021824 utils.py:1231] [12150] progress = 0.10790121044732379
+I1128 19:42:47.497652 137274321021824 utils.py:1231] [12150] epoch = 9.711146165956507
+I1128 19:42:47.497709 137274321021824 utils.py:1231] [12150] img/sec/core = 164.22452088945687
+I1128 19:42:47.497770 137274321021824 utils.py:1231] [12150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.842587903606667
+I1128 19:42:47.497837 137274321021824 utils.py:1231] [12150] core_hours = 21.842587903606667
+I1128 19:42:47.497925 137274321021824 train.py:125] NOTE: Steps:12150/112603 [10.8%]
+Walltime:21h52m (0s eval)
+ETA:7d12h37m
+Total train time:8d10h27m
+I1128 19:47:59.258601 137274321021824 utils.py:1231] [12200] l2_params = 290.16691595717197
+I1128 19:47:59.258892 137274321021824 utils.py:1231] [12200] train/loss = 3.7445023357868195
+I1128 19:47:59.259054 137274321021824 utils.py:1231] [12200] l2_grads = 1.203038215637207
+I1128 19:47:59.259135 137274321021824 utils.py:1231] [12200] lr = 0.0009988670623318197
+I1128 19:47:59.259193 137274321021824 utils.py:1231] [12200] uptime = 79068.621555902
+I1128 19:47:59.259243 137274321021824 utils.py:1231] [12200] examples_seen = 12492800.0
+I1128 19:47:59.259292 137274321021824 utils.py:1231] [12200] progress = 0.10834524835039919
+I1128 19:47:59.259339 137274321021824 utils.py:1231] [12200] epoch = 9.751109730425464
+I1128 19:47:59.259387 137274321021824 utils.py:1231] [12200] img/sec/core = 164.22798036120017
+I1128 19:47:59.259441 137274321021824 utils.py:1231] [12200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 21.92918838539389
+I1128 19:47:59.259490 137274321021824 utils.py:1231] [12200] core_hours = 21.92918838539389
+I1128 19:47:59.259549 137274321021824 train.py:125] NOTE: Steps:12200/112603 [10.8%]
+Walltime:21h57m (0s eval)
+ETA:7d12h30m
+Total train time:8d10h26m
+I1128 19:53:11.023347 137274321021824 utils.py:1231] [12250] l2_params = 290.6898644308516
+I1128 19:53:11.023616 137274321021824 utils.py:1231] [12250] train/loss = 3.655456393957138
+I1128 19:53:11.023737 137274321021824 utils.py:1231] [12250] l2_grads = 1.1968812942504883
+I1128 19:53:11.023811 137274321021824 utils.py:1231] [12250] lr = 0.0009988149766111487
+I1128 19:53:11.023872 137274321021824 utils.py:1231] [12250] uptime = 79380.386234123
+I1128 19:53:11.023950 137274321021824 utils.py:1231] [12250] examples_seen = 12544000.0
+I1128 19:53:11.023998 137274321021824 utils.py:1231] [12250] progress = 0.1087892862534746
+I1128 19:53:11.024046 137274321021824 utils.py:1231] [12250] epoch = 9.79107329489442
+I1128 19:53:11.024096 137274321021824 utils.py:1231] [12250] img/sec/core = 164.2264296653492
+I1128 19:53:11.024151 137274321021824 utils.py:1231] [12250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.01578968489972
+I1128 19:53:11.024200 137274321021824 utils.py:1231] [12250] core_hours = 22.01578968489972
+I1128 19:53:11.024259 137274321021824 train.py:125] NOTE: Steps:12250/112603 [10.9%]
+Walltime:22h3m (0s eval)
+ETA:7d12h23m
+Total train time:8d10h24m
+I1128 19:58:22.797082 137274321021824 utils.py:1231] [12300] l2_params = 291.18812176387684
+I1128 19:58:22.797352 137274321021824 utils.py:1231] [12300] train/loss = 6.021803915500641
+I1128 19:58:22.797448 137274321021824 utils.py:1231] [12300] l2_grads = 1.1020065546035767
+I1128 19:58:22.797511 137274321021824 utils.py:1231] [12300] lr = 0.0009987617217706254
+I1128 19:58:22.797562 137274321021824 utils.py:1231] [12300] uptime = 79692.15992437801
+I1128 19:58:22.797614 137274321021824 utils.py:1231] [12300] examples_seen = 12595200.0
+I1128 19:58:22.797662 137274321021824 utils.py:1231] [12300] progress = 0.10923332415655
+I1128 19:58:22.797710 137274321021824 utils.py:1231] [12300] epoch = 9.831036859363378
+I1128 19:58:22.797760 137274321021824 utils.py:1231] [12300] img/sec/core = 164.22168258688507
+I1128 19:58:22.797815 137274321021824 utils.py:1231] [12300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.10239348774833
+I1128 19:58:22.797864 137274321021824 utils.py:1231] [12300] core_hours = 22.10239348774833
+I1128 19:58:22.797933 137274321021824 train.py:125] NOTE: Steps:12300/112603 [10.9%]
+Walltime:22h8m (0s eval)
+ETA:7d12h16m
+Total train time:8d10h22m
+I1128 20:03:34.573083 137274321021824 utils.py:1231] [12350] l2_params = 291.63019069054246
+I1128 20:03:34.573284 137274321021824 utils.py:1231] [12350] train/loss = 5.948419690132141
+I1128 20:03:34.573386 137274321021824 utils.py:1231] [12350] l2_grads = 0.9374334216117859
+I1128 20:03:34.573455 137274321021824 utils.py:1231] [12350] lr = 0.0009987072979350687
+I1128 20:03:34.573513 137274321021824 utils.py:1231] [12350] uptime = 80003.935874414
+I1128 20:03:34.573573 137274321021824 utils.py:1231] [12350] examples_seen = 12646400.0
+I1128 20:03:34.573628 137274321021824 utils.py:1231] [12350] progress = 0.10967736205962542
+I1128 20:03:34.573682 137274321021824 utils.py:1231] [12350] epoch = 9.871000423832333
+I1128 20:03:34.573738 137274321021824 utils.py:1231] [12350] img/sec/core = 164.2204922929063
+I1128 20:03:34.573805 137274321021824 utils.py:1231] [12350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.188997918313888
+I1128 20:03:34.573861 137274321021824 utils.py:1231] [12350] core_hours = 22.188997918313888
+I1128 20:03:34.573931 137274321021824 train.py:125] NOTE: Steps:12350/112603 [11.0%]
+Walltime:22h13m (0s eval)
+ETA:7d12h9m
+Total train time:8d10h20m
+I1128 20:08:46.346057 137274321021824 utils.py:1231] [12400] l2_params = 292.0087892374303
+I1128 20:08:46.346271 137274321021824 utils.py:1231] [12400] train/loss = 3.742753118276596
+I1128 20:08:46.346377 137274321021824 utils.py:1231] [12400] l2_grads = 1.267112374305725
+I1128 20:08:46.346444 137274321021824 utils.py:1231] [12400] lr = 0.0009986517052320355
+I1128 20:08:46.346504 137274321021824 utils.py:1231] [12400] uptime = 80315.708865741
+I1128 20:08:46.346579 137274321021824 utils.py:1231] [12400] examples_seen = 12697600.0
+I1128 20:08:46.346637 137274321021824 utils.py:1231] [12400] progress = 0.11012139996270082
+I1128 20:08:46.346698 137274321021824 utils.py:1231] [12400] epoch = 9.91096398830129
+I1128 20:08:46.346754 137274321021824 utils.py:1231] [12400] img/sec/core = 164.22205073658597
+I1128 20:08:46.346813 137274321021824 utils.py:1231] [12400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.275601527015834
+I1128 20:08:46.346868 137274321021824 utils.py:1231] [12400] core_hours = 22.275601527015834
+I1128 20:08:46.346942 137274321021824 train.py:125] NOTE: Steps:12400/112603 [11.0%]
+Walltime:22h18m (0s eval)
+ETA:7d12h2m
+Total train time:8d10h18m
+I1128 20:13:57.966075 137274321021824 utils.py:1231] [12450] l2_params = 292.41465255323914
+I1128 20:13:57.966314 137274321021824 utils.py:1231] [12450] train/loss = 3.620831787586212
+I1128 20:13:57.966430 137274321021824 utils.py:1231] [12450] l2_grads = 1.3144190311431885
+I1128 20:13:57.966510 137274321021824 utils.py:1231] [12450] lr = 0.0009985949437918255
+I1128 20:13:57.966592 137274321021824 utils.py:1231] [12450] uptime = 80627.328950799
+I1128 20:13:57.966658 137274321021824 utils.py:1231] [12450] examples_seen = 12748800.0
+I1128 20:13:57.966707 137274321021824 utils.py:1231] [12450] progress = 0.11056543786577622
+I1128 20:13:57.966755 137274321021824 utils.py:1231] [12450] epoch = 9.950927552770247
+I1128 20:13:57.966804 137274321021824 utils.py:1231] [12450] img/sec/core = 164.30263148946838
+I1128 20:13:57.966864 137274321021824 utils.py:1231] [12450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.362162661754162
+I1128 20:13:57.966940 137274321021824 utils.py:1231] [12450] core_hours = 22.362162661754162
+I1128 20:13:57.967001 137274321021824 train.py:125] NOTE: Steps:12450/112603 [11.1%]
+Walltime:22h23m (0s eval)
+ETA:7d11h55m
+Total train time:8d10h17m
+I1128 20:19:09.747977 137274321021824 utils.py:1231] [12500] l2_params = 292.813294467693
+I1128 20:19:09.748186 137274321021824 utils.py:1231] [12500] train/loss = 3.733420252799988
+I1128 20:19:09.748281 137274321021824 utils.py:1231] [12500] l2_grads = 1.1532968282699585
+I1128 20:19:09.748394 137274321021824 utils.py:1231] [12500] lr = 0.0009985370137474742
+I1128 20:19:09.748456 137274321021824 utils.py:1231] [12500] uptime = 80939.110818329
+I1128 20:19:09.748508 137274321021824 utils.py:1231] [12500] examples_seen = 12800000.0
+I1128 20:19:09.748557 137274321021824 utils.py:1231] [12500] progress = 0.11100947576885163
+I1128 20:19:09.748604 137274321021824 utils.py:1231] [12500] epoch = 9.990891117239205
+I1128 20:19:09.748653 137274321021824 utils.py:1231] [12500] img/sec/core = 164.21737545424193
+I1128 20:19:09.748708 137274321021824 utils.py:1231] [12500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.448768736068054
+I1128 20:19:09.748757 137274321021824 utils.py:1231] [12500] core_hours = 22.448768736068054
+I1128 20:19:09.748815 137274321021824 train.py:125] NOTE: Steps:12500/112603 [11.1%]
+Walltime:22h28m (0s eval)
+ETA:7d11h48m
+Total train time:8d10h15m
+I1128 20:19:09.748914 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:12500/112603 [11.1%]
+Walltime:22h28m (0s eval)
+ETA:7d11h48m
+Total train time:8d10h15m
+I1128 20:20:46.327521 137274321021824 utils.py:1231] [12500] val/acc@1 = 0.41256776147959184
+I1128 20:20:46.327795 137274321021824 utils.py:1231] [12500] val/loss = 2.708371726834044
+I1128 20:20:46.328029 137274321021824 utils.py:1231] [12500] z/secs/eval/val = 96.5790341039974
+I1128 20:20:46.328128 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 96.5790341039974
+I1128 20:25:58.088285 137274321021824 utils.py:1231] [12550] l2_params = 293.252010381658
+I1128 20:25:58.088483 137274321021824 utils.py:1231] [12550] train/loss = 3.630053848028183
+I1128 20:25:58.088591 137274321021824 utils.py:1231] [12550] l2_grads = 1.227406620979309
+I1128 20:25:58.088650 137274321021824 utils.py:1231] [12550] lr = 0.0009984779152347589
+I1128 20:25:58.088701 137274321021824 utils.py:1231] [12550] uptime = 81347.45106290201
+I1128 20:25:58.088752 137274321021824 utils.py:1231] [12550] examples_seen = 12851200.0
+I1128 20:25:58.088800 137274321021824 utils.py:1231] [12550] progress = 0.11145351367192703
+I1128 20:25:58.088851 137274321021824 utils.py:1231] [12550] epoch = 10.030854681708162
+I1128 20:25:58.088905 137274321021824 utils.py:1231] [12550] img/sec/core = 125.38563288940414
+I1128 20:25:58.088961 137274321021824 utils.py:1231] [12550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.562196581782775
+I1128 20:25:58.089010 137274321021824 utils.py:1231] [12550] core_hours = 22.562196581782775
+I1128 20:25:58.089068 137274321021824 train.py:125] NOTE: Steps:12550/112603 [11.1%]
+Walltime:22h35m (0s eval)
+ETA:7d11h54m
+Total train time:8d10h28m
+I1128 20:31:09.850614 137274321021824 utils.py:1231] [12600] l2_params = 293.65956907537
+I1128 20:31:09.850907 137274321021824 utils.py:1231] [12600] train/loss = 3.6546559929847717
+I1128 20:31:09.851011 137274321021824 utils.py:1231] [12600] l2_grads = 1.1743565797805786
+I1128 20:31:09.851072 137274321021824 utils.py:1231] [12600] lr = 0.0009984176483921936
+I1128 20:31:09.851138 137274321021824 utils.py:1231] [12600] uptime = 81659.213499342
+I1128 20:31:09.851202 137274321021824 utils.py:1231] [12600] examples_seen = 12902400.0
+I1128 20:31:09.851251 137274321021824 utils.py:1231] [12600] progress = 0.11189755157500245
+I1128 20:31:09.851299 137274321021824 utils.py:1231] [12600] epoch = 10.070818246177119
+I1128 20:31:09.851348 137274321021824 utils.py:1231] [12600] img/sec/core = 164.22761056351405
+I1128 20:31:09.851404 137274321021824 utils.py:1231] [12600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.648797258571665
+I1128 20:31:09.851454 137274321021824 utils.py:1231] [12600] core_hours = 22.648797258571665
+I1128 20:31:09.851513 137274321021824 train.py:125] NOTE: Steps:12600/112603 [11.2%]
+Walltime:22h40m (0s eval)
+ETA:7d11h47m
+Total train time:8d10h26m
+I1128 20:36:21.616026 137274321021824 utils.py:1231] [12650] l2_params = 294.17434034106213
+I1128 20:36:21.616261 137274321021824 utils.py:1231] [12650] train/loss = 3.6920407712459564
+I1128 20:36:21.616357 137274321021824 utils.py:1231] [12650] l2_grads = 1.430911898612976
+I1128 20:36:21.616418 137274321021824 utils.py:1231] [12650] lr = 0.0009983562133610326
+I1128 20:36:21.616468 137274321021824 utils.py:1231] [12650] uptime = 81970.978830306
+I1128 20:36:21.616527 137274321021824 utils.py:1231] [12650] examples_seen = 12953600.0
+I1128 20:36:21.616576 137274321021824 utils.py:1231] [12650] progress = 0.11234158947807785
+I1128 20:36:21.616624 137274321021824 utils.py:1231] [12650] epoch = 10.110781810646076
+I1128 20:36:21.616673 137274321021824 utils.py:1231] [12650] img/sec/core = 164.22608582450752
+I1128 20:36:21.616727 137274321021824 utils.py:1231] [12650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.735398739395
+I1128 20:36:21.616775 137274321021824 utils.py:1231] [12650] core_hours = 22.735398739395
+I1128 20:36:21.616834 137274321021824 train.py:125] NOTE: Steps:12650/112603 [11.2%]
+Walltime:22h46m (0s eval)
+ETA:7d11h40m
+Total train time:8d10h24m
+I1128 20:41:33.514567 137274321021824 utils.py:1231] [12700] l2_params = 294.5771944914736
+I1128 20:41:33.514804 137274321021824 utils.py:1231] [12700] train/loss = 3.62863752245903
+I1128 20:41:33.514945 137274321021824 utils.py:1231] [12700] l2_grads = 1.3353134393692017
+I1128 20:41:33.515024 137274321021824 utils.py:1231] [12700] lr = 0.0009982936102852647
+I1128 20:41:33.515108 137274321021824 utils.py:1231] [12700] uptime = 82282.877459201
+I1128 20:41:33.515184 137274321021824 utils.py:1231] [12700] examples_seen = 13004800.0
+I1128 20:41:33.515247 137274321021824 utils.py:1231] [12700] progress = 0.11278562738115326
+I1128 20:41:33.515300 137274321021824 utils.py:1231] [12700] epoch = 10.150745375115031
+I1128 20:41:33.515366 137274321021824 utils.py:1231] [12700] img/sec/core = 164.15589956708968
+I1128 20:41:33.515429 137274321021824 utils.py:1231] [12700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.822037247421388
+I1128 20:41:33.515487 137274321021824 utils.py:1231] [12700] core_hours = 22.822037247421388
+I1128 20:41:33.515562 137274321021824 train.py:125] NOTE: Steps:12700/112603 [11.3%]
+Walltime:22h51m (0s eval)
+ETA:7d11h33m
+Total train time:8d10h22m
+I1128 20:46:45.282313 137274321021824 utils.py:1231] [12750] l2_params = 294.99882868492824
+I1128 20:46:45.282534 137274321021824 utils.py:1231] [12750] train/loss = 3.6791456937789917
+I1128 20:46:45.282674 137274321021824 utils.py:1231] [12750] l2_grads = 1.1900005340576172
+I1128 20:46:45.282769 137274321021824 utils.py:1231] [12750] lr = 0.0009982298393116204
+I1128 20:46:45.282859 137274321021824 utils.py:1231] [12750] uptime = 82594.645216157
+I1128 20:46:45.282948 137274321021824 utils.py:1231] [12750] examples_seen = 13056000.0
+I1128 20:46:45.283029 137274321021824 utils.py:1231] [12750] progress = 0.11322966528422866
+I1128 20:46:45.283113 137274321021824 utils.py:1231] [12750] epoch = 10.190708939583988
+I1128 20:46:45.283201 137274321021824 utils.py:1231] [12750] img/sec/core = 164.2248079143902
+I1128 20:46:45.283318 137274321021824 utils.py:1231] [12750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.908639402131385
+I1128 20:46:45.283403 137274321021824 utils.py:1231] [12750] core_hours = 22.908639402131385
+I1128 20:46:45.283501 137274321021824 train.py:125] NOTE: Steps:12750/112603 [11.3%]
+Walltime:22h56m (0s eval)
+ETA:7d11h26m
+Total train time:8d10h21m
+I1128 20:51:57.074837 137274321021824 utils.py:1231] [12800] l2_params = 295.3286483685941
+I1128 20:51:57.075066 137274321021824 utils.py:1231] [12800] train/loss = 3.903372347354889
+I1128 20:51:57.075170 137274321021824 utils.py:1231] [12800] l2_grads = 1.4136483669281006
+I1128 20:51:57.075242 137274321021824 utils.py:1231] [12800] lr = 0.0009981649005895668
+I1128 20:51:57.075304 137274321021824 utils.py:1231] [12800] uptime = 82906.43766561401
+I1128 20:51:57.075362 137274321021824 utils.py:1231] [12800] examples_seen = 13107200.0
+I1128 20:51:57.075417 137274321021824 utils.py:1231] [12800] progress = 0.11367370318730406
+I1128 20:51:57.075472 137274321021824 utils.py:1231] [12800] epoch = 10.230672504052945
+I1128 20:51:57.075542 137274321021824 utils.py:1231] [12800] img/sec/core = 164.21180207912812
+I1128 20:51:57.075630 137274321021824 utils.py:1231] [12800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 22.995248415869444
+I1128 20:51:57.075728 137274321021824 utils.py:1231] [12800] core_hours = 22.995248415869444
+I1128 20:51:57.075809 137274321021824 train.py:125] NOTE: Steps:12800/112603 [11.4%]
+Walltime:23h1m (0s eval)
+ETA:7d11h19m
+Total train time:8d10h19m
+I1128 20:57:08.861658 137274321021824 utils.py:1231] [12850] l2_params = 295.74424932333267
+I1128 20:57:08.861896 137274321021824 utils.py:1231] [12850] train/loss = 3.8514674603939056
+I1128 20:57:08.861993 137274321021824 utils.py:1231] [12850] l2_grads = 1.2618463039398193
+I1128 20:57:08.862053 137274321021824 utils.py:1231] [12850] lr = 0.000998098794271305
+I1128 20:57:08.862104 137274321021824 utils.py:1231] [12850] uptime = 83218.224466011
+I1128 20:57:08.862156 137274321021824 utils.py:1231] [12850] examples_seen = 13158400.0
+I1128 20:57:08.862205 137274321021824 utils.py:1231] [12850] progress = 0.11411774109037948
+I1128 20:57:08.862253 137274321021824 utils.py:1231] [12850] epoch = 10.270636068521902
+I1128 20:57:08.862304 137274321021824 utils.py:1231] [12850] img/sec/core = 164.21477732478724
+I1128 20:57:08.862361 137274321021824 utils.py:1231] [12850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.081855860424167
+I1128 20:57:08.862410 137274321021824 utils.py:1231] [12850] core_hours = 23.081855860424167
+I1128 20:57:08.862471 137274321021824 train.py:125] NOTE: Steps:12850/112603 [11.4%]
+Walltime:23h6m (0s eval)
+ETA:7d11h12m
+Total train time:8d10h17m
+I1128 21:02:20.656810 137274321021824 utils.py:1231] [12900] l2_params = 296.14210398269523
+I1128 21:02:20.657077 137274321021824 utils.py:1231] [12900] train/loss = 3.6275351345539093
+I1128 21:02:20.657200 137274321021824 utils.py:1231] [12900] l2_grads = 1.2252936363220215
+I1128 21:02:20.657288 137274321021824 utils.py:1231] [12900] lr = 0.000998031520511775
+I1128 21:02:20.657339 137274321021824 utils.py:1231] [12900] uptime = 83530.019701125
+I1128 21:02:20.657392 137274321021824 utils.py:1231] [12900] examples_seen = 13209600.0
+I1128 21:02:20.657439 137274321021824 utils.py:1231] [12900] progress = 0.11456177899345488
+I1128 21:02:20.657503 137274321021824 utils.py:1231] [12900] epoch = 10.31059963299086
+I1128 21:02:20.657565 137274321021824 utils.py:1231] [12900] img/sec/core = 164.21033496961695
+I1128 21:02:20.657618 137274321021824 utils.py:1231] [12900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.16846564795583
+I1128 21:02:20.657666 137274321021824 utils.py:1231] [12900] core_hours = 23.16846564795583
+I1128 21:02:20.657725 137274321021824 train.py:125] NOTE: Steps:12900/112603 [11.5%]
+Walltime:23h12m (0s eval)
+ETA:7d11h5m
+Total train time:8d10h16m
+I1128 21:07:32.435848 137274321021824 utils.py:1231] [12950] l2_params = 296.60277865776084
+I1128 21:07:32.436102 137274321021824 utils.py:1231] [12950] train/loss = 4.363460719585419
+I1128 21:07:32.436247 137274321021824 utils.py:1231] [12950] l2_grads = 1.2346322536468506
+I1128 21:07:32.436338 137274321021824 utils.py:1231] [12950] lr = 0.0009979630794686534
+I1128 21:07:32.436410 137274321021824 utils.py:1231] [12950] uptime = 83841.798771163
+I1128 21:07:32.436481 137274321021824 utils.py:1231] [12950] examples_seen = 13260800.0
+I1128 21:07:32.436552 137274321021824 utils.py:1231] [12950] progress = 0.11500581689653029
+I1128 21:07:32.436619 137274321021824 utils.py:1231] [12950] epoch = 10.350563197459817
+I1128 21:07:32.436681 137274321021824 utils.py:1231] [12950] img/sec/core = 164.21884892324496
+I1128 21:07:32.436758 137274321021824 utils.py:1231] [12950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.255070945188606
+I1128 21:07:32.436810 137274321021824 utils.py:1231] [12950] core_hours = 23.255070945188606
+I1128 21:07:32.436875 137274321021824 train.py:125] NOTE: Steps:12950/112603 [11.5%]
+Walltime:23h17m (0s eval)
+ETA:7d10h58m
+Total train time:8d10h14m
+I1128 21:12:44.210185 137274321021824 utils.py:1231] [13000] l2_params = 296.9653255548607
+I1128 21:12:44.210453 137274321021824 utils.py:1231] [13000] train/loss = 4.991001605987549
+I1128 21:12:44.210667 137274321021824 utils.py:1231] [13000] l2_grads = 1.111797571182251
+I1128 21:12:44.210782 137274321021824 utils.py:1231] [13000] lr = 0.0009978934713023501
+I1128 21:12:44.210850 137274321021824 utils.py:1231] [13000] uptime = 84153.57321162701
+I1128 21:12:44.210909 137274321021824 utils.py:1231] [13000] examples_seen = 13312000.0
+I1128 21:12:44.210965 137274321021824 utils.py:1231] [13000] progress = 0.11544985479960569
+I1128 21:12:44.211023 137274321021824 utils.py:1231] [13000] epoch = 10.390526761928772
+I1128 21:12:44.211091 137274321021824 utils.py:1231] [13000] img/sec/core = 164.22128742753756
+I1128 21:12:44.211157 137274321021824 utils.py:1231] [13000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.341674956428612
+I1128 21:12:44.211209 137274321021824 utils.py:1231] [13000] core_hours = 23.341674956428612
+I1128 21:12:44.211271 137274321021824 train.py:125] NOTE: Steps:13000/112603 [11.5%]
+Walltime:23h22m (0s eval)
+ETA:7d10h51m
+Total train time:8d10h12m
+I1128 21:17:56.260770 137274321021824 utils.py:1231] [13050] l2_params = 297.2511835719376
+I1128 21:17:56.260995 137274321021824 utils.py:1231] [13050] train/loss = 4.3836864829063416
+I1128 21:17:56.261138 137274321021824 utils.py:1231] [13050] l2_grads = 1.0389577150344849
+I1128 21:17:56.261222 137274321021824 utils.py:1231] [13050] lr = 0.0009978226961760148
+I1128 21:17:56.261289 137274321021824 utils.py:1231] [13050] uptime = 84465.62365012
+I1128 21:17:56.261349 137274321021824 utils.py:1231] [13050] examples_seen = 13363200.0
+I1128 21:17:56.261404 137274321021824 utils.py:1231] [13050] progress = 0.1158938927026811
+I1128 21:17:56.261458 137274321021824 utils.py:1231] [13050] epoch = 10.430490326397729
+I1128 21:17:56.261514 137274321021824 utils.py:1231] [13050] img/sec/core = 164.07603926872636
+I1128 21:17:56.261575 137274321021824 utils.py:1231] [13050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.428355633787774
+I1128 21:17:56.261629 137274321021824 utils.py:1231] [13050] core_hours = 23.428355633787774
+I1128 21:17:56.261704 137274321021824 train.py:125] NOTE: Steps:13050/112603 [11.6%]
+Walltime:23h27m (0s eval)
+ETA:7d10h45m
+Total train time:8d10h11m
+I1128 21:23:08.047614 137274321021824 utils.py:1231] [13100] l2_params = 297.6074314743929
+I1128 21:23:08.047844 137274321021824 utils.py:1231] [13100] train/loss = 5.751530110836029
+I1128 21:23:08.047949 137274321021824 utils.py:1231] [13100] l2_grads = 0.9280180931091309
+I1128 21:23:08.048012 137274321021824 utils.py:1231] [13100] lr = 0.0009977507542555282
+I1128 21:23:08.048065 137274321021824 utils.py:1231] [13100] uptime = 84777.410427209
+I1128 21:23:08.048118 137274321021824 utils.py:1231] [13100] examples_seen = 13414400.0
+I1128 21:23:08.048169 137274321021824 utils.py:1231] [13100] progress = 0.1163379306057565
+I1128 21:23:08.048218 137274321021824 utils.py:1231] [13100] epoch = 10.470453890866686
+I1128 21:23:08.048269 137274321021824 utils.py:1231] [13100] img/sec/core = 164.21478960085844
+I1128 21:23:08.048324 137274321021824 utils.py:1231] [13100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.514963071868056
+I1128 21:23:08.048374 137274321021824 utils.py:1231] [13100] core_hours = 23.514963071868056
+I1128 21:23:08.048438 137274321021824 train.py:125] NOTE: Steps:13100/112603 [11.6%]
+Walltime:23h32m (0s eval)
+ETA:7d10h38m
+Total train time:8d10h9m
+I1128 21:28:17.754393 137274321021824 utils.py:1231] [13150] l2_params = 298.0037804255176
+I1128 21:28:17.754602 137274321021824 utils.py:1231] [13150] train/loss = 5.126711070537567
+I1128 21:28:17.754705 137274321021824 utils.py:1231] [13150] l2_grads = 0.9224568605422974
+I1128 21:28:17.754793 137274321021824 utils.py:1231] [13150] lr = 0.0009976776457095075
+I1128 21:28:17.754855 137274321021824 utils.py:1231] [13150] uptime = 85087.11721602001
+I1128 21:28:17.754921 137274321021824 utils.py:1231] [13150] examples_seen = 13465600.0
+I1128 21:28:17.754980 137274321021824 utils.py:1231] [13150] progress = 0.11678196850883191
+I1128 21:28:17.755039 137274321021824 utils.py:1231] [13150] epoch = 10.510417455335643
+I1128 21:28:17.755098 137274321021824 utils.py:1231] [13150] img/sec/core = 165.31765479394934
+I1128 21:28:17.755165 137274321021824 utils.py:1231] [13150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.600992735426665
+I1128 21:28:17.755217 137274321021824 utils.py:1231] [13150] core_hours = 23.600992735426665
+I1128 21:28:17.755282 137274321021824 train.py:125] NOTE: Steps:13150/112603 [11.7%]
+Walltime:23h38m (0s eval)
+ETA:7d10h31m
+Total train time:8d10h7m
+I1128 21:33:29.528867 137274321021824 utils.py:1231] [13200] l2_params = 298.4238998609136
+I1128 21:33:29.529083 137274321021824 utils.py:1231] [13200] train/loss = 4.0360861122608185
+I1128 21:33:29.529179 137274321021824 utils.py:1231] [13200] l2_grads = 1.2456821203231812
+I1128 21:33:29.529244 137274321021824 utils.py:1231] [13200] lr = 0.0009976033707093054
+I1128 21:33:29.529297 137274321021824 utils.py:1231] [13200] uptime = 85398.891658358
+I1128 21:33:29.529356 137274321021824 utils.py:1231] [13200] examples_seen = 13516800.0
+I1128 21:33:29.529404 137274321021824 utils.py:1231] [13200] progress = 0.11722600641190732
+I1128 21:33:29.529452 137274321021824 utils.py:1231] [13200] epoch = 10.5503810198046
+I1128 21:33:29.529502 137274321021824 utils.py:1231] [13200] img/sec/core = 164.2212864404535
+I1128 21:33:29.529560 137274321021824 utils.py:1231] [13200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.68759674718722
+I1128 21:33:29.529610 137274321021824 utils.py:1231] [13200] core_hours = 23.68759674718722
+I1128 21:33:29.529682 137274321021824 train.py:125] NOTE: Steps:13200/112603 [11.7%]
+Walltime:23h43m (0s eval)
+ETA:7d10h24m
+Total train time:8d10h5m
+I1128 21:38:41.295523 137274321021824 utils.py:1231] [13250] l2_params = 298.7547493552131
+I1128 21:38:41.295770 137274321021824 utils.py:1231] [13250] train/loss = 3.996513694524765
+I1128 21:38:41.295889 137274321021824 utils.py:1231] [13250] l2_grads = 1.2160781621932983
+I1128 21:38:41.295959 137274321021824 utils.py:1231] [13250] lr = 0.0009975279294290053
+I1128 21:38:41.296016 137274321021824 utils.py:1231] [13250] uptime = 85710.65837843
+I1128 21:38:41.296076 137274321021824 utils.py:1231] [13250] examples_seen = 13568000.0
+I1128 21:38:41.296121 137274321021824 utils.py:1231] [13250] progress = 0.11767004431498272
+I1128 21:38:41.296166 137274321021824 utils.py:1231] [13250] epoch = 10.590344584273558
+I1128 21:38:41.296213 137274321021824 utils.py:1231] [13250] img/sec/core = 164.22535409865225
+I1128 21:38:41.296266 137274321021824 utils.py:1231] [13250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.774198613873885
+I1128 21:38:41.296312 137274321021824 utils.py:1231] [13250] core_hours = 23.774198613873885
+I1128 21:38:41.296367 137274321021824 train.py:125] NOTE: Steps:13250/112603 [11.8%]
+Walltime:23h48m (0s eval)
+ETA:7d10h17m
+Total train time:8d10h4m
+I1128 21:43:53.066584 137274321021824 utils.py:1231] [13300] l2_params = 299.05536063236684
+I1128 21:43:53.066797 137274321021824 utils.py:1231] [13300] train/loss = 3.575781524181366
+I1128 21:43:53.066895 137274321021824 utils.py:1231] [13300] l2_grads = 1.2852064371109009
+I1128 21:43:53.066962 137274321021824 utils.py:1231] [13300] lr = 0.0009974513220454282
+I1128 21:43:53.067018 137274321021824 utils.py:1231] [13300] uptime = 86022.42937988
+I1128 21:43:53.067069 137274321021824 utils.py:1231] [13300] examples_seen = 13619200.0
+I1128 21:43:53.067117 137274321021824 utils.py:1231] [13300] progress = 0.11811408221805814
+I1128 21:43:53.067164 137274321021824 utils.py:1231] [13300] epoch = 10.630308148742513
+I1128 21:43:53.067213 137274321021824 utils.py:1231] [13300] img/sec/core = 164.22309888307694
+I1128 21:43:53.067269 137274321021824 utils.py:1231] [13300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.86080166983222
+I1128 21:43:53.067316 137274321021824 utils.py:1231] [13300] core_hours = 23.86080166983222
+I1128 21:43:53.067373 137274321021824 train.py:125] NOTE: Steps:13300/112603 [11.8%]
+Walltime:23h53m (0s eval)
+ETA:7d10h10m
+Total train time:8d10h2m
+I1128 21:49:02.401791 137274321021824 utils.py:1231] [13350] l2_params = 299.4261993111766
+I1128 21:49:02.402041 137274321021824 utils.py:1231] [13350] train/loss = 3.664017379283905
+I1128 21:49:02.402184 137274321021824 utils.py:1231] [13350] l2_grads = 1.3347374200820923
+I1128 21:49:02.402284 137274321021824 utils.py:1231] [13350] lr = 0.0009973735487381237
+I1128 21:49:02.402364 137274321021824 utils.py:1231] [13350] uptime = 86331.764720271
+I1128 21:49:02.402439 137274321021824 utils.py:1231] [13350] examples_seen = 13670400.0
+I1128 21:49:02.402504 137274321021824 utils.py:1231] [13350] progress = 0.11855812012113354
+I1128 21:49:02.402582 137274321021824 utils.py:1231] [13350] epoch = 10.67027171321147
+I1128 21:49:02.402663 137274321021824 utils.py:1231] [13350] img/sec/core = 165.51616745530302
+I1128 21:49:02.402750 137274321021824 utils.py:1231] [13350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 23.94672815327417
+I1128 21:49:02.402813 137274321021824 utils.py:1231] [13350] core_hours = 23.94672815327417
+I1128 21:49:02.402911 137274321021824 train.py:125] NOTE: Steps:13350/112603 [11.9%]
+Walltime:23h58m (0s eval)
+ETA:7d10h3m
+Total train time:8d10h0m
+I1128 21:54:14.177508 137274321021824 utils.py:1231] [13400] l2_params = 299.7482422694318
+I1128 21:54:14.177718 137274321021824 utils.py:1231] [13400] train/loss = 3.6997857689857483
+I1128 21:54:14.177823 137274321021824 utils.py:1231] [13400] l2_grads = 1.2741905450820923
+I1128 21:54:14.177901 137274321021824 utils.py:1231] [13400] lr = 0.00099729460968938
+I1128 21:54:14.177964 137274321021824 utils.py:1231] [13400] uptime = 86643.540325899
+I1128 21:54:14.178024 137274321021824 utils.py:1231] [13400] examples_seen = 13721600.0
+I1128 21:54:14.178080 137274321021824 utils.py:1231] [13400] progress = 0.11900215802420895
+I1128 21:54:14.178137 137274321021824 utils.py:1231] [13400] epoch = 10.710235277680427
+I1128 21:54:14.178194 137274321021824 utils.py:1231] [13400] img/sec/core = 164.22067370175094
+I1128 21:54:14.178257 137274321021824 utils.py:1231] [13400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.03333248817083
+I1128 21:54:14.178312 137274321021824 utils.py:1231] [13400] core_hours = 24.03333248817083
+I1128 21:54:14.178377 137274321021824 train.py:125] NOTE: Steps:13400/112603 [11.9%]
+Walltime:24h4m (0s eval)
+ETA:7d9h57m
+Total train time:8d9h59m
+I1128 21:59:25.969938 137274321021824 utils.py:1231] [13450] l2_params = 300.1009908412553
+I1128 21:59:25.970200 137274321021824 utils.py:1231] [13450] train/loss = 3.554682582616806
+I1128 21:59:25.970346 137274321021824 utils.py:1231] [13450] l2_grads = 1.2062686681747437
+I1128 21:59:25.970445 137274321021824 utils.py:1231] [13450] lr = 0.0009972145050842099
+I1128 21:59:25.970530 137274321021824 utils.py:1231] [13450] uptime = 86955.332889464
+I1128 21:59:25.970638 137274321021824 utils.py:1231] [13450] examples_seen = 13772800.0
+I1128 21:59:25.970747 137274321021824 utils.py:1231] [13450] progress = 0.11944619592728435
+I1128 21:59:25.970815 137274321021824 utils.py:1231] [13450] epoch = 10.750198842149384
+I1128 21:59:25.970884 137274321021824 utils.py:1231] [13450] img/sec/core = 164.21174198186355
+I1128 21:59:25.971005 137274321021824 utils.py:1231] [13450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.11994153360555
+I1128 21:59:25.971082 137274321021824 utils.py:1231] [13450] core_hours = 24.11994153360555
+I1128 21:59:25.971148 137274321021824 train.py:125] NOTE: Steps:13450/112603 [11.9%]
+Walltime:24h9m (0s eval)
+ETA:7d9h50m
+Total train time:8d9h57m
+I1128 22:04:37.743924 137274321021824 utils.py:1231] [13500] l2_params = 300.50248397809713
+I1128 22:04:37.744133 137274321021824 utils.py:1231] [13500] train/loss = 3.6410695910453796
+I1128 22:04:37.744241 137274321021824 utils.py:1231] [13500] l2_grads = 1.1738523244857788
+I1128 22:04:37.744323 137274321021824 utils.py:1231] [13500] lr = 0.0009971332351103644
+I1128 22:04:37.744405 137274321021824 utils.py:1231] [13500] uptime = 87267.10676047401
+I1128 22:04:37.744513 137274321021824 utils.py:1231] [13500] examples_seen = 13824000.0
+I1128 22:04:37.744590 137274321021824 utils.py:1231] [13500] progress = 0.11989023383035977
+I1128 22:04:37.744658 137274321021824 utils.py:1231] [13500] epoch = 10.790162406618341
+I1128 22:04:37.744762 137274321021824 utils.py:1231] [13500] img/sec/core = 164.22158737720602
+I1128 22:04:37.744850 137274321021824 utils.py:1231] [13500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.20654538666389
+I1128 22:04:37.744934 137274321021824 utils.py:1231] [13500] core_hours = 24.20654538666389
+I1128 22:04:37.745016 137274321021824 train.py:125] NOTE: Steps:13500/112603 [12.0%]
+Walltime:24h14m (0s eval)
+ETA:7d9h43m
+Total train time:8d9h56m
+I1128 22:09:49.539186 137274321021824 utils.py:1231] [13550] l2_params = 300.8505600846661
+I1128 22:09:49.539402 137274321021824 utils.py:1231] [13550] train/loss = 5.8953617811203
+I1128 22:09:49.539502 137274321021824 utils.py:1231] [13550] l2_grads = 0.8255525827407837
+I1128 22:09:49.539574 137274321021824 utils.py:1231] [13550] lr = 0.0009970507999583227
+I1128 22:09:49.539627 137274321021824 utils.py:1231] [13550] uptime = 87578.901988909
+I1128 22:09:49.539680 137274321021824 utils.py:1231] [13550] examples_seen = 13875200.0
+I1128 22:09:49.539730 137274321021824 utils.py:1231] [13550] progress = 0.12033427173343517
+I1128 22:09:49.539779 137274321021824 utils.py:1231] [13550] epoch = 10.830125971087298
+I1128 22:09:49.539829 137274321021824 utils.py:1231] [13550] img/sec/core = 164.21033848718938
+I1128 22:09:49.539889 137274321021824 utils.py:1231] [13550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.293155172340274
+I1128 22:09:49.539941 137274321021824 utils.py:1231] [13550] core_hours = 24.293155172340274
+I1128 22:09:49.540002 137274321021824 train.py:125] NOTE: Steps:13550/112603 [12.0%]
+Walltime:24h19m (0s eval)
+ETA:7d9h36m
+Total train time:8d9h54m
+I1128 22:15:00.470076 137274321021824 utils.py:1231] [13600] l2_params = 301.14973824631534
+I1128 22:15:00.470310 137274321021824 utils.py:1231] [13600] train/loss = 3.6305137276649475
+I1128 22:15:00.470417 137274321021824 utils.py:1231] [13600] l2_grads = 1.2226126194000244
+I1128 22:15:00.470494 137274321021824 utils.py:1231] [13600] lr = 0.0009969671998212965
+I1128 22:15:00.470549 137274321021824 utils.py:1231] [13600] uptime = 87889.832911174
+I1128 22:15:00.470614 137274321021824 utils.py:1231] [13600] examples_seen = 13926400.0
+I1128 22:15:00.470663 137274321021824 utils.py:1231] [13600] progress = 0.12077830963651057
+I1128 22:15:00.470711 137274321021824 utils.py:1231] [13600] epoch = 10.870089535556255
+I1128 22:15:00.470762 137274321021824 utils.py:1231] [13600] img/sec/core = 164.66680003078807
+I1128 22:15:00.470818 137274321021824 utils.py:1231] [13600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.379524872969444
+I1128 22:15:00.470880 137274321021824 utils.py:1231] [13600] core_hours = 24.379524872969444
+I1128 22:15:00.470947 137274321021824 train.py:125] NOTE: Steps:13600/112603 [12.1%]
+Walltime:24h24m (0s eval)
+ETA:7d9h30m
+Total train time:8d9h52m
+I1128 22:20:12.258512 137274321021824 utils.py:1231] [13650] l2_params = 301.4855920217445
+I1128 22:20:12.258802 137274321021824 utils.py:1231] [13650] train/loss = 3.555540829896927
+I1128 22:20:12.258988 137274321021824 utils.py:1231] [13650] l2_grads = 1.2526798248291016
+I1128 22:20:12.259062 137274321021824 utils.py:1231] [13650] lr = 0.0009968824348952274
+I1128 22:20:12.259122 137274321021824 utils.py:1231] [13650] uptime = 88201.621484232
+I1128 22:20:12.259187 137274321021824 utils.py:1231] [13650] examples_seen = 13977600.0
+I1128 22:20:12.259236 137274321021824 utils.py:1231] [13650] progress = 0.12122234753958598
+I1128 22:20:12.259283 137274321021824 utils.py:1231] [13650] epoch = 10.91005310002521
+I1128 22:20:12.259331 137274321021824 utils.py:1231] [13650] img/sec/core = 164.21384368847694
+I1128 22:20:12.259385 137274321021824 utils.py:1231] [13650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.46613280993
+I1128 22:20:12.259434 137274321021824 utils.py:1231] [13650] core_hours = 24.46613280993
+I1128 22:20:12.259492 137274321021824 train.py:125] NOTE: Steps:13650/112603 [12.1%]
+Walltime:24h30m (0s eval)
+ETA:7d9h23m
+Total train time:8d9h51m
+I1128 22:25:21.672157 137274321021824 utils.py:1231] [13700] l2_params = 301.8062499560025
+I1128 22:25:21.672364 137274321021824 utils.py:1231] [13700] train/loss = 3.872229665517807
+I1128 22:25:21.672466 137274321021824 utils.py:1231] [13700] l2_grads = 1.1720635890960693
+I1128 22:25:21.672526 137274321021824 utils.py:1231] [13700] lr = 0.0009967965053787867
+I1128 22:25:21.672581 137274321021824 utils.py:1231] [13700] uptime = 88511.034939112
+I1128 22:25:21.672636 137274321021824 utils.py:1231] [13700] examples_seen = 14028800.0
+I1128 22:25:21.672683 137274321021824 utils.py:1231] [13700] progress = 0.12166638544266138
+I1128 22:25:21.672734 137274321021824 utils.py:1231] [13700] epoch = 10.950016664494168
+I1128 22:25:21.672782 137274321021824 utils.py:1231] [13700] img/sec/core = 165.4743812606929
+I1128 22:25:21.672836 137274321021824 utils.py:1231] [13700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.552080991841105
+I1128 22:25:21.672889 137274321021824 utils.py:1231] [13700] core_hours = 24.552080991841105
+I1128 22:25:21.672948 137274321021824 train.py:125] NOTE: Steps:13700/112603 [12.2%]
+Walltime:24h35m (0s eval)
+ETA:7d9h16m
+Total train time:8d9h49m
+I1128 22:30:33.455401 137274321021824 utils.py:1231] [13750] l2_params = 302.1511134068482
+I1128 22:30:33.455612 137274321021824 utils.py:1231] [13750] train/loss = 3.5121888518333435
+I1128 22:30:33.455714 137274321021824 utils.py:1231] [13750] l2_grads = 1.2281590700149536
+I1128 22:30:33.455793 137274321021824 utils.py:1231] [13750] lr = 0.000996709411473376
+I1128 22:30:33.455853 137274321021824 utils.py:1231] [13750] uptime = 88822.818214579
+I1128 22:30:33.455918 137274321021824 utils.py:1231] [13750] examples_seen = 14080000.0
+I1128 22:30:33.455987 137274321021824 utils.py:1231] [13750] progress = 0.1221104233457368
+I1128 22:30:33.456050 137274321021824 utils.py:1231] [13750] epoch = 10.989980228963125
+I1128 22:30:33.456106 137274321021824 utils.py:1231] [13750] img/sec/core = 164.2166338887465
+I1128 22:30:33.456167 137274321021824 utils.py:1231] [13750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.63868745724861
+I1128 22:30:33.456217 137274321021824 utils.py:1231] [13750] core_hours = 24.63868745724861
+I1128 22:30:33.456279 137274321021824 train.py:125] NOTE: Steps:13750/112603 [12.2%]
+Walltime:24h40m (0s eval)
+ETA:7d9h9m
+Total train time:8d9h48m
+I1128 22:35:45.231392 137274321021824 utils.py:1231] [13800] l2_params = 302.51741792087336
+I1128 22:35:45.231602 137274321021824 utils.py:1231] [13800] train/loss = 4.505914092063904
+I1128 22:35:45.231701 137274321021824 utils.py:1231] [13800] l2_grads = 1.200292706489563
+I1128 22:35:45.231793 137274321021824 utils.py:1231] [13800] lr = 0.0009966211533831257
+I1128 22:35:45.231911 137274321021824 utils.py:1231] [13800] uptime = 89134.594250484
+I1128 22:35:45.231992 137274321021824 utils.py:1231] [13800] examples_seen = 14131200.0
+I1128 22:35:45.232051 137274321021824 utils.py:1231] [13800] progress = 0.1225544612488122
+I1128 22:35:45.232110 137274321021824 utils.py:1231] [13800] epoch = 11.029943793432082
+I1128 22:35:45.232167 137274321021824 utils.py:1231] [13800] img/sec/core = 164.2204470634838
+I1128 22:35:45.232231 137274321021824 utils.py:1231] [13800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.725291911666666
+I1128 22:35:45.232305 137274321021824 utils.py:1231] [13800] core_hours = 24.725291911666666
+I1128 22:35:45.232375 137274321021824 train.py:125] NOTE: Steps:13800/112603 [12.3%]
+Walltime:24h45m (0s eval)
+ETA:7d9h2m
+Total train time:8d9h46m
+I1128 22:40:54.564344 137274321021824 utils.py:1231] [13850] l2_params = 302.84895661801414
+I1128 22:40:54.564607 137274321021824 utils.py:1231] [13850] train/loss = 3.8735457956790924
+I1128 22:40:54.564724 137274321021824 utils.py:1231] [13850] l2_grads = 1.0819913148880005
+I1128 22:40:54.564809 137274321021824 utils.py:1231] [13850] lr = 0.0009965317313148938
+I1128 22:40:54.564877 137274321021824 utils.py:1231] [13850] uptime = 89443.927237732
+I1128 22:40:54.564944 137274321021824 utils.py:1231] [13850] examples_seen = 14182400.0
+I1128 22:40:54.565001 137274321021824 utils.py:1231] [13850] progress = 0.12299849915188761
+I1128 22:40:54.565058 137274321021824 utils.py:1231] [13850] epoch = 11.06990735790104
+I1128 22:40:54.565115 137274321021824 utils.py:1231] [13850] img/sec/core = 165.51742656192192
+I1128 22:40:54.565179 137274321021824 utils.py:1231] [13850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.811217741457774
+I1128 22:40:54.565245 137274321021824 utils.py:1231] [13850] core_hours = 24.811217741457774
+I1128 22:40:54.565314 137274321021824 train.py:125] NOTE: Steps:13850/112603 [12.3%]
+Walltime:24h50m (0s eval)
+ETA:7d8h56m
+Total train time:8d9h44m
+I1128 22:46:06.326923 137274321021824 utils.py:1231] [13900] l2_params = 303.26724041189107
+I1128 22:46:06.327200 137274321021824 utils.py:1231] [13900] train/loss = 5.124514400959015
+I1128 22:46:06.327425 137274321021824 utils.py:1231] [13900] l2_grads = 0.9612727165222168
+I1128 22:46:06.327512 137274321021824 utils.py:1231] [13900] lr = 0.000996441145478268
+I1128 22:46:06.327580 137274321021824 utils.py:1231] [13900] uptime = 89755.689941724
+I1128 22:46:06.327644 137274321021824 utils.py:1231] [13900] examples_seen = 14233600.0
+I1128 22:46:06.327705 137274321021824 utils.py:1231] [13900] progress = 0.12344253705496301
+I1128 22:46:06.327760 137274321021824 utils.py:1231] [13900] epoch = 11.109870922369996
+I1128 22:46:06.327829 137274321021824 utils.py:1231] [13900] img/sec/core = 164.22746962482432
+I1128 22:46:06.327906 137274321021824 utils.py:1231] [13900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.897818492566667
+I1128 22:46:06.327973 137274321021824 utils.py:1231] [13900] core_hours = 24.897818492566667
+I1128 22:46:06.328052 137274321021824 train.py:125] NOTE: Steps:13900/112603 [12.3%]
+Walltime:24h55m (0s eval)
+ETA:7d8h49m
+Total train time:8d9h43m
+I1128 22:51:16.113470 137274321021824 utils.py:1231] [13950] l2_params = 303.56607060873273
+I1128 22:51:16.113739 137274321021824 utils.py:1231] [13950] train/loss = 5.867316901683807
+I1128 22:51:16.113873 137274321021824 utils.py:1231] [13950] l2_grads = 1.011837124824524
+I1128 22:51:16.113967 137274321021824 utils.py:1231] [13950] lr = 0.000996349396085562
+I1128 22:51:16.114026 137274321021824 utils.py:1231] [13950] uptime = 90065.476387506
+I1128 22:51:16.114085 137274321021824 utils.py:1231] [13950] examples_seen = 14284800.0
+I1128 22:51:16.114140 137274321021824 utils.py:1231] [13950] progress = 0.12388657495803841
+I1128 22:51:16.114196 137274321021824 utils.py:1231] [13950] epoch = 11.149834486838952
+I1128 22:51:16.114253 137274321021824 utils.py:1231] [13950] img/sec/core = 165.2751458210365
+I1128 22:51:16.114314 137274321021824 utils.py:1231] [13950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 24.983870283061663
+I1128 22:51:16.114364 137274321021824 utils.py:1231] [13950] core_hours = 24.983870283061663
+I1128 22:51:16.114423 137274321021824 train.py:125] NOTE: Steps:13950/112603 [12.4%]
+Walltime:1d1h1m (0s eval)
+ETA:7d8h42m
+Total train time:8d9h41m
+I1128 22:56:27.899731 137274321021824 utils.py:1231] [14000] l2_params = 303.8923993505807
+I1128 22:56:27.900022 137274321021824 utils.py:1231] [14000] train/loss = 3.4753666520118713
+I1128 22:56:27.900200 137274321021824 utils.py:1231] [14000] l2_grads = 1.2792775630950928
+I1128 22:56:27.900281 137274321021824 utils.py:1231] [14000] lr = 0.0009962564833518185
+I1128 22:56:27.900342 137274321021824 utils.py:1231] [14000] uptime = 90377.262703685
+I1128 22:56:27.900419 137274321021824 utils.py:1231] [14000] examples_seen = 14336000.0
+I1128 22:56:27.900475 137274321021824 utils.py:1231] [14000] progress = 0.12433061286111383
+I1128 22:56:27.900530 137274321021824 utils.py:1231] [14000] epoch = 11.189798051307909
+I1128 22:56:27.900585 137274321021824 utils.py:1231] [14000] img/sec/core = 164.21503235762708
+I1128 22:56:27.900648 137274321021824 utils.py:1231] [14000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.070477593111388
+I1128 22:56:27.900705 137274321021824 utils.py:1231] [14000] core_hours = 25.070477593111388
+I1128 22:56:27.900776 137274321021824 train.py:125] NOTE: Steps:14000/112603 [12.4%]
+Walltime:1d1h6m (0s eval)
+ETA:7d8h35m
+Total train time:8d9h40m
+I1128 23:01:38.871555 137274321021824 utils.py:1231] [14050] l2_params = 304.14730894964026
+I1128 23:01:38.871737 137274321021824 utils.py:1231] [14050] train/loss = 3.4101647436618805
+I1128 23:01:38.871826 137274321021824 utils.py:1231] [14050] l2_grads = 1.3313755989074707
+I1128 23:01:38.871892 137274321021824 utils.py:1231] [14050] lr = 0.0009961624074948058
+I1128 23:01:38.871944 137274321021824 utils.py:1231] [14050] uptime = 90688.23430651601
+I1128 23:01:38.871992 137274321021824 utils.py:1231] [14050] examples_seen = 14387200.0
+I1128 23:01:38.872036 137274321021824 utils.py:1231] [14050] progress = 0.12477465076418923
+I1128 23:01:38.872080 137274321021824 utils.py:1231] [14050] epoch = 11.229761615776866
+I1128 23:01:38.872126 137274321021824 utils.py:1231] [14050] img/sec/core = 164.64525871136837
+I1128 23:01:38.872175 137274321021824 utils.py:1231] [14050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.156858593897777
+I1128 23:01:38.872225 137274321021824 utils.py:1231] [14050] core_hours = 25.156858593897777
+I1128 23:01:38.872295 137274321021824 train.py:125] NOTE: Steps:14050/112603 [12.5%]
+Walltime:1d1h11m (0s eval)
+ETA:7d8h29m
+Total train time:8d9h38m
+I1128 23:06:50.652937 137274321021824 utils.py:1231] [14100] l2_params = 304.4377052679507
+I1128 23:06:50.653197 137274321021824 utils.py:1231] [14100] train/loss = 5.20056539773941
+I1128 23:06:50.653330 137274321021824 utils.py:1231] [14100] l2_grads = 0.9058515429496765
+I1128 23:06:50.653398 137274321021824 utils.py:1231] [14100] lr = 0.0009960671687350175
+I1128 23:06:50.653464 137274321021824 utils.py:1231] [14100] uptime = 91000.015826115
+I1128 23:06:50.653524 137274321021824 utils.py:1231] [14100] examples_seen = 14438400.0
+I1128 23:06:50.653569 137274321021824 utils.py:1231] [14100] progress = 0.12521868866726463
+I1128 23:06:50.653614 137274321021824 utils.py:1231] [14100] epoch = 11.269725180245823
+I1128 23:06:50.653661 137274321021824 utils.py:1231] [14100] img/sec/core = 164.21755871179195
+I1128 23:06:50.653713 137274321021824 utils.py:1231] [14100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.243464571564168
+I1128 23:06:50.653759 137274321021824 utils.py:1231] [14100] core_hours = 25.243464571564168
+I1128 23:06:50.653815 137274321021824 train.py:125] NOTE: Steps:14100/112603 [12.5%]
+Walltime:1d1h16m (0s eval)
+ETA:7d8h22m
+Total train time:8d9h37m
+I1128 23:12:02.435775 137274321021824 utils.py:1231] [14150] l2_params = 304.7049227246264
+I1128 23:12:02.436040 137274321021824 utils.py:1231] [14150] train/loss = 5.50589245557785
+I1128 23:12:02.436266 137274321021824 utils.py:1231] [14150] l2_grads = 0.902302086353302
+I1128 23:12:02.436358 137274321021824 utils.py:1231] [14150] lr = 0.0009959707672956735
+I1128 23:12:02.436429 137274321021824 utils.py:1231] [14150] uptime = 91311.79878934601
+I1128 23:12:02.436493 137274321021824 utils.py:1231] [14150] examples_seen = 14489600.0
+I1128 23:12:02.436563 137274321021824 utils.py:1231] [14150] progress = 0.12566272657034006
+I1128 23:12:02.436646 137274321021824 utils.py:1231] [14150] epoch = 11.30968874471478
+I1128 23:12:02.436706 137274321021824 utils.py:1231] [14150] img/sec/core = 164.2167983439992
+I1128 23:12:02.436777 137274321021824 utils.py:1231] [14150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.330070950239442
+I1128 23:12:02.436836 137274321021824 utils.py:1231] [14150] core_hours = 25.330070950239442
+I1128 23:12:02.436910 137274321021824 train.py:125] NOTE: Steps:14150/112603 [12.6%]
+Walltime:1d1h21m (0s eval)
+ETA:7d8h16m
+Total train time:8d9h36m
+I1128 23:17:14.223998 137274321021824 utils.py:1231] [14200] l2_params = 304.9835648217378
+I1128 23:17:14.224193 137274321021824 utils.py:1231] [14200] train/loss = 3.509412169456482
+I1128 23:17:14.224291 137274321021824 utils.py:1231] [14200] l2_grads = 1.239006519317627
+I1128 23:17:14.224362 137274321021824 utils.py:1231] [14200] lr = 0.0009958732034027198
+I1128 23:17:14.224432 137274321021824 utils.py:1231] [14200] uptime = 91623.586793767
+I1128 23:17:14.224492 137274321021824 utils.py:1231] [14200] examples_seen = 14540800.0
+I1128 23:17:14.224552 137274321021824 utils.py:1231] [14200] progress = 0.12610676447341546
+I1128 23:17:14.224619 137274321021824 utils.py:1231] [14200] epoch = 11.349652309183737
+I1128 23:17:14.224672 137274321021824 utils.py:1231] [14200] img/sec/core = 164.2141431806584
+I1128 23:17:14.224734 137274321021824 utils.py:1231] [14200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.416678729245277
+I1128 23:17:14.224783 137274321021824 utils.py:1231] [14200] core_hours = 25.416678729245277
+I1128 23:17:14.224843 137274321021824 train.py:125] NOTE: Steps:14200/112603 [12.6%]
+Walltime:1d1h27m (0s eval)
+ETA:7d8h9m
+Total train time:8d9h34m
+I1128 23:22:26.005184 137274321021824 utils.py:1231] [14250] l2_params = 305.3000251894587
+I1128 23:22:26.005415 137274321021824 utils.py:1231] [14250] train/loss = 3.5314393043518066
+I1128 23:22:26.005520 137274321021824 utils.py:1231] [14250] l2_grads = 1.419582486152649
+I1128 23:22:26.005599 137274321021824 utils.py:1231] [14250] lr = 0.0009957744772848263
+I1128 23:22:26.005658 137274321021824 utils.py:1231] [14250] uptime = 91935.36801993201
+I1128 23:22:26.005717 137274321021824 utils.py:1231] [14250] examples_seen = 14592000.0
+I1128 23:22:26.005785 137274321021824 utils.py:1231] [14250] progress = 0.12655080237649086
+I1128 23:22:26.005845 137274321021824 utils.py:1231] [14250] epoch = 11.389615873652692
+I1128 23:22:26.005911 137274321021824 utils.py:1231] [14250] img/sec/core = 164.21771326571903
+I1128 23:22:26.005975 137274321021824 utils.py:1231] [14250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.50328462540222
+I1128 23:22:26.006026 137274321021824 utils.py:1231] [14250] core_hours = 25.50328462540222
+I1128 23:22:26.006087 137274321021824 train.py:125] NOTE: Steps:14250/112603 [12.7%]
+Walltime:1d1h32m (0s eval)
+ETA:7d8h2m
+Total train time:8d9h33m
+I1128 23:27:37.787985 137274321021824 utils.py:1231] [14300] l2_params = 305.64801156154465
+I1128 23:27:37.788202 137274321021824 utils.py:1231] [14300] train/loss = 5.700393855571747
+I1128 23:27:37.788302 137274321021824 utils.py:1231] [14300] l2_grads = 0.8041854500770569
+I1128 23:27:37.788369 137274321021824 utils.py:1231] [14300] lr = 0.0009956745891733876
+I1128 23:27:37.788427 137274321021824 utils.py:1231] [14300] uptime = 92247.150789705
+I1128 23:27:37.788483 137274321021824 utils.py:1231] [14300] examples_seen = 14643200.0
+I1128 23:27:37.788539 137274321021824 utils.py:1231] [14300] progress = 0.12699484027956626
+I1128 23:27:37.788592 137274321021824 utils.py:1231] [14300] epoch = 11.42957943812165
+I1128 23:27:37.788645 137274321021824 utils.py:1231] [14300] img/sec/core = 164.21690023883613
+I1128 23:27:37.788704 137274321021824 utils.py:1231] [14300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.589890950339164
+I1128 23:27:37.788753 137274321021824 utils.py:1231] [14300] core_hours = 25.589890950339164
+I1128 23:27:37.788814 137274321021824 train.py:125] NOTE: Steps:14300/112603 [12.7%]
+Walltime:1d1h37m (0s eval)
+ETA:7d7h56m
+Total train time:8d9h31m
+I1128 23:32:49.562554 137274321021824 utils.py:1231] [14350] l2_params = 305.8915945152604
+I1128 23:32:49.562762 137274321021824 utils.py:1231] [14350] train/loss = 3.977884531021118
+I1128 23:32:49.562852 137274321021824 utils.py:1231] [14350] l2_grads = 1.2181254625320435
+I1128 23:32:49.562918 137274321021824 utils.py:1231] [14350] lr = 0.0009955735393025191
+I1128 23:32:49.562969 137274321021824 utils.py:1231] [14350] uptime = 92558.92533221301
+I1128 23:32:49.563019 137274321021824 utils.py:1231] [14350] examples_seen = 14694400.0
+I1128 23:32:49.563065 137274321021824 utils.py:1231] [14350] progress = 0.12743887818264166
+I1128 23:32:49.563110 137274321021824 utils.py:1231] [14350] epoch = 11.469543002590607
+I1128 23:32:49.563157 137274321021824 utils.py:1231] [14350] img/sec/core = 164.2212336778111
+I1128 23:32:49.563215 137274321021824 utils.py:1231] [14350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.676494989924723
+I1128 23:32:49.563263 137274321021824 utils.py:1231] [14350] core_hours = 25.676494989924723
+I1128 23:32:49.563318 137274321021824 train.py:125] NOTE: Steps:14350/112603 [12.7%]
+Walltime:1d1h42m (0s eval)
+ETA:7d7h49m
+Total train time:8d9h30m
+I1128 23:38:01.344973 137274321021824 utils.py:1231] [14400] l2_params = 306.1970112565407
+I1128 23:38:01.345199 137274321021824 utils.py:1231] [14400] train/loss = 4.531881630420685
+I1128 23:38:01.345326 137274321021824 utils.py:1231] [14400] l2_grads = 1.232132911682129
+I1128 23:38:01.345407 137274321021824 utils.py:1231] [14400] lr = 0.0009954713279090599
+I1128 23:38:01.345479 137274321021824 utils.py:1231] [14400] uptime = 92870.707838236
+I1128 23:38:01.345548 137274321021824 utils.py:1231] [14400] examples_seen = 14745600.0
+I1128 23:38:01.345609 137274321021824 utils.py:1231] [14400] progress = 0.12788291608571709
+I1128 23:38:01.345676 137274321021824 utils.py:1231] [14400] epoch = 11.509506567059564
+I1128 23:38:01.345751 137274321021824 utils.py:1231] [14400] img/sec/core = 164.21703915685313
+I1128 23:38:01.345826 137274321021824 utils.py:1231] [14400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.763101241597777
+I1128 23:38:01.345906 137274321021824 utils.py:1231] [14400] core_hours = 25.763101241597777
+I1128 23:38:01.346014 137274321021824 train.py:125] NOTE: Steps:14400/112603 [12.8%]
+Walltime:1d1h47m (0s eval)
+ETA:7d7h43m
+Total train time:8d9h29m
+I1128 23:43:13.123144 137274321021824 utils.py:1231] [14450] l2_params = 306.4261421127086
+I1128 23:43:13.123342 137274321021824 utils.py:1231] [14450] train/loss = 3.543203592300415
+I1128 23:43:13.123433 137274321021824 utils.py:1231] [14450] l2_grads = 1.2304396629333496
+I1128 23:43:13.123497 137274321021824 utils.py:1231] [14450] lr = 0.000995367955232576
+I1128 23:43:13.123554 137274321021824 utils.py:1231] [14450] uptime = 93182.48591617201
+I1128 23:43:13.123607 137274321021824 utils.py:1231] [14450] examples_seen = 14796800.0
+I1128 23:43:13.123659 137274321021824 utils.py:1231] [14450] progress = 0.12832695398879249
+I1128 23:43:13.123709 137274321021824 utils.py:1231] [14450] epoch = 11.54947013152852
+I1128 23:43:13.123762 137274321021824 utils.py:1231] [14450] img/sec/core = 164.21937148034326
+I1128 23:43:13.123824 137274321021824 utils.py:1231] [14450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.849706263246667
+I1128 23:43:13.123889 137274321021824 utils.py:1231] [14450] core_hours = 25.849706263246667
+I1128 23:43:13.123954 137274321021824 train.py:125] NOTE: Steps:14450/112603 [12.8%]
+Walltime:1d1h53m (0s eval)
+ETA:7d7h36m
+Total train time:8d9h27m
+I1128 23:48:24.374593 137274321021824 utils.py:1231] [14500] l2_params = 306.65391265594917
+I1128 23:48:24.374826 137274321021824 utils.py:1231] [14500] train/loss = 4.518525660037994
+I1128 23:48:24.374959 137274321021824 utils.py:1231] [14500] l2_grads = 1.122772455215454
+I1128 23:48:24.375058 137274321021824 utils.py:1231] [14500] lr = 0.0009952634215153488
+I1128 23:48:24.375116 137274321021824 utils.py:1231] [14500] uptime = 93493.73747784301
+I1128 23:48:24.375187 137274321021824 utils.py:1231] [14500] examples_seen = 14848000.0
+I1128 23:48:24.375243 137274321021824 utils.py:1231] [14500] progress = 0.1287709918918679
+I1128 23:48:24.375308 137274321021824 utils.py:1231] [14500] epoch = 11.589433695997478
+I1128 23:48:24.375371 137274321021824 utils.py:1231] [14500] img/sec/core = 164.4971666170127
+I1128 23:48:24.375429 137274321021824 utils.py:1231] [14500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 25.9361650303775
+I1128 23:48:24.375484 137274321021824 utils.py:1231] [14500] core_hours = 25.9361650303775
+I1128 23:48:24.375566 137274321021824 train.py:125] NOTE: Steps:14500/112603 [12.9%]
+Walltime:1d1h58m (0s eval)
+ETA:7d7h30m
+Total train time:8d9h26m
+I1128 23:53:35.659335 137274321021824 utils.py:1231] [14550] l2_params = 306.9754400975707
+I1128 23:53:35.659587 137274321021824 utils.py:1231] [14550] train/loss = 3.53346985578537
+I1128 23:53:35.659741 137274321021824 utils.py:1231] [14550] l2_grads = 1.3529921770095825
+I1128 23:53:35.659850 137274321021824 utils.py:1231] [14550] lr = 0.0009951577270023867
+I1128 23:53:35.659945 137274321021824 utils.py:1231] [14550] uptime = 93805.02230158601
+I1128 23:53:35.660027 137274321021824 utils.py:1231] [14550] examples_seen = 14899200.0
+I1128 23:53:35.660110 137274321021824 utils.py:1231] [14550] progress = 0.1292150297949433
+I1128 23:53:35.660204 137274321021824 utils.py:1231] [14550] epoch = 11.629397260466433
+I1128 23:53:35.660288 137274321021824 utils.py:1231] [14550] img/sec/core = 164.4795894138133
+I1128 23:53:35.660375 137274321021824 utils.py:1231] [14550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.022633036972778
+I1128 23:53:35.660470 137274321021824 utils.py:1231] [14550] core_hours = 26.022633036972778
+I1128 23:53:35.660569 137274321021824 train.py:125] NOTE: Steps:14550/112603 [12.9%]
+Walltime:1d2h3m (0s eval)
+ETA:7d7h23m
+Total train time:8d9h25m
+I1128 23:58:46.810286 137274321021824 utils.py:1231] [14600] l2_params = 307.2353705489632
+I1128 23:58:46.810514 137274321021824 utils.py:1231] [14600] train/loss = 3.9702776968479156
+I1128 23:58:46.810627 137274321021824 utils.py:1231] [14600] l2_grads = 1.2085734605789185
+I1128 23:58:46.810694 137274321021824 utils.py:1231] [14600] lr = 0.0009950508719414124
+I1128 23:58:46.810753 137274321021824 utils.py:1231] [14600] uptime = 94116.17311494201
+I1128 23:58:46.810815 137274321021824 utils.py:1231] [14600] examples_seen = 14950400.0
+I1128 23:58:46.810870 137274321021824 utils.py:1231] [14600] progress = 0.12965906769801872
+I1128 23:58:46.810931 137274321021824 utils.py:1231] [14600] epoch = 11.66936082493539
+I1128 23:58:46.810990 137274321021824 utils.py:1231] [14600] img/sec/core = 164.55042957390677
+I1128 23:58:46.811048 137274321021824 utils.py:1231] [14600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.10906381846055
+I1128 23:58:46.811119 137274321021824 utils.py:1231] [14600] core_hours = 26.10906381846055
+I1128 23:58:46.811183 137274321021824 train.py:125] NOTE: Steps:14600/112603 [13.0%]
+Walltime:1d2h8m (0s eval)
+ETA:7d7h16m
+Total train time:8d9h23m
+I1129 00:03:57.065221 137274321021824 utils.py:1231] [14650] l2_params = 307.48990295365155
+I1129 00:03:57.065445 137274321021824 utils.py:1231] [14650] train/loss = 3.482665330171585
+I1129 00:03:57.065546 137274321021824 utils.py:1231] [14650] l2_grads = 1.2575515508651733
+I1129 00:03:57.065614 137274321021824 utils.py:1231] [14650] lr = 0.0009949428565828745
+I1129 00:03:57.065673 137274321021824 utils.py:1231] [14650] uptime = 94426.42803460201
+I1129 00:03:57.065732 137274321021824 utils.py:1231] [14650] examples_seen = 15001600.0
+I1129 00:03:57.065792 137274321021824 utils.py:1231] [14650] progress = 0.13010310560109412
+I1129 00:03:57.065850 137274321021824 utils.py:1231] [14650] epoch = 11.709324389404347
+I1129 00:03:57.065913 137274321021824 utils.py:1231] [14650] img/sec/core = 165.0255862376284
+I1129 00:03:57.065993 137274321021824 utils.py:1231] [14650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.195245740588334
+I1129 00:03:57.066056 137274321021824 utils.py:1231] [14650] core_hours = 26.195245740588334
+I1129 00:03:57.066121 137274321021824 train.py:125] NOTE: Steps:14650/112603 [13.0%]
+Walltime:1d2h13m (0s eval)
+ETA:7d7h10m
+Total train time:8d9h22m
+I1129 00:09:07.361377 137274321021824 utils.py:1231] [14700] l2_params = 307.7345310097962
+I1129 00:09:07.361598 137274321021824 utils.py:1231] [14700] train/loss = 3.922793924808502
+I1129 00:09:07.361700 137274321021824 utils.py:1231] [14700] l2_grads = 1.161606788635254
+I1129 00:09:07.361776 137274321021824 utils.py:1231] [14700] lr = 0.0009948336811799378
+I1129 00:09:07.361834 137274321021824 utils.py:1231] [14700] uptime = 94736.724195692
+I1129 00:09:07.361896 137274321021824 utils.py:1231] [14700] examples_seen = 15052800.0
+I1129 00:09:07.361973 137274321021824 utils.py:1231] [14700] progress = 0.13054714350416952
+I1129 00:09:07.362025 137274321021824 utils.py:1231] [14700] epoch = 11.749287953873305
+I1129 00:09:07.362080 137274321021824 utils.py:1231] [14700] img/sec/core = 165.0036527043934
+I1129 00:09:07.362136 137274321021824 utils.py:1231] [14700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.28143911866889
+I1129 00:09:07.362187 137274321021824 utils.py:1231] [14700] core_hours = 26.28143911866889
+I1129 00:09:07.362272 137274321021824 train.py:125] NOTE: Steps:14700/112603 [13.1%]
+Walltime:1d2h18m (0s eval)
+ETA:7d7h3m
+Total train time:8d9h20m
+I1129 00:14:19.139207 137274321021824 utils.py:1231] [14750] l2_params = 308.04734827687815
+I1129 00:14:19.139457 137274321021824 utils.py:1231] [14750] train/loss = 3.851358652114868
+I1129 00:14:19.139571 137274321021824 utils.py:1231] [14750] l2_grads = 1.185442566871643
+I1129 00:14:19.139647 137274321021824 utils.py:1231] [14750] lr = 0.0009947233459884877
+I1129 00:14:19.139719 137274321021824 utils.py:1231] [14750] uptime = 95048.502080585
+I1129 00:14:19.139782 137274321021824 utils.py:1231] [14750] examples_seen = 15104000.0
+I1129 00:14:19.139839 137274321021824 utils.py:1231] [14750] progress = 0.13099118140724492
+I1129 00:14:19.139898 137274321021824 utils.py:1231] [14750] epoch = 11.789251518342262
+I1129 00:14:19.139962 137274321021824 utils.py:1231] [14750] img/sec/core = 164.2194731597817
+I1129 00:14:19.140022 137274321021824 utils.py:1231] [14750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.368044086694724
+I1129 00:14:19.140076 137274321021824 utils.py:1231] [14750] core_hours = 26.368044086694724
+I1129 00:14:19.140151 137274321021824 train.py:125] NOTE: Steps:14750/112603 [13.1%]
+Walltime:1d2h24m (0s eval)
+ETA:7d6h57m
+Total train time:8d9h19m
+I1129 00:19:28.335879 137274321021824 utils.py:1231] [14800] l2_params = 308.27410190476553
+I1129 00:19:28.336075 137274321021824 utils.py:1231] [14800] train/loss = 3.956108272075653
+I1129 00:19:28.336165 137274321021824 utils.py:1231] [14800] l2_grads = 1.2266638278961182
+I1129 00:19:28.336233 137274321021824 utils.py:1231] [14800] lr = 0.000994611851267128
+I1129 00:19:28.336290 137274321021824 utils.py:1231] [14800] uptime = 95357.698651948
+I1129 00:19:28.336343 137274321021824 utils.py:1231] [14800] examples_seen = 15155200.0
+I1129 00:19:28.336391 137274321021824 utils.py:1231] [14800] progress = 0.13143521931032032
+I1129 00:19:28.336441 137274321021824 utils.py:1231] [14800] epoch = 11.829215082811219
+I1129 00:19:28.336490 137274321021824 utils.py:1231] [14800] img/sec/core = 165.59045197138428
+I1129 00:19:28.336544 137274321021824 utils.py:1231] [14800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.453932023184443
+I1129 00:19:28.336593 137274321021824 utils.py:1231] [14800] core_hours = 26.453932023184443
+I1129 00:19:28.336651 137274321021824 train.py:125] NOTE: Steps:14800/112603 [13.1%]
+Walltime:1d2h29m (0s eval)
+ETA:7d6h50m
+Total train time:8d9h17m
+I1129 00:24:40.120006 137274321021824 utils.py:1231] [14850] l2_params = 308.4642260881235
+I1129 00:24:40.120221 137274321021824 utils.py:1231] [14850] train/loss = 4.976806402206421
+I1129 00:24:40.120316 137274321021824 utils.py:1231] [14850] l2_grads = 0.9948286414146423
+I1129 00:24:40.120393 137274321021824 utils.py:1231] [14850] lr = 0.000994499197277177
+I1129 00:24:40.120445 137274321021824 utils.py:1231] [14850] uptime = 95669.482806678
+I1129 00:24:40.120504 137274321021824 utils.py:1231] [14850] examples_seen = 15206400.0
+I1129 00:24:40.120559 137274321021824 utils.py:1231] [14850] progress = 0.13187925721339575
+I1129 00:24:40.120607 137274321021824 utils.py:1231] [14850] epoch = 11.869178647280176
+I1129 00:24:40.120657 137274321021824 utils.py:1231] [14850] img/sec/core = 164.21617078115324
+I1129 00:24:40.120719 137274321021824 utils.py:1231] [14850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.540538732831667
+I1129 00:24:40.120777 137274321021824 utils.py:1231] [14850] core_hours = 26.540538732831667
+I1129 00:24:40.120836 137274321021824 train.py:125] NOTE: Steps:14850/112603 [13.2%]
+Walltime:1d2h34m (0s eval)
+ETA:7d6h43m
+Total train time:8d9h16m
+I1129 00:29:51.896933 137274321021824 utils.py:1231] [14900] l2_params = 308.77209417305113
+I1129 00:29:51.897174 137274321021824 utils.py:1231] [14900] train/loss = 5.71111786365509
+I1129 00:29:51.897299 137274321021824 utils.py:1231] [14900] l2_grads = 0.882063090801239
+I1129 00:29:51.897382 137274321021824 utils.py:1231] [14900] lr = 0.0009943853842826744
+I1129 00:29:51.897440 137274321021824 utils.py:1231] [14900] uptime = 95981.259797611
+I1129 00:29:51.897505 137274321021824 utils.py:1231] [14900] examples_seen = 15257600.0
+I1129 00:29:51.897552 137274321021824 utils.py:1231] [14900] progress = 0.13232329511647115
+I1129 00:29:51.897599 137274321021824 utils.py:1231] [14900] epoch = 11.909142211749131
+I1129 00:29:51.897649 137274321021824 utils.py:1231] [14900] img/sec/core = 164.21994402724746
+I1129 00:29:51.897703 137274321021824 utils.py:1231] [14900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.627143452535275
+I1129 00:29:51.897752 137274321021824 utils.py:1231] [14900] core_hours = 26.627143452535275
+I1129 00:29:51.897811 137274321021824 train.py:125] NOTE: Steps:14900/112603 [13.2%]
+Walltime:1d2h39m (0s eval)
+ETA:7d6h37m
+Total train time:8d9h15m
+I1129 00:35:03.675204 137274321021824 utils.py:1231] [14950] l2_params = 309.07341102055915
+I1129 00:35:03.675452 137274321021824 utils.py:1231] [14950] train/loss = 4.055813193321228
+I1129 00:35:03.675578 137274321021824 utils.py:1231] [14950] l2_grads = 1.204145908355713
+I1129 00:35:03.675656 137274321021824 utils.py:1231] [14950] lr = 0.0009942704125503736
+I1129 00:35:03.675727 137274321021824 utils.py:1231] [14950] uptime = 96293.03808876201
+I1129 00:35:03.675785 137274321021824 utils.py:1231] [14950] examples_seen = 15308800.0
+I1129 00:35:03.675839 137274321021824 utils.py:1231] [14950] progress = 0.13276733301954655
+I1129 00:35:03.675904 137274321021824 utils.py:1231] [14950] epoch = 11.949105776218088
+I1129 00:35:03.675960 137274321021824 utils.py:1231] [14950] img/sec/core = 164.21925917606768
+I1129 00:35:03.676019 137274321021824 utils.py:1231] [14950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.71374853341056
+I1129 00:35:03.676074 137274321021824 utils.py:1231] [14950] core_hours = 26.71374853341056
+I1129 00:35:03.676134 137274321021824 train.py:125] NOTE: Steps:14950/112603 [13.3%]
+Walltime:1d2h44m (0s eval)
+ETA:7d6h31m
+Total train time:8d9h14m
+I1129 00:40:15.454668 137274321021824 utils.py:1231] [15000] l2_params = 309.316807849467
+I1129 00:40:15.454981 137274321021824 utils.py:1231] [15000] train/loss = 4.671006500720978
+I1129 00:40:15.455221 137274321021824 utils.py:1231] [15000] l2_grads = 1.1762081384658813
+I1129 00:40:15.455361 137274321021824 utils.py:1231] [15000] lr = 0.0009941542823497457
+I1129 00:40:15.455467 137274321021824 utils.py:1231] [15000] uptime = 96604.817824732
+I1129 00:40:15.455578 137274321021824 utils.py:1231] [15000] examples_seen = 15360000.0
+I1129 00:40:15.455688 137274321021824 utils.py:1231] [15000] progress = 0.13321137092262195
+I1129 00:40:15.455805 137274321021824 utils.py:1231] [15000] epoch = 11.989069340687045
+I1129 00:40:15.455924 137274321021824 utils.py:1231] [15000] img/sec/core = 164.21849816733464
+I1129 00:40:15.456018 137274321021824 utils.py:1231] [15000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.800354015624443
+I1129 00:40:15.456103 137274321021824 utils.py:1231] [15000] core_hours = 26.800354015624443
+I1129 00:40:15.456205 137274321021824 train.py:125] NOTE: Steps:15000/112603 [13.3%]
+Walltime:1d2h50m (0s eval)
+ETA:7d6h24m
+Total train time:8d9h12m
+I1129 00:40:15.776956 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:15000/112603 [13.3%]
+Walltime:1d2h50m (0s eval)
+ETA:7d6h24m
+Total train time:8d9h12m
+I1129 00:41:46.545991 137274321021824 utils.py:1231] [15000] val/acc@1 = 0.4536232461734694
+I1129 00:41:46.546264 137274321021824 utils.py:1231] [15000] val/loss = 2.49043986353339
+I1129 00:41:46.546488 137274321021824 utils.py:1231] [15000] z/secs/eval/val = 90.76930089699454
+I1129 00:41:46.546602 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 90.76930089699454
+I1129 00:46:57.401211 137274321021824 utils.py:1231] [15050] l2_params = 309.603293961595
+I1129 00:46:57.401480 137274321021824 utils.py:1231] [15050] train/loss = 5.254890978336334
+I1129 00:46:57.401655 137274321021824 utils.py:1231] [15050] l2_grads = 1.0487709045410156
+I1129 00:46:57.401777 137274321021824 utils.py:1231] [15050] lr = 0.000994036993952975
+I1129 00:46:57.401850 137274321021824 utils.py:1231] [15050] uptime = 97006.764207899
+I1129 00:46:57.401933 137274321021824 utils.py:1231] [15050] examples_seen = 15411200.0
+I1129 00:46:57.401992 137274321021824 utils.py:1231] [15050] progress = 0.13365540882569737
+I1129 00:46:57.402049 137274321021824 utils.py:1231] [15050] epoch = 12.029032905156003
+I1129 00:46:57.402103 137274321021824 utils.py:1231] [15050] img/sec/core = 127.38017343653397
+I1129 00:46:57.402163 137274321021824 utils.py:1231] [15050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.912005788726386
+I1129 00:46:57.402218 137274321021824 utils.py:1231] [15050] core_hours = 26.912005788726386
+I1129 00:46:57.402287 137274321021824 train.py:125] NOTE: Steps:15050/112603 [13.4%]
+Walltime:1d2h56m (0s eval)
+ETA:7d6h27m
+Total train time:8d9h22m
+I1129 00:52:07.739644 137274321021824 utils.py:1231] [15100] l2_params = 309.9177571679187
+I1129 00:52:07.739832 137274321021824 utils.py:1231] [15100] train/loss = 3.4176210165023804
+I1129 00:52:07.739938 137274321021824 utils.py:1231] [15100] l2_grads = 1.2965785264968872
+I1129 00:52:07.740027 137274321021824 utils.py:1231] [15100] lr = 0.000993918547634963
+I1129 00:52:07.740125 137274321021824 utils.py:1231] [15100] uptime = 97317.102485908
+I1129 00:52:07.740189 137274321021824 utils.py:1231] [15100] examples_seen = 15462400.0
+I1129 00:52:07.740258 137274321021824 utils.py:1231] [15100] progress = 0.13409944672877278
+I1129 00:52:07.740312 137274321021824 utils.py:1231] [15100] epoch = 12.06899646962496
+I1129 00:52:07.740369 137274321021824 utils.py:1231] [15100] img/sec/core = 164.98125957415616
+I1129 00:52:07.740440 137274321021824 utils.py:1231] [15100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 26.998210865951112
+I1129 00:52:07.740491 137274321021824 utils.py:1231] [15100] core_hours = 26.998210865951112
+I1129 00:52:07.740553 137274321021824 train.py:125] NOTE: Steps:15100/112603 [13.4%]
+Walltime:1d3h1m (0s eval)
+ETA:7d6h21m
+Total train time:8d9h21m
+I1129 00:57:19.514045 137274321021824 utils.py:1231] [15150] l2_params = 310.20922334668023
+I1129 00:57:19.514286 137274321021824 utils.py:1231] [15150] train/loss = 3.419518083333969
+I1129 00:57:19.514407 137274321021824 utils.py:1231] [15150] l2_grads = 1.216475248336792
+I1129 00:57:19.514483 137274321021824 utils.py:1231] [15150] lr = 0.0009937989436733232
+I1129 00:57:19.514538 137274321021824 utils.py:1231] [15150] uptime = 97628.876900557
+I1129 00:57:19.514603 137274321021824 utils.py:1231] [15150] examples_seen = 15513600.0
+I1129 00:57:19.514653 137274321021824 utils.py:1231] [15150] progress = 0.13454348463184818
+I1129 00:57:19.514704 137274321021824 utils.py:1231] [15150] epoch = 12.108960034093917
+I1129 00:57:19.514760 137274321021824 utils.py:1231] [15150] img/sec/core = 164.22130102510692
+I1129 00:57:19.514820 137274321021824 utils.py:1231] [15150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.084814870020278
+I1129 00:57:19.514874 137274321021824 utils.py:1231] [15150] core_hours = 27.084814870020278
+I1129 00:57:19.514942 137274321021824 train.py:125] NOTE: Steps:15150/112603 [13.5%]
+Walltime:1d3h7m (0s eval)
+ETA:7d6h14m
+Total train time:8d9h20m
+I1129 01:02:28.240278 137274321021824 utils.py:1231] [15200] l2_params = 310.4672218832298
+I1129 01:02:28.240484 137274321021824 utils.py:1231] [15200] train/loss = 4.303513944149017
+I1129 01:02:28.240607 137274321021824 utils.py:1231] [15200] l2_grads = 1.1172679662704468
+I1129 01:02:28.240682 137274321021824 utils.py:1231] [15200] lr = 0.0009936781823483803
+I1129 01:02:28.240750 137274321021824 utils.py:1231] [15200] uptime = 97937.603099863
+I1129 01:02:28.240805 137274321021824 utils.py:1231] [15200] examples_seen = 15564800.0
+I1129 01:02:28.240854 137274321021824 utils.py:1231] [15200] progress = 0.13498752253492358
+I1129 01:02:28.240906 137274321021824 utils.py:1231] [15200] epoch = 12.148923598562872
+I1129 01:02:28.240956 137274321021824 utils.py:1231] [15200] img/sec/core = 165.84274387821878
+I1129 01:02:28.241011 137274321021824 utils.py:1231] [15200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.170572147605277
+I1129 01:02:28.241059 137274321021824 utils.py:1231] [15200] core_hours = 27.170572147605277
+I1129 01:02:28.241117 137274321021824 train.py:125] NOTE: Steps:15200/112603 [13.5%]
+Walltime:1d3h12m (0s eval)
+ETA:7d6h8m
+Total train time:8d9h18m
+I1129 01:07:33.302341 137274321021824 utils.py:1231] [15250] l2_params = 310.7328056223982
+I1129 01:07:33.302587 137274321021824 utils.py:1231] [15250] train/loss = 3.4687947928905487
+I1129 01:07:33.302711 137274321021824 utils.py:1231] [15250] l2_grads = 1.2753156423568726
+I1129 01:07:33.302805 137274321021824 utils.py:1231] [15250] lr = 0.0009935562639431773
+I1129 01:07:33.302874 137274321021824 utils.py:1231] [15250] uptime = 98242.66523271399
+I1129 01:07:33.302981 137274321021824 utils.py:1231] [15250] examples_seen = 15616000.0
+I1129 01:07:33.303043 137274321021824 utils.py:1231] [15250] progress = 0.13543156043799898
+I1129 01:07:33.303105 137274321021824 utils.py:1231] [15250] epoch = 12.18888716303183
+I1129 01:07:33.303180 137274321021824 utils.py:1231] [15250] img/sec/core = 167.8346621440866
+I1129 01:07:33.303256 137274321021824 utils.py:1231] [15250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.25531162895277
+I1129 01:07:33.303314 137274321021824 utils.py:1231] [15250] core_hours = 27.25531162895277
+I1129 01:07:33.303377 137274321021824 train.py:125] NOTE: Steps:15250/112603 [13.5%]
+Walltime:1d3h17m (0s eval)
+ETA:7d6h0m
+Total train time:8d9h16m
+I1129 01:12:42.127608 137274321021824 utils.py:1231] [15300] l2_params = 311.0124691483756
+I1129 01:12:42.127852 137274321021824 utils.py:1231] [15300] train/loss = 5.834261119365692
+I1129 01:12:42.128004 137274321021824 utils.py:1231] [15300] l2_grads = 1.1510124206542969
+I1129 01:12:42.128085 137274321021824 utils.py:1231] [15300] lr = 0.0009934331887434637
+I1129 01:12:42.128160 137274321021824 utils.py:1231] [15300] uptime = 98551.490518674
+I1129 01:12:42.128238 137274321021824 utils.py:1231] [15300] examples_seen = 15667200.0
+I1129 01:12:42.128311 137274321021824 utils.py:1231] [15300] progress = 0.1358755983410744
+I1129 01:12:42.128401 137274321021824 utils.py:1231] [15300] epoch = 12.228850727500786
+I1129 01:12:42.128483 137274321021824 utils.py:1231] [15300] img/sec/core = 165.78953320107541
+I1129 01:12:42.128565 137274321021824 utils.py:1231] [15300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.34109643060833
+I1129 01:12:42.128638 137274321021824 utils.py:1231] [15300] core_hours = 27.34109643060833
+I1129 01:12:42.128729 137274321021824 train.py:125] NOTE: Steps:15300/112603 [13.6%]
+Walltime:1d3h22m (0s eval)
+ETA:7d5h54m
+Total train time:8d9h14m
+I1129 01:17:48.390550 137274321021824 utils.py:1231] [15350] l2_params = 311.2404446405252
+I1129 01:17:48.390746 137274321021824 utils.py:1231] [15350] train/loss = 4.235298216342926
+I1129 01:17:48.390836 137274321021824 utils.py:1231] [15350] l2_grads = 1.0472989082336426
+I1129 01:17:48.390910 137274321021824 utils.py:1231] [15350] lr = 0.000993308957037704
+I1129 01:17:48.390982 137274321021824 utils.py:1231] [15350] uptime = 98857.753343751
+I1129 01:17:48.391050 137274321021824 utils.py:1231] [15350] examples_seen = 15718400.0
+I1129 01:17:48.391104 137274321021824 utils.py:1231] [15350] progress = 0.1363196362441498
+I1129 01:17:48.391151 137274321021824 utils.py:1231] [15350] epoch = 12.268814291969743
+I1129 01:17:48.391200 137274321021824 utils.py:1231] [15350] img/sec/core = 167.17667247772795
+I1129 01:17:48.391256 137274321021824 utils.py:1231] [15350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.426169437574163
+I1129 01:17:48.391306 137274321021824 utils.py:1231] [15350] core_hours = 27.426169437574163
+I1129 01:17:48.391364 137274321021824 train.py:125] NOTE: Steps:15350/112603 [13.6%]
+Walltime:1d3h27m (0s eval)
+ETA:7d5h47m
+Total train time:8d9h12m
+I1129 01:22:53.322449 137274321021824 utils.py:1231] [15400] l2_params = 311.492754030393
+I1129 01:22:53.322753 137274321021824 utils.py:1231] [15400] train/loss = 5.6028202176094055
+I1129 01:22:53.322935 137274321021824 utils.py:1231] [15400] l2_grads = 1.1167521476745605
+I1129 01:22:53.323000 137274321021824 utils.py:1231] [15400] lr = 0.000993183569117071
+I1129 01:22:53.323053 137274321021824 utils.py:1231] [15400] uptime = 99162.68541464301
+I1129 01:22:53.323103 137274321021824 utils.py:1231] [15400] examples_seen = 15769600.0
+I1129 01:22:53.323151 137274321021824 utils.py:1231] [15400] progress = 0.1367636741472252
+I1129 01:22:53.323198 137274321021824 utils.py:1231] [15400] epoch = 12.3087778564387
+I1129 01:22:53.323247 137274321021824 utils.py:1231] [15400] img/sec/core = 167.9062482677713
+I1129 01:22:53.323302 137274321021824 utils.py:1231] [15400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.510872790599723
+I1129 01:22:53.323351 137274321021824 utils.py:1231] [15400] core_hours = 27.510872790599723
+I1129 01:22:53.323422 137274321021824 train.py:125] NOTE: Steps:15400/112603 [13.7%]
+Walltime:1d3h32m (0s eval)
+ETA:7d5h40m
+Total train time:8d9h10m
+I1129 01:28:05.089132 137274321021824 utils.py:1231] [15450] l2_params = 311.69684910024375
+I1129 01:28:05.089380 137274321021824 utils.py:1231] [15450] train/loss = 4.218046069145203
+I1129 01:28:05.089480 137274321021824 utils.py:1231] [15450] l2_grads = 1.1173484325408936
+I1129 01:28:05.089554 137274321021824 utils.py:1231] [15450] lr = 0.0009930570252754476
+I1129 01:28:05.089647 137274321021824 utils.py:1231] [15450] uptime = 99474.452005935
+I1129 01:28:05.089757 137274321021824 utils.py:1231] [15450] examples_seen = 15820800.0
+I1129 01:28:05.089834 137274321021824 utils.py:1231] [15450] progress = 0.1372077120503006
+I1129 01:28:05.089922 137274321021824 utils.py:1231] [15450] epoch = 12.348741420907658
+I1129 01:28:05.089998 137274321021824 utils.py:1231] [15450] img/sec/core = 164.22542193447418
+I1129 01:28:05.090079 137274321021824 utils.py:1231] [15450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.597474621514163
+I1129 01:28:05.090149 137274321021824 utils.py:1231] [15450] core_hours = 27.597474621514163
+I1129 01:28:05.090247 137274321021824 train.py:125] NOTE: Steps:15450/112603 [13.7%]
+Walltime:1d3h37m (0s eval)
+ETA:7d5h33m
+Total train time:8d9h9m
+I1129 01:33:16.877836 137274321021824 utils.py:1231] [15500] l2_params = 311.9325128605957
+I1129 01:33:16.878158 137274321021824 utils.py:1231] [15500] train/loss = 3.4657942950725555
+I1129 01:33:16.878322 137274321021824 utils.py:1231] [15500] l2_grads = 1.381850242614746
+I1129 01:33:16.878395 137274321021824 utils.py:1231] [15500] lr = 0.0009929293258094272
+I1129 01:33:16.878467 137274321021824 utils.py:1231] [15500] uptime = 99786.240829042
+I1129 01:33:16.878528 137274321021824 utils.py:1231] [15500] examples_seen = 15872000.0
+I1129 01:33:16.878584 137274321021824 utils.py:1231] [15500] progress = 0.137651749953376
+I1129 01:33:16.878649 137274321021824 utils.py:1231] [15500] epoch = 12.388704985376613
+I1129 01:33:16.878707 137274321021824 utils.py:1231] [15500] img/sec/core = 164.2137119919461
+I1129 01:33:16.878791 137274321021824 utils.py:1231] [15500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.684082627932774
+I1129 01:33:16.878873 137274321021824 utils.py:1231] [15500] core_hours = 27.684082627932774
+I1129 01:33:16.878980 137274321021824 train.py:125] NOTE: Steps:15500/112603 [13.8%]
+Walltime:1d3h43m (0s eval)
+ETA:7d5h27m
+Total train time:8d9h8m
+I1129 01:38:23.003363 137274321021824 utils.py:1231] [15550] l2_params = 312.0905839960563
+I1129 01:38:23.003701 137274321021824 utils.py:1231] [15550] train/loss = 3.5062576830387115
+I1129 01:38:23.003880 137274321021824 utils.py:1231] [15550] l2_grads = 1.221704363822937
+I1129 01:38:23.003955 137274321021824 utils.py:1231] [15550] lr = 0.0009928004710183112
+I1129 01:38:23.004007 137274321021824 utils.py:1231] [15550] uptime = 100092.366368926
+I1129 01:38:23.004058 137274321021824 utils.py:1231] [15550] examples_seen = 15923200.0
+I1129 01:38:23.004106 137274321021824 utils.py:1231] [15550] progress = 0.13809578785645144
+I1129 01:38:23.004153 137274321021824 utils.py:1231] [15550] epoch = 12.42866854984557
+I1129 01:38:23.004203 137274321021824 utils.py:1231] [15550] img/sec/core = 167.2516446011043
+I1129 01:38:23.004259 137274321021824 utils.py:1231] [15550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.769117500122775
+I1129 01:38:23.004308 137274321021824 utils.py:1231] [15550] core_hours = 27.769117500122775
+I1129 01:38:23.004367 137274321021824 train.py:125] NOTE: Steps:15550/112603 [13.8%]
+Walltime:1d3h48m (0s eval)
+ETA:7d5h20m
+Total train time:8d9h6m
+I1129 01:43:30.939528 137274321021824 utils.py:1231] [15600] l2_params = 312.3427753585944
+I1129 01:43:30.939805 137274321021824 utils.py:1231] [15600] train/loss = 3.5666694343090057
+I1129 01:43:30.940037 137274321021824 utils.py:1231] [15600] l2_grads = 1.1587508916854858
+I1129 01:43:30.940149 137274321021824 utils.py:1231] [15600] lr = 0.000992670461204108
+I1129 01:43:30.940212 137274321021824 utils.py:1231] [15600] uptime = 100400.30257277701
+I1129 01:43:30.940292 137274321021824 utils.py:1231] [15600] examples_seen = 15974400.0
+I1129 01:43:30.940344 137274321021824 utils.py:1231] [15600] progress = 0.13853982575952684
+I1129 01:43:30.940396 137274321021824 utils.py:1231] [15600] epoch = 12.468632114314527
+I1129 01:43:30.940446 137274321021824 utils.py:1231] [15600] img/sec/core = 166.26820542598952
+I1129 01:43:30.940502 137274321021824 utils.py:1231] [15600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.854655334525834
+I1129 01:43:30.940552 137274321021824 utils.py:1231] [15600] core_hours = 27.854655334525834
+I1129 01:43:30.940611 137274321021824 train.py:125] NOTE: Steps:15600/112603 [13.9%]
+Walltime:1d3h53m (0s eval)
+ETA:7d5h13m
+Total train time:8d9h5m
+I1129 01:48:42.702056 137274321021824 utils.py:1231] [15650] l2_params = 312.5744710043646
+I1129 01:48:42.702266 137274321021824 utils.py:1231] [15650] train/loss = 5.715206444263458
+I1129 01:48:42.702356 137274321021824 utils.py:1231] [15650] l2_grads = 0.9037434458732605
+I1129 01:48:42.702418 137274321021824 utils.py:1231] [15650] lr = 0.000992539296671535
+I1129 01:48:42.702465 137274321021824 utils.py:1231] [15650] uptime = 100712.064827565
+I1129 01:48:42.702512 137274321021824 utils.py:1231] [15650] examples_seen = 16025600.0
+I1129 01:48:42.702558 137274321021824 utils.py:1231] [15650] progress = 0.13898386366260224
+I1129 01:48:42.702604 137274321021824 utils.py:1231] [15650] epoch = 12.508595678783484
+I1129 01:48:42.702650 137274321021824 utils.py:1231] [15650] img/sec/core = 164.2277062526968
+I1129 01:48:42.702705 137274321021824 utils.py:1231] [15650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 27.94125596085583
+I1129 01:48:42.702757 137274321021824 utils.py:1231] [15650] core_hours = 27.94125596085583
+I1129 01:48:42.702812 137274321021824 train.py:125] NOTE: Steps:15650/112603 [13.9%]
+Walltime:1d3h58m (0s eval)
+ETA:7d5h7m
+Total train time:8d9h3m
+I1129 01:53:50.102831 137274321021824 utils.py:1231] [15700] l2_params = 312.7945833302778
+I1129 01:53:50.103084 137274321021824 utils.py:1231] [15700] train/loss = 3.63117179274559
+I1129 01:53:50.103215 137274321021824 utils.py:1231] [15700] l2_grads = 1.379328727722168
+I1129 01:53:50.103288 137274321021824 utils.py:1231] [15700] lr = 0.0009924069777280152
+I1129 01:53:50.103355 137274321021824 utils.py:1231] [15700] uptime = 101019.46571404701
+I1129 01:53:50.103405 137274321021824 utils.py:1231] [15700] examples_seen = 16076800.0
+I1129 01:53:50.103452 137274321021824 utils.py:1231] [15700] progress = 0.13942790156567764
+I1129 01:53:50.103498 137274321021824 utils.py:1231] [15700] epoch = 12.548559243252441
+I1129 01:53:50.103548 137274321021824 utils.py:1231] [15700] img/sec/core = 166.55774999854142
+I1129 01:53:50.103603 137274321021824 utils.py:1231] [15700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.026645095989725
+I1129 01:53:50.103651 137274321021824 utils.py:1231] [15700] core_hours = 28.026645095989725
+I1129 01:53:50.103709 137274321021824 train.py:125] NOTE: Steps:15700/112603 [13.9%]
+Walltime:1d4h3m (0s eval)
+ETA:7d5h0m
+Total train time:8d9h2m
+I1129 01:58:55.744061 137274321021824 utils.py:1231] [15750] l2_params = 313.0644521058371
+I1129 01:58:55.744276 137274321021824 utils.py:1231] [15750] train/loss = 3.3524648547172546
+I1129 01:58:55.744398 137274321021824 utils.py:1231] [15750] l2_grads = 1.189744234085083
+I1129 01:58:55.744476 137274321021824 utils.py:1231] [15750] lr = 0.0009922735046836758
+I1129 01:58:55.744547 137274321021824 utils.py:1231] [15750] uptime = 101325.106905548
+I1129 01:58:55.744622 137274321021824 utils.py:1231] [15750] examples_seen = 16128000.0
+I1129 01:58:55.744682 137274321021824 utils.py:1231] [15750] progress = 0.13987193946875306
+I1129 01:58:55.744741 137274321021824 utils.py:1231] [15750] epoch = 12.588522807721398
+I1129 01:58:55.744801 137274321021824 utils.py:1231] [15750] img/sec/core = 167.51668761843965
+I1129 01:58:55.744873 137274321021824 utils.py:1231] [15750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.11154542696222
+I1129 01:58:55.744953 137274321021824 utils.py:1231] [15750] core_hours = 28.11154542696222
+I1129 01:58:55.745052 137274321021824 train.py:125] NOTE: Steps:15750/112603 [14.0%]
+Walltime:1d4h8m (0s eval)
+ETA:7d4h53m
+Total train time:8d9h0m
+I1129 02:04:03.923583 137274321021824 utils.py:1231] [15800] l2_params = 313.27655894836477
+I1129 02:04:03.923803 137274321021824 utils.py:1231] [15800] train/loss = 5.27291476726532
+I1129 02:04:03.923912 137274321021824 utils.py:1231] [15800] l2_grads = 0.9498845934867859
+I1129 02:04:03.923989 137274321021824 utils.py:1231] [15800] lr = 0.0009921388778513503
+I1129 02:04:03.924048 137274321021824 utils.py:1231] [15800] uptime = 101633.28640893802
+I1129 02:04:03.924114 137274321021824 utils.py:1231] [15800] examples_seen = 16179200.0
+I1129 02:04:03.924183 137274321021824 utils.py:1231] [15800] progress = 0.14031597737182847
+I1129 02:04:03.924244 137274321021824 utils.py:1231] [15800] epoch = 12.628486372190354
+I1129 02:04:03.924303 137274321021824 utils.py:1231] [15800] img/sec/core = 166.13694109048035
+I1129 02:04:03.924373 137274321021824 utils.py:1231] [15800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.197150844570558
+I1129 02:04:03.924435 137274321021824 utils.py:1231] [15800] core_hours = 28.197150844570558
+I1129 02:04:03.924508 137274321021824 train.py:125] NOTE: Steps:15800/112603 [14.0%]
+Walltime:1d4h13m (0s eval)
+ETA:7d4h46m
+Total train time:8d8h58m
+I1129 02:09:10.244450 137274321021824 utils.py:1231] [15850] l2_params = 313.56084507790996
+I1129 02:09:10.244752 137274321021824 utils.py:1231] [15850] train/loss = 3.4783254861831665
+I1129 02:09:10.244927 137274321021824 utils.py:1231] [15850] l2_grads = 1.3942322731018066
+I1129 02:09:10.244991 137274321021824 utils.py:1231] [15850] lr = 0.0009920030975465765
+I1129 02:09:10.245042 137274321021824 utils.py:1231] [15850] uptime = 101939.607404408
+I1129 02:09:10.245115 137274321021824 utils.py:1231] [15850] examples_seen = 16230400.0
+I1129 02:09:10.245163 137274321021824 utils.py:1231] [15850] progress = 0.14076001527490387
+I1129 02:09:10.245210 137274321021824 utils.py:1231] [15850] epoch = 12.66844993665931
+I1129 02:09:10.245260 137274321021824 utils.py:1231] [15850] img/sec/core = 167.14492560800275
+I1129 02:09:10.245315 137274321021824 utils.py:1231] [15850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.282240009978885
+I1129 02:09:10.245363 137274321021824 utils.py:1231] [15850] core_hours = 28.282240009978885
+I1129 02:09:10.245422 137274321021824 train.py:125] NOTE: Steps:15850/112603 [14.1%]
+Walltime:1d4h18m (0s eval)
+ETA:7d4h39m
+Total train time:8d8h57m
+I1129 02:14:13.494946 137274321021824 utils.py:1231] [15900] l2_params = 313.7658586152619
+I1129 02:14:13.495218 137274321021824 utils.py:1231] [15900] train/loss = 3.3640577495098114
+I1129 02:14:13.495345 137274321021824 utils.py:1231] [15900] l2_grads = 1.45441734790802
+I1129 02:14:13.495433 137274321021824 utils.py:1231] [15900] lr = 0.0009918661640875956
+I1129 02:14:13.495491 137274321021824 utils.py:1231] [15900] uptime = 102242.85785289701
+I1129 02:14:13.495558 137274321021824 utils.py:1231] [15900] examples_seen = 16281600.0
+I1129 02:14:13.495622 137274321021824 utils.py:1231] [15900] progress = 0.14120405317797927
+I1129 02:14:13.495687 137274321021824 utils.py:1231] [15900] epoch = 12.708413501128268
+I1129 02:14:13.495747 137274321021824 utils.py:1231] [15900] img/sec/core = 168.83734304470696
+I1129 02:14:13.495807 137274321021824 utils.py:1231] [15900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.366476245670277
+I1129 02:14:13.495858 137274321021824 utils.py:1231] [15900] core_hours = 28.366476245670277
+I1129 02:14:13.495953 137274321021824 train.py:125] NOTE: Steps:15900/112603 [14.1%]
+Walltime:1d4h24m (0s eval)
+ETA:7d4h32m
+Total train time:8d8h54m
+I1129 02:19:15.256669 137274321021824 utils.py:1231] [15950] l2_params = 313.90860591505634
+I1129 02:19:15.256943 137274321021824 utils.py:1231] [15950] train/loss = 3.358236610889435
+I1129 02:19:15.257065 137274321021824 utils.py:1231] [15950] l2_grads = 1.324599266052246
+I1129 02:19:15.257143 137274321021824 utils.py:1231] [15950] lr = 0.0009917280777953523
+I1129 02:19:15.257207 137274321021824 utils.py:1231] [15950] uptime = 102544.61956843799
+I1129 02:19:15.257267 137274321021824 utils.py:1231] [15950] examples_seen = 16332800.0
+I1129 02:19:15.257323 137274321021824 utils.py:1231] [15950] progress = 0.14164809108105467
+I1129 02:19:15.257377 137274321021824 utils.py:1231] [15950] epoch = 12.748377065597225
+I1129 02:19:15.257451 137274321021824 utils.py:1231] [15950] img/sec/core = 169.67029733447575
+I1129 02:19:15.257513 137274321021824 utils.py:1231] [15950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.450298944431662
+I1129 02:19:15.257569 137274321021824 utils.py:1231] [15950] core_hours = 28.450298944431662
+I1129 02:19:15.257634 137274321021824 train.py:125] NOTE: Steps:15950/112603 [14.2%]
+Walltime:1d4h29m (0s eval)
+ETA:7d4h25m
+Total train time:8d8h52m
+I1129 02:24:21.979004 137274321021824 utils.py:1231] [16000] l2_params = 314.09909304267313
+I1129 02:24:21.979353 137274321021824 utils.py:1231] [16000] train/loss = 3.4163320064544678
+I1129 02:24:21.979556 137274321021824 utils.py:1231] [16000] l2_grads = 1.252766728401184
+I1129 02:24:21.979638 137274321021824 utils.py:1231] [16000] lr = 0.0009915888389934913
+I1129 02:24:21.979721 137274321021824 utils.py:1231] [16000] uptime = 102851.34208172
+I1129 02:24:21.979784 137274321021824 utils.py:1231] [16000] examples_seen = 16384000.0
+I1129 02:24:21.979844 137274321021824 utils.py:1231] [16000] progress = 0.1420921289841301
+I1129 02:24:21.979906 137274321021824 utils.py:1231] [16000] epoch = 12.788340630066182
+I1129 02:24:21.979958 137274321021824 utils.py:1231] [16000] img/sec/core = 166.92612306852894
+I1129 02:24:21.980014 137274321021824 utils.py:1231] [16000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.53549964256555
+I1129 02:24:21.980065 137274321021824 utils.py:1231] [16000] core_hours = 28.53549964256555
+I1129 02:24:21.980127 137274321021824 train.py:125] NOTE: Steps:16000/112603 [14.2%]
+Walltime:1d4h34m (0s eval)
+ETA:7d4h18m
+Total train time:8d8h50m
+I1129 02:29:29.801501 137274321021824 utils.py:1231] [16050] l2_params = 314.3202345901107
+I1129 02:29:29.801710 137274321021824 utils.py:1231] [16050] train/loss = 3.3930163979530334
+I1129 02:29:29.801805 137274321021824 utils.py:1231] [16050] l2_grads = 1.3853121995925903
+I1129 02:29:29.801863 137274321021824 utils.py:1231] [16050] lr = 0.000991448448008361
+I1129 02:29:29.801920 137274321021824 utils.py:1231] [16050] uptime = 103159.16428202299
+I1129 02:29:29.801971 137274321021824 utils.py:1231] [16050] examples_seen = 16435200.0
+I1129 02:29:29.802036 137274321021824 utils.py:1231] [16050] progress = 0.1425361668872055
+I1129 02:29:29.802084 137274321021824 utils.py:1231] [16050] epoch = 12.82830419453514
+I1129 02:29:29.802131 137274321021824 utils.py:1231] [16050] img/sec/core = 166.32978371801286
+I1129 02:29:29.802187 137274321021824 utils.py:1231] [16050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.621005809316383
+I1129 02:29:29.802235 137274321021824 utils.py:1231] [16050] core_hours = 28.621005809316383
+I1129 02:29:29.802294 137274321021824 train.py:125] NOTE: Steps:16050/112603 [14.3%]
+Walltime:1d4h39m (0s eval)
+ETA:7d4h11m
+Total train time:8d8h49m
+I1129 02:34:38.286020 137274321021824 utils.py:1231] [16100] l2_params = 314.5159806024896
+I1129 02:34:38.286224 137274321021824 utils.py:1231] [16100] train/loss = 3.519822120666504
+I1129 02:34:38.286318 137274321021824 utils.py:1231] [16100] l2_grads = 1.2945350408554077
+I1129 02:34:38.286383 137274321021824 utils.py:1231] [16100] lr = 0.0009913069051690076
+I1129 02:34:38.286441 137274321021824 utils.py:1231] [16100] uptime = 103467.648802772
+I1129 02:34:38.286497 137274321021824 utils.py:1231] [16100] examples_seen = 16486400.0
+I1129 02:34:38.286549 137274321021824 utils.py:1231] [16100] progress = 0.1429802047902809
+I1129 02:34:38.286598 137274321021824 utils.py:1231] [16100] epoch = 12.868267759004096
+I1129 02:34:38.286668 137274321021824 utils.py:1231] [16100] img/sec/core = 165.97267141859854
+I1129 02:34:38.286747 137274321021824 utils.py:1231] [16100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.70669595396889
+I1129 02:34:38.286796 137274321021824 utils.py:1231] [16100] core_hours = 28.70669595396889
+I1129 02:34:38.286871 137274321021824 train.py:125] NOTE: Steps:16100/112603 [14.3%]
+Walltime:1d4h44m (0s eval)
+ETA:7d4h5m
+Total train time:8d8h47m
+I1129 02:39:48.623704 137274321021824 utils.py:1231] [16150] l2_params = 314.70257556227716
+I1129 02:39:48.623903 137274321021824 utils.py:1231] [16150] train/loss = 5.504897654056549
+I1129 02:39:48.624009 137274321021824 utils.py:1231] [16150] l2_grads = 0.963716447353363
+I1129 02:39:48.624069 137274321021824 utils.py:1231] [16150] lr = 0.0009911642108071806
+I1129 02:39:48.624120 137274321021824 utils.py:1231] [16150] uptime = 103777.986482389
+I1129 02:39:48.624171 137274321021824 utils.py:1231] [16150] examples_seen = 16537600.0
+I1129 02:39:48.624220 137274321021824 utils.py:1231] [16150] progress = 0.1434242426933563
+I1129 02:39:48.624268 137274321021824 utils.py:1231] [16150] epoch = 12.908231323473052
+I1129 02:39:48.624317 137274321021824 utils.py:1231] [16150] img/sec/core = 164.9815776904355
+I1129 02:39:48.624372 137274321021824 utils.py:1231] [16150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.792900864973607
+I1129 02:39:48.624420 137274321021824 utils.py:1231] [16150] core_hours = 28.792900864973607
+I1129 02:39:48.624479 137274321021824 train.py:125] NOTE: Steps:16150/112603 [14.3%]
+Walltime:1d4h49m (0s eval)
+ETA:7d3h58m
+Total train time:8d8h46m
+I1129 02:44:57.719712 137274321021824 utils.py:1231] [16200] l2_params = 314.82202969581715
+I1129 02:44:57.719985 137274321021824 utils.py:1231] [16200] train/loss = 3.355706602334976
+I1129 02:44:57.720164 137274321021824 utils.py:1231] [16200] l2_grads = 1.3242958784103394
+I1129 02:44:57.720252 137274321021824 utils.py:1231] [16200] lr = 0.000991020365257324
+I1129 02:44:57.720314 137274321021824 utils.py:1231] [16200] uptime = 104087.082673015
+I1129 02:44:57.720377 137274321021824 utils.py:1231] [16200] examples_seen = 16588800.0
+I1129 02:44:57.720430 137274321021824 utils.py:1231] [16200] progress = 0.14386828059643172
+I1129 02:44:57.720498 137274321021824 utils.py:1231] [16200] epoch = 12.948194887942009
+I1129 02:44:57.720551 137274321021824 utils.py:1231] [16200] img/sec/core = 165.644228407687
+I1129 02:44:57.720607 137274321021824 utils.py:1231] [16200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.878760917925273
+I1129 02:44:57.720666 137274321021824 utils.py:1231] [16200] core_hours = 28.878760917925273
+I1129 02:44:57.720726 137274321021824 train.py:125] NOTE: Steps:16200/112603 [14.4%]
+Walltime:1d4h54m (0s eval)
+ETA:7d3h52m
+Total train time:8d8h45m
+I1129 02:50:04.565563 137274321021824 utils.py:1231] [16250] l2_params = 315.0279934750939
+I1129 02:50:04.565761 137274321021824 utils.py:1231] [16250] train/loss = 5.566577136516571
+I1129 02:50:04.565862 137274321021824 utils.py:1231] [16250] l2_grads = 0.9092855453491211
+I1129 02:50:04.565936 137274321021824 utils.py:1231] [16250] lr = 0.0009908753688565829
+I1129 02:50:04.565993 137274321021824 utils.py:1231] [16250] uptime = 104393.92835481201
+I1129 02:50:04.566048 137274321021824 utils.py:1231] [16250] examples_seen = 16640000.0
+I1129 02:50:04.566105 137274321021824 utils.py:1231] [16250] progress = 0.14431231849950713
+I1129 02:50:04.566160 137274321021824 utils.py:1231] [16250] epoch = 12.988158452410966
+I1129 02:50:04.566222 137274321021824 utils.py:1231] [16250] img/sec/core = 166.85911856458836
+I1129 02:50:04.566286 137274321021824 utils.py:1231] [16250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 28.96399582953556
+I1129 02:50:04.566345 137274321021824 utils.py:1231] [16250] core_hours = 28.96399582953556
+I1129 02:50:04.566411 137274321021824 train.py:125] NOTE: Steps:16250/112603 [14.4%]
+Walltime:1d4h59m (0s eval)
+ETA:7d3h45m
+Total train time:8d8h43m
+I1129 02:55:16.346104 137274321021824 utils.py:1231] [16300] l2_params = 315.2501405059364
+I1129 02:55:16.346410 137274321021824 utils.py:1231] [16300] train/loss = 3.3803176283836365
+I1129 02:55:16.346565 137274321021824 utils.py:1231] [16300] l2_grads = 1.4234602451324463
+I1129 02:55:16.346650 137274321021824 utils.py:1231] [16300] lr = 0.0009907292219447994
+I1129 02:55:16.346725 137274321021824 utils.py:1231] [16300] uptime = 104705.709080758
+I1129 02:55:16.346779 137274321021824 utils.py:1231] [16300] examples_seen = 16691200.0
+I1129 02:55:16.346840 137274321021824 utils.py:1231] [16300] progress = 0.14475635640258253
+I1129 02:55:16.346901 137274321021824 utils.py:1231] [16300] epoch = 13.028122016879923
+I1129 02:55:16.346976 137274321021824 utils.py:1231] [16300] img/sec/core = 164.21797673558234
+I1129 02:55:16.347033 137274321021824 utils.py:1231] [16300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.050601586742776
+I1129 02:55:16.347086 137274321021824 utils.py:1231] [16300] core_hours = 29.050601586742776
+I1129 02:55:16.347146 137274321021824 train.py:125] NOTE: Steps:16300/112603 [14.5%]
+Walltime:1d5h5m (0s eval)
+ETA:7d3h39m
+Total train time:8d8h42m
+I1129 03:00:22.194376 137274321021824 utils.py:1231] [16350] l2_params = 315.4530566805577
+I1129 03:00:22.194587 137274321021824 utils.py:1231] [16350] train/loss = 3.2801783680915833
+I1129 03:00:22.194692 137274321021824 utils.py:1231] [16350] l2_grads = 1.2387765645980835
+I1129 03:00:22.194754 137274321021824 utils.py:1231] [16350] lr = 0.0009905819248645124
+I1129 03:00:22.194810 137274321021824 utils.py:1231] [16350] uptime = 105011.557171557
+I1129 03:00:22.194871 137274321021824 utils.py:1231] [16350] examples_seen = 16742400.0
+I1129 03:00:22.194932 137274321021824 utils.py:1231] [16350] progress = 0.14520039430565793
+I1129 03:00:22.194981 137274321021824 utils.py:1231] [16350] epoch = 13.06808558134888
+I1129 03:00:22.195032 137274321021824 utils.py:1231] [16350] img/sec/core = 167.40336637787883
+I1129 03:00:22.195086 137274321021824 utils.py:1231] [16350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.135559389742493
+I1129 03:00:22.195136 137274321021824 utils.py:1231] [16350] core_hours = 29.135559389742493
+I1129 03:00:22.195196 137274321021824 train.py:125] NOTE: Steps:16350/112603 [14.5%]
+Walltime:1d5h10m (0s eval)
+ETA:7d3h32m
+Total train time:8d8h40m
+I1129 03:05:27.139619 137274321021824 utils.py:1231] [16400] l2_params = 315.66991216264273
+I1129 03:05:27.139866 137274321021824 utils.py:1231] [16400] train/loss = 3.389899343252182
+I1129 03:05:27.139997 137274321021824 utils.py:1231] [16400] l2_grads = 1.487139344215393
+I1129 03:05:27.140086 137274321021824 utils.py:1231] [16400] lr = 0.0009904334779609547
+I1129 03:05:27.140153 137274321021824 utils.py:1231] [16400] uptime = 105316.502514445
+I1129 03:05:27.140238 137274321021824 utils.py:1231] [16400] examples_seen = 16793600.0
+I1129 03:05:27.140307 137274321021824 utils.py:1231] [16400] progress = 0.14564443220873333
+I1129 03:05:27.140380 137274321021824 utils.py:1231] [16400] epoch = 13.108049145817837
+I1129 03:05:27.140440 137274321021824 utils.py:1231] [16400] img/sec/core = 167.89894056130368
+I1129 03:05:27.140499 137274321021824 utils.py:1231] [16400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.220266429433607
+I1129 03:05:27.140554 137274321021824 utils.py:1231] [16400] core_hours = 29.220266429433607
+I1129 03:05:27.140619 137274321021824 train.py:125] NOTE: Steps:16400/112603 [14.6%]
+Walltime:1d5h15m (0s eval)
+ETA:7d3h25m
+Total train time:8d8h39m
+I1129 03:10:34.523647 137274321021824 utils.py:1231] [16450] l2_params = 315.89345306288715
+I1129 03:10:34.523856 137274321021824 utils.py:1231] [16450] train/loss = 3.339263379573822
+I1129 03:10:34.523962 137274321021824 utils.py:1231] [16450] l2_grads = 1.2635157108306885
+I1129 03:10:34.524035 137274321021824 utils.py:1231] [16450] lr = 0.0009902838815820574
+I1129 03:10:34.524119 137274321021824 utils.py:1231] [16450] uptime = 105623.886475538
+I1129 03:10:34.524179 137274321021824 utils.py:1231] [16450] examples_seen = 16844800.0
+I1129 03:10:34.524235 137274321021824 utils.py:1231] [16450] progress = 0.14608847011180875
+I1129 03:10:34.524289 137274321021824 utils.py:1231] [16450] epoch = 13.148012710286793
+I1129 03:10:34.524352 137274321021824 utils.py:1231] [16450] img/sec/core = 166.56692111697114
+I1129 03:10:34.524426 137274321021824 utils.py:1231] [16450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.305650863070557
+I1129 03:10:34.524490 137274321021824 utils.py:1231] [16450] core_hours = 29.305650863070557
+I1129 03:10:34.524571 137274321021824 train.py:125] NOTE: Steps:16450/112603 [14.6%]
+Walltime:1d5h20m (0s eval)
+ETA:7d3h19m
+Total train time:8d8h37m
+I1129 03:15:34.137668 137274321021824 utils.py:1231] [16500] l2_params = 316.05822474440726
+I1129 03:15:34.137917 137274321021824 utils.py:1231] [16500] train/loss = 3.3589644134044647
+I1129 03:15:34.138060 137274321021824 utils.py:1231] [16500] l2_grads = 1.348484992980957
+I1129 03:15:34.138170 137274321021824 utils.py:1231] [16500] lr = 0.0009901331360784409
+I1129 03:15:34.138234 137274321021824 utils.py:1231] [16500] uptime = 105923.500595615
+I1129 03:15:34.138307 137274321021824 utils.py:1231] [16500] examples_seen = 16896000.0
+I1129 03:15:34.138371 137274321021824 utils.py:1231] [16500] progress = 0.14653250801488416
+I1129 03:15:34.138432 137274321021824 utils.py:1231] [16500] epoch = 13.18797627475575
+I1129 03:15:34.138504 137274321021824 utils.py:1231] [16500] img/sec/core = 170.88647219577973
+I1129 03:15:34.138584 137274321021824 utils.py:1231] [16500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.388877007536387
+I1129 03:15:34.138638 137274321021824 utils.py:1231] [16500] core_hours = 29.388877007536387
+I1129 03:15:34.138703 137274321021824 train.py:125] NOTE: Steps:16500/112603 [14.7%]
+Walltime:1d5h25m (0s eval)
+ETA:7d3h11m
+Total train time:8d8h35m
+I1129 03:20:35.720903 137274321021824 utils.py:1231] [16550] l2_params = 316.22816957706766
+I1129 03:20:35.721148 137274321021824 utils.py:1231] [16550] train/loss = 4.21619188785553
+I1129 03:20:35.721297 137274321021824 utils.py:1231] [16550] l2_grads = 1.1552009582519531
+I1129 03:20:35.721397 137274321021824 utils.py:1231] [16550] lr = 0.0009899812418034237
+I1129 03:20:35.721451 137274321021824 utils.py:1231] [16550] uptime = 106225.08381375
+I1129 03:20:35.721524 137274321021824 utils.py:1231] [16550] examples_seen = 16947200.0
+I1129 03:20:35.721587 137274321021824 utils.py:1231] [16550] progress = 0.14697654591795956
+I1129 03:20:35.721638 137274321021824 utils.py:1231] [16550] epoch = 13.227939839224707
+I1129 03:20:35.721691 137274321021824 utils.py:1231] [16550] img/sec/core = 169.77071972579492
+I1129 03:20:35.721749 137274321021824 utils.py:1231] [16550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.472650123684996
+I1129 03:20:35.721801 137274321021824 utils.py:1231] [16550] core_hours = 29.472650123684996
+I1129 03:20:35.721862 137274321021824 train.py:125] NOTE: Steps:16550/112603 [14.7%]
+Walltime:1d5h30m (0s eval)
+ETA:7d3h4m
+Total train time:8d8h33m
+I1129 03:25:37.943953 137274321021824 utils.py:1231] [16600] l2_params = 316.39598936883294
+I1129 03:25:37.944219 137274321021824 utils.py:1231] [16600] train/loss = 3.389721155166626
+I1129 03:25:37.944317 137274321021824 utils.py:1231] [16600] l2_grads = 1.1944913864135742
+I1129 03:25:37.944398 137274321021824 utils.py:1231] [16600] lr = 0.0009898281991130135
+I1129 03:25:37.944453 137274321021824 utils.py:1231] [16600] uptime = 106527.30681551101
+I1129 03:25:37.944507 137274321021824 utils.py:1231] [16600] examples_seen = 16998400.0
+I1129 03:25:37.944558 137274321021824 utils.py:1231] [16600] progress = 0.14742058382103496
+I1129 03:25:37.944621 137274321021824 utils.py:1231] [16600] epoch = 13.267903403693664
+I1129 03:25:37.944672 137274321021824 utils.py:1231] [16600] img/sec/core = 169.41132773370822
+I1129 03:25:37.944728 137274321021824 utils.py:1231] [16600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.5566009575075
+I1129 03:25:37.944780 137274321021824 utils.py:1231] [16600] core_hours = 29.5566009575075
+I1129 03:25:37.944841 137274321021824 train.py:125] NOTE: Steps:16600/112603 [14.7%]
+Walltime:1d5h35m (0s eval)
+ETA:7d2h57m
+Total train time:8d8h30m
+I1129 03:30:49.379411 137274321021824 utils.py:1231] [16650] l2_params = 316.64363665773266
+I1129 03:30:49.379655 137274321021824 utils.py:1231] [16650] train/loss = 3.300117999315262
+I1129 03:30:49.379786 137274321021824 utils.py:1231] [16650] l2_grads = 1.3117649555206299
+I1129 03:30:49.379852 137274321021824 utils.py:1231] [16650] lr = 0.0009896740083659115
+I1129 03:30:49.379918 137274321021824 utils.py:1231] [16650] uptime = 106838.74227957001
+I1129 03:30:49.379969 137274321021824 utils.py:1231] [16650] examples_seen = 17049600.0
+I1129 03:30:49.380016 137274321021824 utils.py:1231] [16650] progress = 0.14786462172411036
+I1129 03:30:49.380063 137274321021824 utils.py:1231] [16650] epoch = 13.307866968162621
+I1129 03:30:49.380113 137274321021824 utils.py:1231] [16650] img/sec/core = 164.40003117403467
+I1129 03:30:49.380167 137274321021824 utils.py:1231] [16650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.643110808635
+I1129 03:30:49.380218 137274321021824 utils.py:1231] [16650] core_hours = 29.643110808635
+I1129 03:30:49.380278 137274321021824 train.py:125] NOTE: Steps:16650/112603 [14.8%]
+Walltime:1d5h40m (0s eval)
+ETA:7d2h51m
+Total train time:8d8h29m
+I1129 03:35:52.440434 137274321021824 utils.py:1231] [16700] l2_params = 316.83349996326837
+I1129 03:35:52.440709 137274321021824 utils.py:1231] [16700] train/loss = 3.4688801765441895
+I1129 03:35:52.440832 137274321021824 utils.py:1231] [16700] l2_grads = 1.3007616996765137
+I1129 03:35:52.440918 137274321021824 utils.py:1231] [16700] lr = 0.0009895186699235097
+I1129 03:35:52.440983 137274321021824 utils.py:1231] [16700] uptime = 107141.80334445501
+I1129 03:35:52.441034 137274321021824 utils.py:1231] [16700] examples_seen = 17100800.0
+I1129 03:35:52.441081 137274321021824 utils.py:1231] [16700] progress = 0.14830865962718578
+I1129 03:35:52.441129 137274321021824 utils.py:1231] [16700] epoch = 13.347830532631578
+I1129 03:35:52.441180 137274321021824 utils.py:1231] [16700] img/sec/core = 168.94284991517728
+I1129 03:35:52.441232 137274321021824 utils.py:1231] [16700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.72729443776972
+I1129 03:35:52.441281 137274321021824 utils.py:1231] [16700] core_hours = 29.72729443776972
+I1129 03:35:52.441338 137274321021824 train.py:125] NOTE: Steps:16700/112603 [14.8%]
+Walltime:1d5h45m (0s eval)
+ETA:7d2h44m
+Total train time:8d8h27m
+I1129 03:41:04.211247 137274321021824 utils.py:1231] [16750] l2_params = 317.01751595915135
+I1129 03:41:04.211503 137274321021824 utils.py:1231] [16750] train/loss = 3.487646371126175
+I1129 03:41:04.211606 137274321021824 utils.py:1231] [16750] l2_grads = 1.2336912155151367
+I1129 03:41:04.211694 137274321021824 utils.py:1231] [16750] lr = 0.0009893621841498893
+I1129 03:41:04.211759 137274321021824 utils.py:1231] [16750] uptime = 107453.57411793202
+I1129 03:41:04.211815 137274321021824 utils.py:1231] [16750] examples_seen = 17152000.0
+I1129 03:41:04.211869 137274321021824 utils.py:1231] [16750] progress = 0.14875269753026119
+I1129 03:41:04.211932 137274321021824 utils.py:1231] [16750] epoch = 13.387794097100533
+I1129 03:41:04.211986 137274321021824 utils.py:1231] [16750] img/sec/core = 164.22321896627608
+I1129 03:41:04.212044 137274321021824 utils.py:1231] [16750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.813897430402225
+I1129 03:41:04.212091 137274321021824 utils.py:1231] [16750] core_hours = 29.813897430402225
+I1129 03:41:04.212178 137274321021824 train.py:125] NOTE: Steps:16750/112603 [14.9%]
+Walltime:1d5h50m (0s eval)
+ETA:7d2h37m
+Total train time:8d8h26m
+I1129 03:46:16.006887 137274321021824 utils.py:1231] [16800] l2_params = 317.26758389548075
+I1129 03:46:16.007186 137274321021824 utils.py:1231] [16800] train/loss = 5.770748794078827
+I1129 03:46:16.007372 137274321021824 utils.py:1231] [16800] l2_grads = 1.0397971868515015
+I1129 03:46:16.007485 137274321021824 utils.py:1231] [16800] lr = 0.0009892045514118205
+I1129 03:46:16.007576 137274321021824 utils.py:1231] [16800] uptime = 107765.36993497702
+I1129 03:46:16.007662 137274321021824 utils.py:1231] [16800] examples_seen = 17203200.0
+I1129 03:46:16.007784 137274321021824 utils.py:1231] [16800] progress = 0.14919673543333659
+I1129 03:46:16.007898 137274321021824 utils.py:1231] [16800] epoch = 13.42775766156949
+I1129 03:46:16.007977 137274321021824 utils.py:1231] [16800] img/sec/core = 164.21002848992848
+I1129 03:46:16.008066 137274321021824 utils.py:1231] [16800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.90050737958139
+I1129 03:46:16.008138 137274321021824 utils.py:1231] [16800] core_hours = 29.90050737958139
+I1129 03:46:16.008234 137274321021824 train.py:125] NOTE: Steps:16800/112603 [14.9%]
+Walltime:1d5h56m (0s eval)
+ETA:7d2h31m
+Total train time:8d8h26m
+I1129 03:51:27.782218 137274321021824 utils.py:1231] [16850] l2_params = 317.4366089515813
+I1129 03:51:27.782475 137274321021824 utils.py:1231] [16850] train/loss = 3.216576784849167
+I1129 03:51:27.782584 137274321021824 utils.py:1231] [16850] l2_grads = 1.2194390296936035
+I1129 03:51:27.782664 137274321021824 utils.py:1231] [16850] lr = 0.0009890457720787625
+I1129 03:51:27.782728 137274321021824 utils.py:1231] [16850] uptime = 108077.14508915199
+I1129 03:51:27.782794 137274321021824 utils.py:1231] [16850] examples_seen = 17254400.0
+I1129 03:51:27.782855 137274321021824 utils.py:1231] [16850] progress = 0.149640773336412
+I1129 03:51:27.782923 137274321021824 utils.py:1231] [16850] epoch = 13.467721226038448
+I1129 03:51:27.782987 137274321021824 utils.py:1231] [16850] img/sec/core = 164.2209114946519
+I1129 03:51:27.783050 137274321021824 utils.py:1231] [16850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 29.98711158907444
+I1129 03:51:27.783112 137274321021824 utils.py:1231] [16850] core_hours = 29.98711158907444
+I1129 03:51:27.783187 137274321021824 train.py:125] NOTE: Steps:16850/112603 [15.0%]
+Walltime:1d6h1m (0s eval)
+ETA:7d2h25m
+Total train time:8d8h25m
+I1129 03:56:37.460857 137274321021824 utils.py:1231] [16900] l2_params = 317.6133842418364
+I1129 03:56:37.461063 137274321021824 utils.py:1231] [16900] train/loss = 3.5760781168937683
+I1129 03:56:37.461167 137274321021824 utils.py:1231] [16900] l2_grads = 1.1909781694412231
+I1129 03:56:37.461252 137274321021824 utils.py:1231] [16900] lr = 0.0009888858465228604
+I1129 03:56:37.461341 137274321021824 utils.py:1231] [16900] uptime = 108386.823696155
+I1129 03:56:37.461432 137274321021824 utils.py:1231] [16900] examples_seen = 17305600.0
+I1129 03:56:37.461493 137274321021824 utils.py:1231] [16900] progress = 0.15008481123948741
+I1129 03:56:37.461569 137274321021824 utils.py:1231] [16900] epoch = 13.507684790507405
+I1129 03:56:37.461646 137274321021824 utils.py:1231] [16900] img/sec/core = 165.3326992636043
+I1129 03:56:37.461721 137274321021824 utils.py:1231] [16900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.073133424353053
+I1129 03:56:37.461773 137274321021824 utils.py:1231] [16900] core_hours = 30.073133424353053
+I1129 03:56:37.461837 137274321021824 train.py:125] NOTE: Steps:16900/112603 [15.0%]
+Walltime:1d6h6m (0s eval)
+ETA:7d2h19m
+Total train time:8d8h23m
+I1129 04:01:49.236423 137274321021824 utils.py:1231] [16950] l2_params = 317.7746242222494
+I1129 04:01:49.236632 137274321021824 utils.py:1231] [16950] train/loss = 3.2084808349609375
+I1129 04:01:49.236724 137274321021824 utils.py:1231] [16950] l2_grads = 1.394953966140747
+I1129 04:01:49.236782 137274321021824 utils.py:1231] [16950] lr = 0.0009887247751189483
+I1129 04:01:49.236844 137274321021824 utils.py:1231] [16950] uptime = 108698.599205285
+I1129 04:01:49.236907 137274321021824 utils.py:1231] [16950] examples_seen = 17356800.0
+I1129 04:01:49.236958 137274321021824 utils.py:1231] [16950] progress = 0.15052884914256282
+I1129 04:01:49.237003 137274321021824 utils.py:1231] [16950] epoch = 13.547648354976362
+I1129 04:01:49.237052 137274321021824 utils.py:1231] [16950] img/sec/core = 164.22072452987425
+I1129 04:01:49.237106 137274321021824 utils.py:1231] [16950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.15973773244472
+I1129 04:01:49.237154 137274321021824 utils.py:1231] [16950] core_hours = 30.15973773244472
+I1129 04:01:49.237213 137274321021824 train.py:125] NOTE: Steps:16950/112603 [15.1%]
+Walltime:1d6h11m (0s eval)
+ETA:7d2h13m
+Total train time:8d8h22m
+I1129 04:06:56.557935 137274321021824 utils.py:1231] [17000] l2_params = 317.8855785914794
+I1129 04:06:56.558215 137274321021824 utils.py:1231] [17000] train/loss = 5.40420937538147
+I1129 04:06:56.558340 137274321021824 utils.py:1231] [17000] l2_grads = 0.9089248180389404
+I1129 04:06:56.558434 137274321021824 utils.py:1231] [17000] lr = 0.0009885625582445436
+I1129 04:06:56.558506 137274321021824 utils.py:1231] [17000] uptime = 109005.92086582699
+I1129 04:06:56.558569 137274321021824 utils.py:1231] [17000] examples_seen = 17408000.0
+I1129 04:06:56.558635 137274321021824 utils.py:1231] [17000] progress = 0.15097288704563822
+I1129 04:06:56.558691 137274321021824 utils.py:1231] [17000] epoch = 13.587611919445319
+I1129 04:06:56.558746 137274321021824 utils.py:1231] [17000] img/sec/core = 166.60068772798022
+I1129 04:06:56.558809 137274321021824 utils.py:1231] [17000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.24510486037305
+I1129 04:06:56.558862 137274321021824 utils.py:1231] [17000] core_hours = 30.24510486037305
+I1129 04:06:56.558946 137274321021824 train.py:125] NOTE: Steps:17000/112603 [15.1%]
+Walltime:1d6h16m (0s eval)
+ETA:7d2h6m
+Total train time:8d8h21m
+I1129 04:12:06.330390 137274321021824 utils.py:1231] [17050] l2_params = 317.9826027723461
+I1129 04:12:06.330630 137274321021824 utils.py:1231] [17050] train/loss = 5.710434377193451
+I1129 04:12:06.330767 137274321021824 utils.py:1231] [17050] l2_grads = 0.8155744075775146
+I1129 04:12:06.330861 137274321021824 utils.py:1231] [17050] lr = 0.0009883991962798494
+I1129 04:12:06.330943 137274321021824 utils.py:1231] [17050] uptime = 109315.693300356
+I1129 04:12:06.331000 137274321021824 utils.py:1231] [17050] examples_seen = 17459200.0
+I1129 04:12:06.331053 137274321021824 utils.py:1231] [17050] progress = 0.15141692494871362
+I1129 04:12:06.331128 137274321021824 utils.py:1231] [17050] epoch = 13.627575483914274
+I1129 04:12:06.331182 137274321021824 utils.py:1231] [17050] img/sec/core = 165.28262134701362
+I1129 04:12:06.331243 137274321021824 utils.py:1231] [17050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.331152758853328
+I1129 04:12:06.331298 137274321021824 utils.py:1231] [17050] core_hours = 30.331152758853328
+I1129 04:12:06.331363 137274321021824 train.py:125] NOTE: Steps:17050/112603 [15.1%]
+Walltime:1d6h21m (0s eval)
+ETA:7d2h0m
+Total train time:8d8h20m
+I1129 04:17:14.016817 137274321021824 utils.py:1231] [17100] l2_params = 318.1339412051936
+I1129 04:17:14.017028 137274321021824 utils.py:1231] [17100] train/loss = 3.3361688554286957
+I1129 04:17:14.017129 137274321021824 utils.py:1231] [17100] l2_grads = 1.2925643920898438
+I1129 04:17:14.017198 137274321021824 utils.py:1231] [17100] lr = 0.0009882346896077532
+I1129 04:17:14.017293 137274321021824 utils.py:1231] [17100] uptime = 109623.37965067
+I1129 04:17:14.017374 137274321021824 utils.py:1231] [17100] examples_seen = 17510400.0
+I1129 04:17:14.017457 137274321021824 utils.py:1231] [17100] progress = 0.15186096285178902
+I1129 04:17:14.017532 137274321021824 utils.py:1231] [17100] epoch = 13.667539048383231
+I1129 04:17:14.017619 137274321021824 utils.py:1231] [17100] img/sec/core = 166.40322181256525
+I1129 04:17:14.017688 137274321021824 utils.py:1231] [17100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.41662118949611
+I1129 04:17:14.017753 137274321021824 utils.py:1231] [17100] core_hours = 30.41662118949611
+I1129 04:17:14.017840 137274321021824 train.py:125] NOTE: Steps:17100/112603 [15.2%]
+Walltime:1d6h27m (0s eval)
+ETA:7d1h53m
+Total train time:8d8h18m
+I1129 04:22:20.641681 137274321021824 utils.py:1231] [17150] l2_params = 318.3373023643371
+I1129 04:22:20.641999 137274321021824 utils.py:1231] [17150] train/loss = 4.973368525505066
+I1129 04:22:20.642172 137274321021824 utils.py:1231] [17150] l2_grads = 0.9035207629203796
+I1129 04:22:20.642250 137274321021824 utils.py:1231] [17150] lr = 0.0009880690386138236
+I1129 04:22:20.642321 137274321021824 utils.py:1231] [17150] uptime = 109930.00466849601
+I1129 04:22:20.642380 137274321021824 utils.py:1231] [17150] examples_seen = 17561600.0
+I1129 04:22:20.642427 137274321021824 utils.py:1231] [17150] progress = 0.15230500075486444
+I1129 04:22:20.642474 137274321021824 utils.py:1231] [17150] epoch = 13.707502612852188
+I1129 04:22:20.642523 137274321021824 utils.py:1231] [17150] img/sec/core = 166.97919942414126
+I1129 04:22:20.642577 137274321021824 utils.py:1231] [17150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.50179480555889
+I1129 04:22:20.642627 137274321021824 utils.py:1231] [17150] core_hours = 30.50179480555889
+I1129 04:22:20.642686 137274321021824 train.py:125] NOTE: Steps:17150/112603 [15.2%]
+Walltime:1d6h32m (0s eval)
+ETA:7d1h47m
+Total train time:8d8h17m
+I1129 04:27:26.182435 137274321021824 utils.py:1231] [17200] l2_params = 318.50164139989397
+I1129 04:27:26.182731 137274321021824 utils.py:1231] [17200] train/loss = 3.1729671359062195
+I1129 04:27:26.182914 137274321021824 utils.py:1231] [17200] l2_grads = 1.1775619983673096
+I1129 04:27:26.182990 137274321021824 utils.py:1231] [17200] lr = 0.0009879022436863124
+I1129 04:27:26.183048 137274321021824 utils.py:1231] [17200] uptime = 110235.545410209
+I1129 04:27:26.183107 137274321021824 utils.py:1231] [17200] examples_seen = 17612800.0
+I1129 04:27:26.183164 137274321021824 utils.py:1231] [17200] progress = 0.15274903865793985
+I1129 04:27:26.183218 137274321021824 utils.py:1231] [17200] epoch = 13.747466177321146
+I1129 04:27:26.183286 137274321021824 utils.py:1231] [17200] img/sec/core = 167.57176052185793
+I1129 04:27:26.183349 137274321021824 utils.py:1231] [17200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.586667233812495
+I1129 04:27:26.183426 137274321021824 utils.py:1231] [17200] core_hours = 30.586667233812495
+I1129 04:27:26.183494 137274321021824 train.py:125] NOTE: Steps:17200/112603 [15.3%]
+Walltime:1d6h37m (0s eval)
+ETA:7d1h40m
+Total train time:8d8h15m
+I1129 04:32:33.105918 137274321021824 utils.py:1231] [17250] l2_params = 318.65615666195345
+I1129 04:32:33.106123 137274321021824 utils.py:1231] [17250] train/loss = 5.648161172866821
+I1129 04:32:33.106222 137274321021824 utils.py:1231] [17250] l2_grads = 1.1393297910690308
+I1129 04:32:33.106301 137274321021824 utils.py:1231] [17250] lr = 0.0009877343052161537
+I1129 04:32:33.106379 137274321021824 utils.py:1231] [17250] uptime = 110542.468739083
+I1129 04:32:33.106449 137274321021824 utils.py:1231] [17250] examples_seen = 17664000.0
+I1129 04:32:33.106521 137274321021824 utils.py:1231] [17250] progress = 0.15319307656101525
+I1129 04:32:33.106591 137274321021824 utils.py:1231] [17250] epoch = 13.787429741790103
+I1129 04:32:33.106663 137274321021824 utils.py:1231] [17250] img/sec/core = 166.816905667735
+I1129 04:32:33.106721 137274321021824 utils.py:1231] [17250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.67192371405527
+I1129 04:32:33.106788 137274321021824 utils.py:1231] [17250] core_hours = 30.67192371405527
+I1129 04:32:33.106888 137274321021824 train.py:125] NOTE: Steps:17250/112603 [15.3%]
+Walltime:1d6h42m (0s eval)
+ETA:7d1h33m
+Total train time:8d8h14m
+I1129 04:37:37.321775 137274321021824 utils.py:1231] [17300] l2_params = 318.80256041590445
+I1129 04:37:37.322054 137274321021824 utils.py:1231] [17300] train/loss = 5.766488790512085
+I1129 04:37:37.322170 137274321021824 utils.py:1231] [17300] l2_grads = 1.1119122505187988
+I1129 04:37:37.322250 137274321021824 utils.py:1231] [17300] lr = 0.0009875652235969612
+I1129 04:37:37.322329 137274321021824 utils.py:1231] [17300] uptime = 110846.68468601601
+I1129 04:37:37.322395 137274321021824 utils.py:1231] [17300] examples_seen = 17715200.0
+I1129 04:37:37.322449 137274321021824 utils.py:1231] [17300] progress = 0.15363711446409065
+I1129 04:37:37.322504 137274321021824 utils.py:1231] [17300] epoch = 13.82739330625906
+I1129 04:37:37.322567 137274321021824 utils.py:1231] [17300] img/sec/core = 168.30149936641072
+I1129 04:37:37.322632 137274321021824 utils.py:1231] [17300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.756428143758892
+I1129 04:37:37.322713 137274321021824 utils.py:1231] [17300] core_hours = 30.756428143758892
+I1129 04:37:37.322798 137274321021824 train.py:125] NOTE: Steps:17300/112603 [15.4%]
+Walltime:1d6h47m (0s eval)
+ETA:7d1h27m
+Total train time:8d8h12m
+I1129 04:42:45.362921 137274321021824 utils.py:1231] [17350] l2_params = 318.91575115195343
+I1129 04:42:45.363158 137274321021824 utils.py:1231] [17350] train/loss = 3.303148001432419
+I1129 04:42:45.363255 137274321021824 utils.py:1231] [17350] l2_grads = 1.2201558351516724
+I1129 04:42:45.363348 137274321021824 utils.py:1231] [17350] lr = 0.000987394999225027
+I1129 04:42:45.363420 137274321021824 utils.py:1231] [17350] uptime = 111154.72577742401
+I1129 04:42:45.363492 137274321021824 utils.py:1231] [17350] examples_seen = 17766400.0
+I1129 04:42:45.363582 137274321021824 utils.py:1231] [17350] progress = 0.15408115236716607
+I1129 04:42:45.363641 137274321021824 utils.py:1231] [17350] epoch = 13.867356870728017
+I1129 04:42:45.363697 137274321021824 utils.py:1231] [17350] img/sec/core = 166.21159133664386
+I1129 04:42:45.363758 137274321021824 utils.py:1231] [17350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.841995113594447
+I1129 04:42:45.363818 137274321021824 utils.py:1231] [17350] core_hours = 30.841995113594447
+I1129 04:42:45.363888 137274321021824 train.py:125] NOTE: Steps:17350/112603 [15.4%]
+Walltime:1d6h52m (0s eval)
+ETA:7d1h20m
+Total train time:8d8h11m
+I1129 04:47:54.823014 137274321021824 utils.py:1231] [17400] l2_params = 319.12184649532253
+I1129 04:47:54.823273 137274321021824 utils.py:1231] [17400] train/loss = 5.0657148361206055
+I1129 04:47:54.823402 137274321021824 utils.py:1231] [17400] l2_grads = 0.9492163062095642
+I1129 04:47:54.823504 137274321021824 utils.py:1231] [17400] lr = 0.0009872236324993213
+I1129 04:47:54.823572 137274321021824 utils.py:1231] [17400] uptime = 111464.18593349202
+I1129 04:47:54.823647 137274321021824 utils.py:1231] [17400] examples_seen = 17817600.0
+I1129 04:47:54.823710 137274321021824 utils.py:1231] [17400] progress = 0.15452519027024147
+I1129 04:47:54.823769 137274321021824 utils.py:1231] [17400] epoch = 13.907320435196972
+I1129 04:47:54.823843 137274321021824 utils.py:1231] [17400] img/sec/core = 165.4494092245888
+I1129 04:47:54.823916 137274321021824 utils.py:1231] [17400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 30.927956268057777
+I1129 04:47:54.823984 137274321021824 utils.py:1231] [17400] core_hours = 30.927956268057777
+I1129 04:47:54.824047 137274321021824 train.py:125] NOTE: Steps:17400/112603 [15.5%]
+Walltime:1d6h57m (0s eval)
+ETA:7d1h14m
+Total train time:8d8h10m
+I1129 04:52:58.875834 137274321021824 utils.py:1231] [17450] l2_params = 319.2675129542399
+I1129 04:52:58.876085 137274321021824 utils.py:1231] [17450] train/loss = 5.630066215991974
+I1129 04:52:58.876229 137274321021824 utils.py:1231] [17450] l2_grads = 1.0190834999084473
+I1129 04:52:58.876311 137274321021824 utils.py:1231] [17450] lr = 0.0009870511238214935
+I1129 04:52:58.876371 137274321021824 utils.py:1231] [17450] uptime = 111768.23873325602
+I1129 04:52:58.876427 137274321021824 utils.py:1231] [17450] examples_seen = 17868800.0
+I1129 04:52:58.876475 137274321021824 utils.py:1231] [17450] progress = 0.15496922817331688
+I1129 04:52:58.876522 137274321021824 utils.py:1231] [17450] epoch = 13.94728399966593
+I1129 04:52:58.876572 137274321021824 utils.py:1231] [17450] img/sec/core = 168.39180576445978
+I1129 04:52:58.876628 137274321021824 utils.py:1231] [17450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.012415379103334
+I1129 04:52:58.876677 137274321021824 utils.py:1231] [17450] core_hours = 31.012415379103334
+I1129 04:52:58.876744 137274321021824 train.py:125] NOTE: Steps:17450/112603 [15.5%]
+Walltime:1d7h2m (0s eval)
+ETA:7d1h7m
+Total train time:8d8h8m
+I1129 04:58:00.802217 137274321021824 utils.py:1231] [17500] l2_params = 319.3860030207234
+I1129 04:58:00.802478 137274321021824 utils.py:1231] [17500] train/loss = 3.289493441581726
+I1129 04:58:00.802585 137274321021824 utils.py:1231] [17500] l2_grads = 1.2030105590820312
+I1129 04:58:00.802648 137274321021824 utils.py:1231] [17500] lr = 0.0009868774735958675
+I1129 04:58:00.802721 137274321021824 utils.py:1231] [17500] uptime = 112070.16508081001
+I1129 04:58:00.802775 137274321021824 utils.py:1231] [17500] examples_seen = 17920000.0
+I1129 04:58:00.802824 137274321021824 utils.py:1231] [17500] progress = 0.15541326607639228
+I1129 04:58:00.802873 137274321021824 utils.py:1231] [17500] epoch = 13.987247564134886
+I1129 04:58:00.802932 137274321021824 utils.py:1231] [17500] img/sec/core = 169.57778085545903
+I1129 04:58:00.802989 137274321021824 utils.py:1231] [17500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.096283808979443
+I1129 04:58:00.803039 137274321021824 utils.py:1231] [17500] core_hours = 31.096283808979443
+I1129 04:58:00.803099 137274321021824 train.py:125] NOTE: Steps:17500/112603 [15.5%]
+Walltime:1d7h7m (0s eval)
+ETA:7d1h0m
+Total train time:8d8h6m
+I1129 04:58:00.803240 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:17500/112603 [15.5%]
+Walltime:1d7h7m (0s eval)
+ETA:7d1h0m
+Total train time:8d8h6m
+I1129 04:59:31.376578 137274321021824 utils.py:1231] [17500] val/acc@1 = 0.4928451849489796
+I1129 04:59:31.376859 137274321021824 utils.py:1231] [17500] val/loss = 2.2551835757129046
+I1129 04:59:31.377046 137274321021824 utils.py:1231] [17500] z/secs/eval/val = 90.57373338600155
+I1129 04:59:31.377116 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 90.57373338600155
+I1129 05:04:38.131863 137274321021824 utils.py:1231] [17550] l2_params = 319.5349547208244
+I1129 05:04:38.132066 137274321021824 utils.py:1231] [17550] train/loss = 3.180873990058899
+I1129 05:04:38.132172 137274321021824 utils.py:1231] [17550] l2_grads = 1.2764650583267212
+I1129 05:04:38.132247 137274321021824 utils.py:1231] [17550] lr = 0.000986702682229445
+I1129 05:04:38.132341 137274321021824 utils.py:1231] [17550] uptime = 112467.49469916
+I1129 05:04:38.132416 137274321021824 utils.py:1231] [17550] examples_seen = 17971200.0
+I1129 05:04:38.132484 137274321021824 utils.py:1231] [17550] progress = 0.15585730397946768
+I1129 05:04:38.132556 137274321021824 utils.py:1231] [17550] epoch = 14.027211128603843
+I1129 05:04:38.132630 137274321021824 utils.py:1231] [17550] img/sec/core = 128.86026521914104
+I1129 05:04:38.132718 137274321021824 utils.py:1231] [17550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.206653147409998
+I1129 05:04:38.132782 137274321021824 utils.py:1231] [17550] core_hours = 31.206653147409998
+I1129 05:04:38.132865 137274321021824 train.py:125] NOTE: Steps:17550/112603 [15.6%]
+Walltime:1d7h14m (0s eval)
+ETA:7d1h2m
+Total train time:8d8h14m
+I1129 05:09:41.216279 137274321021824 utils.py:1231] [17600] l2_params = 319.7440798018368
+I1129 05:09:41.216478 137274321021824 utils.py:1231] [17600] train/loss = 5.6964821219444275
+I1129 05:09:41.216571 137274321021824 utils.py:1231] [17600] l2_grads = 0.9387259483337402
+I1129 05:09:41.216637 137274321021824 utils.py:1231] [17600] lr = 0.0009865267501319
+I1129 05:09:41.216691 137274321021824 utils.py:1231] [17600] uptime = 112770.57905037502
+I1129 05:09:41.216765 137274321021824 utils.py:1231] [17600] examples_seen = 18022400.0
+I1129 05:09:41.216824 137274321021824 utils.py:1231] [17600] progress = 0.1563013418825431
+I1129 05:09:41.216874 137274321021824 utils.py:1231] [17600] epoch = 14.0671746930728
+I1129 05:09:41.216932 137274321021824 utils.py:1231] [17600] img/sec/core = 168.92986983573178
+I1129 05:09:41.216985 137274321021824 utils.py:1231] [17600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.290843244969725
+I1129 05:09:41.217033 137274321021824 utils.py:1231] [17600] core_hours = 31.290843244969725
+I1129 05:09:41.217091 137274321021824 train.py:125] NOTE: Steps:17600/112603 [15.6%]
+Walltime:1d7h19m (0s eval)
+ETA:7d0h55m
+Total train time:8d8h13m
+I1129 05:14:47.969307 137274321021824 utils.py:1231] [17650] l2_params = 319.9263646783997
+I1129 05:14:47.969561 137274321021824 utils.py:1231] [17650] train/loss = 3.237509608268738
+I1129 05:14:47.969697 137274321021824 utils.py:1231] [17650] l2_grads = 1.230712890625
+I1129 05:14:47.969798 137274321021824 utils.py:1231] [17650] lr = 0.0009863496777155823
+I1129 05:14:47.969898 137274321021824 utils.py:1231] [17650] uptime = 113077.33224988
+I1129 05:14:47.969968 137274321021824 utils.py:1231] [17650] examples_seen = 18073600.0
+I1129 05:14:47.970034 137274321021824 utils.py:1231] [17650] progress = 0.1567453797856185
+I1129 05:14:47.970103 137274321021824 utils.py:1231] [17650] epoch = 14.107138257541758
+I1129 05:14:47.970161 137274321021824 utils.py:1231] [17650] img/sec/core = 166.9094245231137
+I1129 05:14:47.970230 137274321021824 utils.py:1231] [17650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.37605246705444
+I1129 05:14:47.970344 137274321021824 utils.py:1231] [17650] core_hours = 31.37605246705444
+I1129 05:14:47.970441 137274321021824 train.py:125] NOTE: Steps:17650/112603 [15.7%]
+Walltime:1d7h24m (0s eval)
+ETA:7d0h48m
+Total train time:8d8h11m
+I1129 05:19:45.046945 137274321021824 utils.py:1231] [17700] l2_params = 320.05451398043937
+I1129 05:19:45.047212 137274321021824 utils.py:1231] [17700] train/loss = 4.448845028877258
+I1129 05:19:45.047344 137274321021824 utils.py:1231] [17700] l2_grads = 1.0667986869812012
+I1129 05:19:45.047442 137274321021824 utils.py:1231] [17700] lr = 0.0009861714653955128
+I1129 05:19:45.047557 137274321021824 utils.py:1231] [17700] uptime = 113374.409900799
+I1129 05:19:45.047636 137274321021824 utils.py:1231] [17700] examples_seen = 18124800.0
+I1129 05:19:45.047715 137274321021824 utils.py:1231] [17700] progress = 0.1571894176886939
+I1129 05:19:45.047791 137274321021824 utils.py:1231] [17700] epoch = 14.147101822010713
+I1129 05:19:45.047888 137274321021824 utils.py:1231] [17700] img/sec/core = 172.34551250022963
+I1129 05:19:45.047973 137274321021824 utils.py:1231] [17700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.458574036754165
+I1129 05:19:45.048047 137274321021824 utils.py:1231] [17700] core_hours = 31.458574036754165
+I1129 05:19:45.048146 137274321021824 train.py:125] NOTE: Steps:17700/112603 [15.7%]
+Walltime:1d7h29m (0s eval)
+ETA:7d0h41m
+Total train time:8d8h9m
+I1129 05:24:48.771365 137274321021824 utils.py:1231] [17750] l2_params = 320.19293913557914
+I1129 05:24:48.771547 137274321021824 utils.py:1231] [17750] train/loss = 3.2581010460853577
+I1129 05:24:48.771633 137274321021824 utils.py:1231] [17750] l2_grads = 1.297914981842041
+I1129 05:24:48.771705 137274321021824 utils.py:1231] [17750] lr = 0.000985992113589384
+I1129 05:24:48.771754 137274321021824 utils.py:1231] [17750] uptime = 113678.134116388
+I1129 05:24:48.771806 137274321021824 utils.py:1231] [17750] examples_seen = 18176000.0
+I1129 05:24:48.771851 137274321021824 utils.py:1231] [17750] progress = 0.1576334555917693
+I1129 05:24:48.771899 137274321021824 utils.py:1231] [17750] epoch = 14.18706538647967
+I1129 05:24:48.771946 137274321021824 utils.py:1231] [17750] img/sec/core = 168.5739805129145
+I1129 05:24:48.771996 137274321021824 utils.py:1231] [17750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.542941874417775
+I1129 05:24:48.772042 137274321021824 utils.py:1231] [17750] core_hours = 31.542941874417775
+I1129 05:24:48.772097 137274321021824 train.py:125] NOTE: Steps:17750/112603 [15.8%]
+Walltime:1d7h34m (0s eval)
+ETA:7d0h34m
+Total train time:8d8h7m
+I1129 05:29:52.939223 137274321021824 utils.py:1231] [17800] l2_params = 320.39757407441977
+I1129 05:29:52.939513 137274321021824 utils.py:1231] [17800] train/loss = 3.2075773775577545
+I1129 05:29:52.939672 137274321021824 utils.py:1231] [17800] l2_grads = 1.2773860692977905
+I1129 05:29:52.939749 137274321021824 utils.py:1231] [17800] lr = 0.0009858116227175599
+I1129 05:29:52.939855 137274321021824 utils.py:1231] [17800] uptime = 113982.302201862
+I1129 05:29:52.939940 137274321021824 utils.py:1231] [17800] examples_seen = 18227200.0
+I1129 05:29:52.939996 137274321021824 utils.py:1231] [17800] progress = 0.1580774934948447
+I1129 05:29:52.940046 137274321021824 utils.py:1231] [17800] epoch = 14.227028950948627
+I1129 05:29:52.940095 137274321021824 utils.py:1231] [17800] img/sec/core = 168.32798194528868
+I1129 05:29:52.940163 137274321021824 utils.py:1231] [17800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.627433009271662
+I1129 05:29:52.940212 137274321021824 utils.py:1231] [17800] core_hours = 31.627433009271662
+I1129 05:29:52.940273 137274321021824 train.py:125] NOTE: Steps:17800/112603 [15.8%]
+Walltime:1d7h39m (0s eval)
+ETA:7d0h28m
+Total train time:8d8h5m
+I1129 05:34:54.842818 137274321021824 utils.py:1231] [17850] l2_params = 320.5645511535636
+I1129 05:34:54.843032 137274321021824 utils.py:1231] [17850] train/loss = 5.264568865299225
+I1129 05:34:54.843145 137274321021824 utils.py:1231] [17850] l2_grads = 0.8732937574386597
+I1129 05:34:54.843213 137274321021824 utils.py:1231] [17850] lr = 0.0009856299932030743
+I1129 05:34:54.843271 137274321021824 utils.py:1231] [17850] uptime = 114284.20563322499
+I1129 05:34:54.843345 137274321021824 utils.py:1231] [17850] examples_seen = 18278400.0
+I1129 05:34:54.843405 137274321021824 utils.py:1231] [17850] progress = 0.15852153139792013
+I1129 05:34:54.843462 137274321021824 utils.py:1231] [17850] epoch = 14.266992515417584
+I1129 05:34:54.843518 137274321021824 utils.py:1231] [17850] img/sec/core = 169.59065277545392
+I1129 05:34:54.843580 137274321021824 utils.py:1231] [17850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.71129507353916
+I1129 05:34:54.843641 137274321021824 utils.py:1231] [17850] core_hours = 31.71129507353916
+I1129 05:34:54.843705 137274321021824 train.py:125] NOTE: Steps:17850/112603 [15.9%]
+Walltime:1d7h44m (0s eval)
+ETA:7d0h21m
+Total train time:8d8h3m
+I1129 05:39:51.907481 137274321021824 utils.py:1231] [17900] l2_params = 320.68335054164044
+I1129 05:39:51.907690 137274321021824 utils.py:1231] [17900] train/loss = 4.846194267272949
+I1129 05:39:51.907794 137274321021824 utils.py:1231] [17900] l2_grads = 1.115458369255066
+I1129 05:39:51.907854 137274321021824 utils.py:1231] [17900] lr = 0.0009854472254716294
+I1129 05:39:51.907909 137274321021824 utils.py:1231] [17900] uptime = 114581.27027140201
+I1129 05:39:51.907959 137274321021824 utils.py:1231] [17900] examples_seen = 18329600.0
+I1129 05:39:51.908009 137274321021824 utils.py:1231] [17900] progress = 0.15896556930099554
+I1129 05:39:51.908056 137274321021824 utils.py:1231] [17900] epoch = 14.306956079886541
+I1129 05:39:51.908106 137274321021824 utils.py:1231] [17900] img/sec/core = 172.35306199417118
+I1129 05:39:51.908159 137274321021824 utils.py:1231] [17900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.793813028588332
+I1129 05:39:51.908207 137274321021824 utils.py:1231] [17900] core_hours = 31.793813028588332
+I1129 05:39:51.908267 137274321021824 train.py:125] NOTE: Steps:17900/112603 [15.9%]
+Walltime:1d7h49m (0s eval)
+ETA:7d0h13m
+Total train time:8d8h1m
+I1129 05:44:55.891903 137274321021824 utils.py:1231] [17950] l2_params = 320.83422705478887
+I1129 05:44:55.892120 137274321021824 utils.py:1231] [17950] train/loss = 3.6851434409618378
+I1129 05:44:55.892230 137274321021824 utils.py:1231] [17950] l2_grads = 1.181270956993103
+I1129 05:44:55.892314 137274321021824 utils.py:1231] [17950] lr = 0.0009852633199515948
+I1129 05:44:55.892375 137274321021824 utils.py:1231] [17950] uptime = 114885.25473690401
+I1129 05:44:55.892454 137274321021824 utils.py:1231] [17950] examples_seen = 18380800.0
+I1129 05:44:55.892511 137274321021824 utils.py:1231] [17950] progress = 0.15940960720407094
+I1129 05:44:55.892567 137274321021824 utils.py:1231] [17950] epoch = 14.346919644355498
+I1129 05:44:55.892625 137274321021824 utils.py:1231] [17950] img/sec/core = 168.4296594414735
+I1129 05:44:55.892689 137274321021824 utils.py:1231] [17950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.878253157894445
+I1129 05:44:55.892743 137274321021824 utils.py:1231] [17950] core_hours = 31.878253157894445
+I1129 05:44:55.892807 137274321021824 train.py:125] NOTE: Steps:17950/112603 [15.9%]
+Walltime:1d7h54m (0s eval)
+ETA:7d0h7m
+Total train time:8d7h59m
+I1129 05:49:53.826012 137274321021824 utils.py:1231] [18000] l2_params = 320.95132700623225
+I1129 05:49:53.826251 137274321021824 utils.py:1231] [18000] train/loss = 3.1563914716243744
+I1129 05:49:53.826353 137274321021824 utils.py:1231] [18000] l2_grads = 1.2430477142333984
+I1129 05:49:53.826415 137274321021824 utils.py:1231] [18000] lr = 0.0009850782770740069
+I1129 05:49:53.826468 137274321021824 utils.py:1231] [18000] uptime = 115183.18882950199
+I1129 05:49:53.826519 137274321021824 utils.py:1231] [18000] examples_seen = 18432000.0
+I1129 05:49:53.826570 137274321021824 utils.py:1231] [18000] progress = 0.15985364510714634
+I1129 05:49:53.826623 137274321021824 utils.py:1231] [18000] epoch = 14.386883208824454
+I1129 05:49:53.826674 137274321021824 utils.py:1231] [18000] img/sec/core = 171.85008789540433
+I1129 05:49:53.826739 137274321021824 utils.py:1231] [18000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 31.961012628060548
+I1129 05:49:53.826789 137274321021824 utils.py:1231] [18000] core_hours = 31.961012628060548
+I1129 05:49:53.826852 137274321021824 train.py:125] NOTE: Steps:18000/112603 [16.0%]
+Walltime:1d7h59m (0s eval)
+ETA:6d23h59m
+Total train time:8d7h57m
+I1129 05:55:00.747133 137274321021824 utils.py:1231] [18050] l2_params = 321.0428023709147
+I1129 05:55:00.747347 137274321021824 utils.py:1231] [18050] train/loss = 5.531985342502594
+I1129 05:55:00.747433 137274321021824 utils.py:1231] [18050] l2_grads = 1.0065834522247314
+I1129 05:55:00.747521 137274321021824 utils.py:1231] [18050] lr = 0.00098489209727257
+I1129 05:55:00.747579 137274321021824 utils.py:1231] [18050] uptime = 115490.109933262
+I1129 05:55:00.747638 137274321021824 utils.py:1231] [18050] examples_seen = 18483200.0
+I1129 05:55:00.747687 137274321021824 utils.py:1231] [18050] progress = 0.16029768301022176
+I1129 05:55:00.747735 137274321021824 utils.py:1231] [18050] epoch = 14.426846773293411
+I1129 05:55:00.747782 137274321021824 utils.py:1231] [18050] img/sec/core = 166.81811505550084
+I1129 05:55:00.747836 137274321021824 utils.py:1231] [18050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.046268490216114
+I1129 05:55:00.747887 137274321021824 utils.py:1231] [18050] core_hours = 32.046268490216114
+I1129 05:55:00.747948 137274321021824 train.py:125] NOTE: Steps:18050/112603 [16.0%]
+Walltime:1d8h4m (0s eval)
+ETA:6d23h53m
+Total train time:8d7h56m
+I1129 06:00:02.663678 137274321021824 utils.py:1231] [18100] l2_params = 321.22735388477145
+I1129 06:00:02.663896 137274321021824 utils.py:1231] [18100] train/loss = 3.1963729560375214
+I1129 06:00:02.664001 137274321021824 utils.py:1231] [18100] l2_grads = 1.227257251739502
+I1129 06:00:02.664081 137274321021824 utils.py:1231] [18100] lr = 0.0009847047809836505
+I1129 06:00:02.664146 137274321021824 utils.py:1231] [18100] uptime = 115792.026507168
+I1129 06:00:02.664245 137274321021824 utils.py:1231] [18100] examples_seen = 18534400.0
+I1129 06:00:02.664314 137274321021824 utils.py:1231] [18100] progress = 0.16074172091329716
+I1129 06:00:02.664369 137274321021824 utils.py:1231] [18100] epoch = 14.466810337762368
+I1129 06:00:02.664426 137274321021824 utils.py:1231] [18100] img/sec/core = 169.58327043000065
+I1129 06:00:02.664489 137274321021824 utils.py:1231] [18100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.13013420519
+I1129 06:00:02.664544 137274321021824 utils.py:1231] [18100] core_hours = 32.13013420519
+I1129 06:00:02.664610 137274321021824 train.py:125] NOTE: Steps:18100/112603 [16.1%]
+Walltime:1d8h9m (0s eval)
+ETA:6d23h46m
+Total train time:8d7h54m
+I1129 06:05:03.812079 137274321021824 utils.py:1231] [18150] l2_params = 321.346717483929
+I1129 06:05:03.812285 137274321021824 utils.py:1231] [18150] train/loss = 4.406110167503357
+I1129 06:05:03.812417 137274321021824 utils.py:1231] [18150] l2_grads = 1.0018118619918823
+I1129 06:05:03.812496 137274321021824 utils.py:1231] [18150] lr = 0.0009845163286462783
+I1129 06:05:03.812557 137274321021824 utils.py:1231] [18150] uptime = 116093.17491846399
+I1129 06:05:03.812617 137274321021824 utils.py:1231] [18150] examples_seen = 18585600.0
+I1129 06:05:03.812674 137274321021824 utils.py:1231] [18150] progress = 0.16118575881637257
+I1129 06:05:03.812730 137274321021824 utils.py:1231] [18150] epoch = 14.506773902231325
+I1129 06:05:03.812786 137274321021824 utils.py:1231] [18150] img/sec/core = 170.01583963090334
+I1129 06:05:03.812847 137274321021824 utils.py:1231] [18150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.21378654166111
+I1129 06:05:03.812908 137274321021824 utils.py:1231] [18150] core_hours = 32.21378654166111
+I1129 06:05:03.812973 137274321021824 train.py:125] NOTE: Steps:18150/112603 [16.1%]
+Walltime:1d8h14m (0s eval)
+ETA:6d23h39m
+Total train time:8d7h52m
+I1129 06:10:07.471990 137274321021824 utils.py:1231] [18200] l2_params = 321.47457361165783
+I1129 06:10:07.472211 137274321021824 utils.py:1231] [18200] train/loss = 3.2832146286964417
+I1129 06:10:07.472315 137274321021824 utils.py:1231] [18200] l2_grads = 1.2852996587753296
+I1129 06:10:07.472397 137274321021824 utils.py:1231] [18200] lr = 0.000984326740702148
+I1129 06:10:07.472470 137274321021824 utils.py:1231] [18200] uptime = 116396.834832873
+I1129 06:10:07.472529 137274321021824 utils.py:1231] [18200] examples_seen = 18636800.0
+I1129 06:10:07.472589 137274321021824 utils.py:1231] [18200] progress = 0.16162979671944797
+I1129 06:10:07.472647 137274321021824 utils.py:1231] [18200] epoch = 14.546737466700282
+I1129 06:10:07.472701 137274321021824 utils.py:1231] [18200] img/sec/core = 168.6096767156308
+I1129 06:10:07.472762 137274321021824 utils.py:1231] [18200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.29813651788583
+I1129 06:10:07.472826 137274321021824 utils.py:1231] [18200] core_hours = 32.29813651788583
+I1129 06:10:07.472893 137274321021824 train.py:125] NOTE: Steps:18200/112603 [16.2%]
+Walltime:1d8h19m (0s eval)
+ETA:6d23h32m
+Total train time:8d7h50m
+I1129 06:15:14.198989 137274321021824 utils.py:1231] [18250] l2_params = 321.600982607992
+I1129 06:15:14.199257 137274321021824 utils.py:1231] [18250] train/loss = 3.156219631433487
+I1129 06:15:14.199382 137274321021824 utils.py:1231] [18250] l2_grads = 1.3311270475387573
+I1129 06:15:14.199452 137274321021824 utils.py:1231] [18250] lr = 0.0009841360175956153
+I1129 06:15:14.199511 137274321021824 utils.py:1231] [18250] uptime = 116703.56186943901
+I1129 06:15:14.199580 137274321021824 utils.py:1231] [18250] examples_seen = 18688000.0
+I1129 06:15:14.199638 137274321021824 utils.py:1231] [18250] progress = 0.16207383462252337
+I1129 06:15:14.199706 137274321021824 utils.py:1231] [18250] epoch = 14.58670103116924
+I1129 06:15:14.199826 137274321021824 utils.py:1231] [18250] img/sec/core = 166.92366141966008
+I1129 06:15:14.199942 137274321021824 utils.py:1231] [18250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.3833384724875
+I1129 06:15:14.200046 137274321021824 utils.py:1231] [18250] core_hours = 32.3833384724875
+I1129 06:15:14.200170 137274321021824 train.py:125] NOTE: Steps:18250/112603 [16.2%]
+Walltime:1d8h25m (0s eval)
+ETA:6d23h26m
+Total train time:8d7h49m
+I1129 06:20:21.035622 137274321021824 utils.py:1231] [18300] l2_params = 321.790360409067
+I1129 06:20:21.035868 137274321021824 utils.py:1231] [18300] train/loss = 4.379779160022736
+I1129 06:20:21.035999 137274321021824 utils.py:1231] [18300] l2_grads = 1.1429023742675781
+I1129 06:20:21.036086 137274321021824 utils.py:1231] [18300] lr = 0.0009839441597736943
+I1129 06:20:21.036165 137274321021824 utils.py:1231] [18300] uptime = 117010.398522939
+I1129 06:20:21.036216 137274321021824 utils.py:1231] [18300] examples_seen = 18739200.0
+I1129 06:20:21.036264 137274321021824 utils.py:1231] [18300] progress = 0.1625178725255988
+I1129 06:20:21.036311 137274321021824 utils.py:1231] [18300] epoch = 14.626664595638195
+I1129 06:20:21.036360 137274321021824 utils.py:1231] [18300] img/sec/core = 166.86402819213143
+I1129 06:20:21.036414 137274321021824 utils.py:1231] [18300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.4685708762375
+I1129 06:20:21.036463 137274321021824 utils.py:1231] [18300] core_hours = 32.4685708762375
+I1129 06:20:21.036520 137274321021824 train.py:125] NOTE: Steps:18300/112603 [16.3%]
+Walltime:1d8h30m (0s eval)
+ETA:6d23h20m
+Total train time:8d7h48m
+I1129 06:25:24.126417 137274321021824 utils.py:1231] [18350] l2_params = 321.844082061327
+I1129 06:25:24.126631 137274321021824 utils.py:1231] [18350] train/loss = 3.0951808094978333
+I1129 06:25:24.126728 137274321021824 utils.py:1231] [18350] l2_grads = 1.3220947980880737
+I1129 06:25:24.126795 137274321021824 utils.py:1231] [18350] lr = 0.0009837511676860629
+I1129 06:25:24.126864 137274321021824 utils.py:1231] [18350] uptime = 117313.48922602301
+I1129 06:25:24.126949 137274321021824 utils.py:1231] [18350] examples_seen = 18790400.0
+I1129 06:25:24.127003 137274321021824 utils.py:1231] [18350] progress = 0.1629619104286742
+I1129 06:25:24.127056 137274321021824 utils.py:1231] [18350] epoch = 14.666628160107152
+I1129 06:25:24.127126 137274321021824 utils.py:1231] [18350] img/sec/core = 168.926329574046
+I1129 06:25:24.127188 137274321021824 utils.py:1231] [18350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.55276273820528
+I1129 06:25:24.127253 137274321021824 utils.py:1231] [18350] core_hours = 32.55276273820528
+I1129 06:25:24.127316 137274321021824 train.py:125] NOTE: Steps:18350/112603 [16.3%]
+Walltime:1d8h35m (0s eval)
+ETA:6d23h13m
+Total train time:8d7h46m
+I1129 06:30:21.153211 137274321021824 utils.py:1231] [18400] l2_params = 321.93274429233287
+I1129 06:30:21.153472 137274321021824 utils.py:1231] [18400] train/loss = 4.404770612716675
+I1129 06:30:21.153594 137274321021824 utils.py:1231] [18400] l2_grads = 1.1274456977844238
+I1129 06:30:21.153672 137274321021824 utils.py:1231] [18400] lr = 0.0009835570417850513
+I1129 06:30:21.153727 137274321021824 utils.py:1231] [18400] uptime = 117610.51608904499
+I1129 06:30:21.153787 137274321021824 utils.py:1231] [18400] examples_seen = 18841600.0
+I1129 06:30:21.153835 137274321021824 utils.py:1231] [18400] progress = 0.1634059483317496
+I1129 06:30:21.153888 137274321021824 utils.py:1231] [18400] epoch = 14.706591724576109
+I1129 06:30:21.153940 137274321021824 utils.py:1231] [18400] img/sec/core = 172.37498143799618
+I1129 06:30:21.153995 137274321021824 utils.py:1231] [18400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.63527020015583
+I1129 06:30:21.154045 137274321021824 utils.py:1231] [18400] core_hours = 32.63527020015583
+I1129 06:30:21.154105 137274321021824 train.py:125] NOTE: Steps:18400/112603 [16.3%]
+Walltime:1d8h40m (0s eval)
+ETA:6d23h6m
+Total train time:8d7h44m
+I1129 06:35:24.296840 137274321021824 utils.py:1231] [18450] l2_params = 322.1055236302114
+I1129 06:35:24.297041 137274321021824 utils.py:1231] [18450] train/loss = 3.3459450602531433
+I1129 06:35:24.297141 137274321021824 utils.py:1231] [18450] l2_grads = 1.3297605514526367
+I1129 06:35:24.297210 137274321021824 utils.py:1231] [18450] lr = 0.0009833617825256529
+I1129 06:35:24.297268 137274321021824 utils.py:1231] [18450] uptime = 117913.65962962901
+I1129 06:35:24.297325 137274321021824 utils.py:1231] [18450] examples_seen = 18892800.0
+I1129 06:35:24.297379 137274321021824 utils.py:1231] [18450] progress = 0.163849986234825
+I1129 06:35:24.297434 137274321021824 utils.py:1231] [18450] epoch = 14.746555289045066
+I1129 06:35:24.297503 137274321021824 utils.py:1231] [18450] img/sec/core = 168.89688594835772
+I1129 06:35:24.297561 137274321021824 utils.py:1231] [18450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.71947673920695
+I1129 06:35:24.297612 137274321021824 utils.py:1231] [18450] core_hours = 32.71947673920695
+I1129 06:35:24.297675 137274321021824 train.py:125] NOTE: Steps:18450/112603 [16.4%]
+Walltime:1d8h45m (0s eval)
+ETA:6d22h59m
+Total train time:8d7h42m
+I1129 06:40:30.081297 137274321021824 utils.py:1231] [18500] l2_params = 322.2853720861509
+I1129 06:40:30.081549 137274321021824 utils.py:1231] [18500] train/loss = 5.242862522602081
+I1129 06:40:30.081637 137274321021824 utils.py:1231] [18500] l2_grads = 1.0130794048309326
+I1129 06:40:30.081695 137274321021824 utils.py:1231] [18500] lr = 0.0009831653903655162
+I1129 06:40:30.081745 137274321021824 utils.py:1231] [18500] uptime = 118219.44410707701
+I1129 06:40:30.081797 137274321021824 utils.py:1231] [18500] examples_seen = 18944000.0
+I1129 06:40:30.081845 137274321021824 utils.py:1231] [18500] progress = 0.16429402413790042
+I1129 06:40:30.081897 137274321021824 utils.py:1231] [18500] epoch = 14.786518853514023
+I1129 06:40:30.081948 137274321021824 utils.py:1231] [18500] img/sec/core = 167.43819185101108
+I1129 06:40:30.082002 137274321021824 utils.py:1231] [18500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.804416871831386
+I1129 06:40:30.082052 137274321021824 utils.py:1231] [18500] core_hours = 32.804416871831386
+I1129 06:40:30.082112 137274321021824 train.py:125] NOTE: Steps:18500/112603 [16.4%]
+Walltime:1d8h50m (0s eval)
+ETA:6d22h52m
+Total train time:8d7h41m
+I1129 06:45:30.149340 137274321021824 utils.py:1231] [18550] l2_params = 322.4356400280789
+I1129 06:45:30.149537 137274321021824 utils.py:1231] [18550] train/loss = 3.1845160722732544
+I1129 06:45:30.149637 137274321021824 utils.py:1231] [18550] l2_grads = 1.3196790218353271
+I1129 06:45:30.149699 137274321021824 utils.py:1231] [18550] lr = 0.0009829678657649428
+I1129 06:45:30.149751 137274321021824 utils.py:1231] [18550] uptime = 118519.51211281701
+I1129 06:45:30.149803 137274321021824 utils.py:1231] [18550] examples_seen = 18995200.0
+I1129 06:45:30.149853 137274321021824 utils.py:1231] [18550] progress = 0.16473806204097582
+I1129 06:45:30.149906 137274321021824 utils.py:1231] [18550] epoch = 14.82648241798298
+I1129 06:45:30.149958 137274321021824 utils.py:1231] [18550] img/sec/core = 170.6279877247656
+I1129 06:45:30.150014 137274321021824 utils.py:1231] [18550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.88776909564805
+I1129 06:45:30.150065 137274321021824 utils.py:1231] [18550] core_hours = 32.88776909564805
+I1129 06:45:30.150125 137274321021824 train.py:125] NOTE: Steps:18550/112603 [16.5%]
+Walltime:1d8h55m (0s eval)
+ETA:6d22h46m
+Total train time:8d7h39m
+I1129 06:50:31.592134 137274321021824 utils.py:1231] [18600] l2_params = 322.6215623832165
+I1129 06:50:31.592394 137274321021824 utils.py:1231] [18600] train/loss = 5.576362133026123
+I1129 06:50:31.592502 137274321021824 utils.py:1231] [18600] l2_grads = 0.9599713087081909
+I1129 06:50:31.592568 137274321021824 utils.py:1231] [18600] lr = 0.0009827692091868908
+I1129 06:50:31.592623 137274321021824 utils.py:1231] [18600] uptime = 118820.954985162
+I1129 06:50:31.592680 137274321021824 utils.py:1231] [18600] examples_seen = 19046400.0
+I1129 06:50:31.592730 137274321021824 utils.py:1231] [18600] progress = 0.16518209994405122
+I1129 06:50:31.592780 137274321021824 utils.py:1231] [18600] epoch = 14.866445982451937
+I1129 06:50:31.592830 137274321021824 utils.py:1231] [18600] img/sec/core = 169.84976158734617
+I1129 06:50:31.592897 137274321021824 utils.py:1231] [18600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 32.971503226855
+I1129 06:50:31.592947 137274321021824 utils.py:1231] [18600] core_hours = 32.971503226855
+I1129 06:50:31.593006 137274321021824 train.py:125] NOTE: Steps:18600/112603 [16.5%]
+Walltime:1d9h0m (0s eval)
+ETA:6d22h39m
+Total train time:8d7h37m
+I1129 06:55:33.423526 137274321021824 utils.py:1231] [18650] l2_params = 322.7452104173945
+I1129 06:55:33.423758 137274321021824 utils.py:1231] [18650] train/loss = 4.353415012359619
+I1129 06:55:33.423906 137274321021824 utils.py:1231] [18650] l2_grads = 0.9283971786499023
+I1129 06:55:33.424001 137274321021824 utils.py:1231] [18650] lr = 0.0009825694210969683
+I1129 06:55:33.424086 137274321021824 utils.py:1231] [18650] uptime = 119122.786442987
+I1129 06:55:33.424169 137274321021824 utils.py:1231] [18650] examples_seen = 19097600.0
+I1129 06:55:33.424245 137274321021824 utils.py:1231] [18650] progress = 0.16562613784712663
+I1129 06:55:33.424333 137274321021824 utils.py:1231] [18650] epoch = 14.906409546920893
+I1129 06:55:33.424416 137274321021824 utils.py:1231] [18650] img/sec/core = 169.6310926930803
+I1129 06:55:33.424507 137274321021824 utils.py:1231] [18650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.05534529847305
+I1129 06:55:33.424588 137274321021824 utils.py:1231] [18650] core_hours = 33.05534529847305
+I1129 06:55:33.424684 137274321021824 train.py:125] NOTE: Steps:18650/112603 [16.6%]
+Walltime:1d9h5m (0s eval)
+ETA:6d22h32m
+Total train time:8d7h35m
+I1129 07:00:41.274791 137274321021824 utils.py:1231] [18700] l2_params = 322.87524736061175
+I1129 07:00:41.275026 137274321021824 utils.py:1231] [18700] train/loss = 3.0229262709617615
+I1129 07:00:41.275125 137274321021824 utils.py:1231] [18700] l2_grads = 1.280059576034546
+I1129 07:00:41.275187 137274321021824 utils.py:1231] [18700] lr = 0.0009823685019634397
+I1129 07:00:41.275239 137274321021824 utils.py:1231] [18700] uptime = 119430.63760103301
+I1129 07:00:41.275294 137274321021824 utils.py:1231] [18700] examples_seen = 19148800.0
+I1129 07:00:41.275342 137274321021824 utils.py:1231] [18700] progress = 0.16607017575020203
+I1129 07:00:41.275391 137274321021824 utils.py:1231] [18700] epoch = 14.94637311138985
+I1129 07:00:41.275442 137274321021824 utils.py:1231] [18700] img/sec/core = 166.31413805611393
+I1129 07:00:41.275497 137274321021824 utils.py:1231] [18700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.14085950904139
+I1129 07:00:41.275546 137274321021824 utils.py:1231] [18700] core_hours = 33.14085950904139
+I1129 07:00:41.275607 137274321021824 train.py:125] NOTE: Steps:18700/112603 [16.6%]
+Walltime:1d9h10m (0s eval)
+ETA:6d22h26m
+Total train time:8d7h34m
+I1129 07:05:41.979706 137274321021824 utils.py:1231] [18750] l2_params = 322.9470353662298
+I1129 07:05:41.979964 137274321021824 utils.py:1231] [18750] train/loss = 3.1142594516277313
+I1129 07:05:41.980060 137274321021824 utils.py:1231] [18750] l2_grads = 1.2284916639328003
+I1129 07:05:41.980122 137274321021824 utils.py:1231] [18750] lr = 0.0009821664522572176
+I1129 07:05:41.980173 137274321021824 utils.py:1231] [18750] uptime = 119731.342535689
+I1129 07:05:41.980223 137274321021824 utils.py:1231] [18750] examples_seen = 19200000.0
+I1129 07:05:41.980273 137274321021824 utils.py:1231] [18750] progress = 0.16651421365327745
+I1129 07:05:41.980320 137274321021824 utils.py:1231] [18750] epoch = 14.986336675858807
+I1129 07:05:41.980370 137274321021824 utils.py:1231] [18750] img/sec/core = 170.26657729636247
+I1129 07:05:41.980431 137274321021824 utils.py:1231] [18750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.22438865755694
+I1129 07:05:41.980479 137274321021824 utils.py:1231] [18750] core_hours = 33.22438865755694
+I1129 07:05:41.980539 137274321021824 train.py:125] NOTE: Steps:18750/112603 [16.7%]
+Walltime:1d9h15m (0s eval)
+ETA:6d22h19m
+Total train time:8d7h33m
+I1129 07:10:46.371378 137274321021824 utils.py:1231] [18800] l2_params = 323.0937528182695
+I1129 07:10:46.371602 137274321021824 utils.py:1231] [18800] train/loss = 3.021900087594986
+I1129 07:10:46.371724 137274321021824 utils.py:1231] [18800] l2_grads = 1.2774666547775269
+I1129 07:10:46.371813 137274321021824 utils.py:1231] [18800] lr = 0.000981963272451864
+I1129 07:10:46.371905 137274321021824 utils.py:1231] [18800] uptime = 120035.734250803
+I1129 07:10:46.371973 137274321021824 utils.py:1231] [18800] examples_seen = 19251200.0
+I1129 07:10:46.372035 137274321021824 utils.py:1231] [18800] progress = 0.16695825155635285
+I1129 07:10:46.372097 137274321021824 utils.py:1231] [18800] epoch = 15.026300240327764
+I1129 07:10:46.372159 137274321021824 utils.py:1231] [18800] img/sec/core = 168.20431522199573
+I1129 07:10:46.372224 137274321021824 utils.py:1231] [18800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.30894191175528
+I1129 07:10:46.372285 137274321021824 utils.py:1231] [18800] core_hours = 33.30894191175528
+I1129 07:10:46.372358 137274321021824 train.py:125] NOTE: Steps:18800/112603 [16.7%]
+Walltime:1d9h20m (0s eval)
+ETA:6d22h12m
+Total train time:8d7h31m
+I1129 07:15:54.432335 137274321021824 utils.py:1231] [18850] l2_params = 323.27345612028364
+I1129 07:15:54.432539 137274321021824 utils.py:1231] [18850] train/loss = 3.338065266609192
+I1129 07:15:54.432639 137274321021824 utils.py:1231] [18850] l2_grads = 1.2981210947036743
+I1129 07:15:54.432708 137274321021824 utils.py:1231] [18850] lr = 0.000981758963023592
+I1129 07:15:54.432775 137274321021824 utils.py:1231] [18850] uptime = 120343.795135868
+I1129 07:15:54.432835 137274321021824 utils.py:1231] [18850] examples_seen = 19302400.0
+I1129 07:15:54.432898 137274321021824 utils.py:1231] [18850] progress = 0.16740228945942826
+I1129 07:15:54.432957 137274321021824 utils.py:1231] [18850] epoch = 15.066263804796721
+I1129 07:15:54.433023 137274321021824 utils.py:1231] [18850] img/sec/core = 166.20091183986744
+I1129 07:15:54.433109 137274321021824 utils.py:1231] [18850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.394514379828884
+I1129 07:15:54.433181 137274321021824 utils.py:1231] [18850] core_hours = 33.394514379828884
+I1129 07:15:54.433266 137274321021824 train.py:125] NOTE: Steps:18850/112603 [16.7%]
+Walltime:1d9h25m (0s eval)
+ETA:6d22h6m
+Total train time:8d7h30m
+I1129 07:21:00.089938 137274321021824 utils.py:1231] [18900] l2_params = 323.41325971873874
+I1129 07:21:00.090214 137274321021824 utils.py:1231] [18900] train/loss = 3.583155870437622
+I1129 07:21:00.090324 137274321021824 utils.py:1231] [18900] l2_grads = 1.1914260387420654
+I1129 07:21:00.090392 137274321021824 utils.py:1231] [18900] lr = 0.0009815535244512609
+I1129 07:21:00.090448 137274321021824 utils.py:1231] [18900] uptime = 120649.452810306
+I1129 07:21:00.090501 137274321021824 utils.py:1231] [18900] examples_seen = 19353600.0
+I1129 07:21:00.090556 137274321021824 utils.py:1231] [18900] progress = 0.16784632736250366
+I1129 07:21:00.090603 137274321021824 utils.py:1231] [18900] epoch = 15.106227369265678
+I1129 07:21:00.090653 137274321021824 utils.py:1231] [18900] img/sec/core = 167.5076540909346
+I1129 07:21:00.090708 137274321021824 utils.py:1231] [18900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.479419289395
+I1129 07:21:00.090756 137274321021824 utils.py:1231] [18900] core_hours = 33.479419289395
+I1129 07:21:00.090823 137274321021824 train.py:125] NOTE: Steps:18900/112603 [16.8%]
+Walltime:1d9h30m (0s eval)
+ETA:6d22h0m
+Total train time:8d7h29m
+I1129 07:26:00.043874 137274321021824 utils.py:1231] [18950] l2_params = 323.5234977008382
+I1129 07:26:00.044157 137274321021824 utils.py:1231] [18950] train/loss = 3.4805631041526794
+I1129 07:26:00.044315 137274321021824 utils.py:1231] [18950] l2_grads = 1.215900182723999
+I1129 07:26:00.044412 137274321021824 utils.py:1231] [18950] lr = 0.0009813469572163755
+I1129 07:26:00.044467 137274321021824 utils.py:1231] [18950] uptime = 120949.406829576
+I1129 07:26:00.044519 137274321021824 utils.py:1231] [18950] examples_seen = 19404800.0
+I1129 07:26:00.044568 137274321021824 utils.py:1231] [18950] progress = 0.16829036526557906
+I1129 07:26:00.044614 137274321021824 utils.py:1231] [18950] epoch = 15.146190933734633
+I1129 07:26:00.044664 137274321021824 utils.py:1231] [18950] img/sec/core = 170.6928286028869
+I1129 07:26:00.044717 137274321021824 utils.py:1231] [18950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.56273985030333
+I1129 07:26:00.044764 137274321021824 utils.py:1231] [18950] core_hours = 33.56273985030333
+I1129 07:26:00.044824 137274321021824 train.py:125] NOTE: Steps:18950/112603 [16.8%]
+Walltime:1d9h35m (0s eval)
+ETA:6d21h53m
+Total train time:8d7h27m
+I1129 07:31:10.946420 137274321021824 utils.py:1231] [19000] l2_params = 323.67315211397465
+I1129 07:31:10.946663 137274321021824 utils.py:1231] [19000] train/loss = 4.426890730857849
+I1129 07:31:10.946770 137274321021824 utils.py:1231] [19000] l2_grads = 1.1019713878631592
+I1129 07:31:10.946846 137274321021824 utils.py:1231] [19000] lr = 0.0009811392618030873
+I1129 07:31:10.946909 137274321021824 utils.py:1231] [19000] uptime = 121260.309271146
+I1129 07:31:10.946967 137274321021824 utils.py:1231] [19000] examples_seen = 19456000.0
+I1129 07:31:10.947028 137274321021824 utils.py:1231] [19000] progress = 0.16873440316865448
+I1129 07:31:10.947079 137274321021824 utils.py:1231] [19000] epoch = 15.18615449820359
+I1129 07:31:10.947126 137274321021824 utils.py:1231] [19000] img/sec/core = 164.6818845855554
+I1129 07:31:10.947180 137274321021824 utils.py:1231] [19000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.649101639628334
+I1129 07:31:10.947228 137274321021824 utils.py:1231] [19000] core_hours = 33.649101639628334
+I1129 07:31:10.947285 137274321021824 train.py:125] NOTE: Steps:19000/112603 [16.9%]
+Walltime:1d9h41m (0s eval)
+ETA:6d21h47m
+Total train time:8d7h26m
+I1129 07:36:16.644010 137274321021824 utils.py:1231] [19050] l2_params = 323.7689061031967
+I1129 07:36:16.644223 137274321021824 utils.py:1231] [19050] train/loss = 5.666278958320618
+I1129 07:36:16.644330 137274321021824 utils.py:1231] [19050] l2_grads = 1.25330650806427
+I1129 07:36:16.644399 137274321021824 utils.py:1231] [19050] lr = 0.0009809304386981918
+I1129 07:36:16.644457 137274321021824 utils.py:1231] [19050] uptime = 121566.00681870001
+I1129 07:36:16.644517 137274321021824 utils.py:1231] [19050] examples_seen = 19507200.0
+I1129 07:36:16.644573 137274321021824 utils.py:1231] [19050] progress = 0.16917844107172988
+I1129 07:36:16.644628 137274321021824 utils.py:1231] [19050] epoch = 15.226118062672548
+I1129 07:36:16.644698 137274321021824 utils.py:1231] [19050] img/sec/core = 167.48580552794326
+I1129 07:36:16.644755 137274321021824 utils.py:1231] [19050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.73401762506
+I1129 07:36:16.644807 137274321021824 utils.py:1231] [19050] core_hours = 33.73401762506
+I1129 07:36:16.644890 137274321021824 train.py:125] NOTE: Steps:19050/112603 [16.9%]
+Walltime:1d9h46m (0s eval)
+ETA:6d21h40m
+Total train time:8d7h25m
+I1129 07:41:20.665333 137274321021824 utils.py:1231] [19100] l2_params = 323.893278284618
+I1129 07:41:20.665541 137274321021824 utils.py:1231] [19100] train/loss = 3.25099578499794
+I1129 07:41:20.889024 137274321021824 utils.py:1231] [19100] l2_grads = 1.251084327697754
+I1129 07:41:20.889332 137274321021824 utils.py:1231] [19100] lr = 0.0009807204883911281
+I1129 07:41:20.889422 137274321021824 utils.py:1231] [19100] uptime = 121870.25178101899
+I1129 07:41:20.889519 137274321021824 utils.py:1231] [19100] examples_seen = 19558400.0
+I1129 07:41:20.889590 137274321021824 utils.py:1231] [19100] progress = 0.16962247897480529
+I1129 07:41:20.889643 137274321021824 utils.py:1231] [19100] epoch = 15.266081627141505
+I1129 07:41:20.889718 137274321021824 utils.py:1231] [19100] img/sec/core = 168.28544870472217
+I1129 07:41:20.889780 137274321021824 utils.py:1231] [19100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.81853011459305
+I1129 07:41:20.889836 137274321021824 utils.py:1231] [19100] core_hours = 33.81853011459305
+I1129 07:41:20.889935 137274321021824 train.py:125] NOTE: Steps:19100/112603 [17.0%]
+Walltime:1d9h51m (0s eval)
+ETA:6d21h34m
+Total train time:8d7h23m
+I1129 07:46:25.882047 137274321021824 utils.py:1231] [19150] l2_params = 324.03763295778555
+I1129 07:46:25.882255 137274321021824 utils.py:1231] [19150] train/loss = 3.7820450961589813
+I1129 07:46:25.882363 137274321021824 utils.py:1231] [19150] l2_grads = 1.1300508975982666
+I1129 07:46:25.882430 137274321021824 utils.py:1231] [19150] lr = 0.0009805094113739755
+I1129 07:46:25.882487 137274321021824 utils.py:1231] [19150] uptime = 122175.244849186
+I1129 07:46:25.882547 137274321021824 utils.py:1231] [19150] examples_seen = 19609600.0
+I1129 07:46:25.882610 137274321021824 utils.py:1231] [19150] progress = 0.17006651687788069
+I1129 07:46:25.882668 137274321021824 utils.py:1231] [19150] epoch = 15.306045191610462
+I1129 07:46:25.882723 137274321021824 utils.py:1231] [19150] img/sec/core = 167.87266775507123
+I1129 07:46:25.882785 137274321021824 utils.py:1231] [19150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.90325041130611
+I1129 07:46:25.882839 137274321021824 utils.py:1231] [19150] core_hours = 33.90325041130611
+I1129 07:46:25.882907 137274321021824 train.py:125] NOTE: Steps:19150/112603 [17.0%]
+Walltime:1d9h56m (0s eval)
+ETA:6d21h28m
+Total train time:8d7h22m
+I1129 07:51:26.480602 137274321021824 utils.py:1231] [19200] l2_params = 324.1745338502359
+I1129 07:51:26.480865 137274321021824 utils.py:1231] [19200] train/loss = 3.4902206659317017
+I1129 07:51:26.481014 137274321021824 utils.py:1231] [19200] l2_grads = 1.1135077476501465
+I1129 07:51:26.481106 137274321021824 utils.py:1231] [19200] lr = 0.0009802972081414554
+I1129 07:51:26.481170 137274321021824 utils.py:1231] [19200] uptime = 122475.84353180199
+I1129 07:51:26.481241 137274321021824 utils.py:1231] [19200] examples_seen = 19660800.0
+I1129 07:51:26.481312 137274321021824 utils.py:1231] [19200] progress = 0.17051055478095611
+I1129 07:51:26.481375 137274321021824 utils.py:1231] [19200] epoch = 15.346008756079419
+I1129 07:51:26.481445 137274321021824 utils.py:1231] [19200] img/sec/core = 170.32676109697599
+I1129 07:51:26.481517 137274321021824 utils.py:1231] [19200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 33.98675004536611
+I1129 07:51:26.481573 137274321021824 utils.py:1231] [19200] core_hours = 33.98675004536611
+I1129 07:51:26.481653 137274321021824 train.py:125] NOTE: Steps:19200/112603 [17.1%]
+Walltime:1d10h1m (0s eval)
+ETA:6d21h21m
+Total train time:8d7h20m
+I1129 07:56:24.914550 137274321021824 utils.py:1231] [19250] l2_params = 324.27822872138296
+I1129 07:56:24.914795 137274321021824 utils.py:1231] [19250] train/loss = 3.3841663897037506
+I1129 07:56:24.914936 137274321021824 utils.py:1231] [19250] l2_grads = 1.2661609649658203
+I1129 07:56:24.915052 137274321021824 utils.py:1231] [19250] lr = 0.000980083879190929
+I1129 07:56:24.915119 137274321021824 utils.py:1231] [19250] uptime = 122774.27748067501
+I1129 07:56:24.915193 137274321021824 utils.py:1231] [19250] examples_seen = 19712000.0
+I1129 07:56:24.915280 137274321021824 utils.py:1231] [19250] progress = 0.17095459268403151
+I1129 07:56:24.915349 137274321021824 utils.py:1231] [19250] epoch = 15.385972320548374
+I1129 07:56:24.915419 137274321021824 utils.py:1231] [19250] img/sec/core = 171.5622508543264
+I1129 07:56:24.915506 137274321021824 utils.py:1231] [19250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.0696483644975
+I1129 07:56:24.915616 137274321021824 utils.py:1231] [19250] core_hours = 34.0696483644975
+I1129 07:56:24.915709 137274321021824 train.py:125] NOTE: Steps:19250/112603 [17.1%]
+Walltime:1d10h6m (0s eval)
+ETA:6d21h14m
+Total train time:8d7h18m
+I1129 08:01:26.589665 137274321021824 utils.py:1231] [19300] l2_params = 324.3838580784831
+I1129 08:01:26.589891 137274321021824 utils.py:1231] [19300] train/loss = 3.1460430026054382
+I1129 08:01:26.590005 137274321021824 utils.py:1231] [19300] l2_grads = 1.2570748329162598
+I1129 08:01:26.590083 137274321021824 utils.py:1231] [19300] lr = 0.000979869425022396
+I1129 08:01:26.590142 137274321021824 utils.py:1231] [19300] uptime = 123075.952503422
+I1129 08:01:26.590199 137274321021824 utils.py:1231] [19300] examples_seen = 19763200.0
+I1129 08:01:26.590270 137274321021824 utils.py:1231] [19300] progress = 0.17139863058710691
+I1129 08:01:26.590326 137274321021824 utils.py:1231] [19300] epoch = 15.425935885017331
+I1129 08:01:26.590396 137274321021824 utils.py:1231] [19300] img/sec/core = 169.71905573681101
+I1129 08:01:26.590452 137274321021824 utils.py:1231] [19300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.15344698192722
+I1129 08:01:26.590513 137274321021824 utils.py:1231] [19300] core_hours = 34.15344698192722
+I1129 08:01:26.590583 137274321021824 train.py:125] NOTE: Steps:19300/112603 [17.1%]
+Walltime:1d10h11m (0s eval)
+ETA:6d21h7m
+Total train time:8d7h17m
+I1129 08:06:36.263630 137274321021824 utils.py:1231] [19350] l2_params = 324.45366083023276
+I1129 08:06:36.263917 137274321021824 utils.py:1231] [19350] train/loss = 5.419668972492218
+I1129 08:06:36.264118 137274321021824 utils.py:1231] [19350] l2_grads = 0.9881940484046936
+I1129 08:06:36.264190 137274321021824 utils.py:1231] [19350] lr = 0.0009796538461384913
+I1129 08:06:36.264249 137274321021824 utils.py:1231] [19350] uptime = 123385.62661006399
+I1129 08:06:36.264307 137274321021824 utils.py:1231] [19350] examples_seen = 19814400.0
+I1129 08:06:36.264361 137274321021824 utils.py:1231] [19350] progress = 0.17184266849018232
+I1129 08:06:36.264416 137274321021824 utils.py:1231] [19350] epoch = 15.465899449486288
+I1129 08:06:36.264472 137274321021824 utils.py:1231] [19350] img/sec/core = 165.33510197283357
+I1129 08:06:36.264533 137274321021824 utils.py:1231] [19350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.23946756710555
+I1129 08:06:36.264591 137274321021824 utils.py:1231] [19350] core_hours = 34.23946756710555
+I1129 08:06:36.264655 137274321021824 train.py:125] NOTE: Steps:19350/112603 [17.2%]
+Walltime:1d10h16m (0s eval)
+ETA:6d21h1m
+Total train time:8d7h16m
+I1129 08:11:39.561952 137274321021824 utils.py:1231] [19400] l2_params = 324.54969527896264
+I1129 08:11:39.562187 137274321021824 utils.py:1231] [19400] train/loss = 3.6339070200920105
+I1129 08:11:39.562283 137274321021824 utils.py:1231] [19400] l2_grads = 1.1634674072265625
+I1129 08:11:39.562350 137274321021824 utils.py:1231] [19400] lr = 0.0009794371430444893
+I1129 08:11:39.562400 137274321021824 utils.py:1231] [19400] uptime = 123688.92476189301
+I1129 08:11:39.562450 137274321021824 utils.py:1231] [19400] examples_seen = 19865600.0
+I1129 08:11:39.562498 137274321021824 utils.py:1231] [19400] progress = 0.17228670639325772
+I1129 08:11:39.562545 137274321021824 utils.py:1231] [19400] epoch = 15.505863013955246
+I1129 08:11:39.562596 137274321021824 utils.py:1231] [19400] img/sec/core = 168.81078796966625
+I1129 08:11:39.562650 137274321021824 utils.py:1231] [19400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.32371705372472
+I1129 08:11:39.562697 137274321021824 utils.py:1231] [19400] core_hours = 34.32371705372472
+I1129 08:11:39.562778 137274321021824 train.py:125] NOTE: Steps:19400/112603 [17.2%]
+Walltime:1d10h21m (0s eval)
+ETA:6d20h55m
+Total train time:8d7h14m
+I1129 08:16:45.838711 137274321021824 utils.py:1231] [19450] l2_params = 324.653283886749
+I1129 08:16:45.838981 137274321021824 utils.py:1231] [19450] train/loss = 3.254388839006424
+I1129 08:16:45.839123 137274321021824 utils.py:1231] [19450] l2_grads = 1.2282414436340332
+I1129 08:16:45.839236 137274321021824 utils.py:1231] [19450] lr = 0.0009792193162482972
+I1129 08:16:45.839294 137274321021824 utils.py:1231] [19450] uptime = 123995.20165591499
+I1129 08:16:45.839355 137274321021824 utils.py:1231] [19450] examples_seen = 19916800.0
+I1129 08:16:45.839410 137274321021824 utils.py:1231] [19450] progress = 0.17273074429633314
+I1129 08:16:45.839464 137274321021824 utils.py:1231] [19450] epoch = 15.545826578424203
+I1129 08:16:45.839533 137274321021824 utils.py:1231] [19450] img/sec/core = 167.16899315403506
+I1129 08:16:45.839617 137274321021824 utils.py:1231] [19450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.408793968730826
+I1129 08:16:45.839679 137274321021824 utils.py:1231] [19450] core_hours = 34.408793968730826
+I1129 08:16:45.839751 137274321021824 train.py:125] NOTE: Steps:19450/112603 [17.3%]
+Walltime:1d10h26m (0s eval)
+ETA:6d20h48m
+Total train time:8d7h13m
+I1129 08:21:46.001810 137274321021824 utils.py:1231] [19500] l2_params = 324.752490936003
+I1129 08:21:46.002079 137274321021824 utils.py:1231] [19500] train/loss = 3.2480768263339996
+I1129 08:21:46.002218 137274321021824 utils.py:1231] [19500] l2_grads = 1.2484298944473267
+I1129 08:21:46.002303 137274321021824 utils.py:1231] [19500] lr = 0.0009790003662604556
+I1129 08:21:46.002369 137274321021824 utils.py:1231] [19500] uptime = 124295.36472607501
+I1129 08:21:46.002437 137274321021824 utils.py:1231] [19500] examples_seen = 19968000.0
+I1129 08:21:46.002506 137274321021824 utils.py:1231] [19500] progress = 0.17317478219940854
+I1129 08:21:46.002574 137274321021824 utils.py:1231] [19500] epoch = 15.58579014289316
+I1129 08:21:46.002641 137274321021824 utils.py:1231] [19500] img/sec/core = 170.5739482632035
+I1129 08:21:46.002720 137274321021824 utils.py:1231] [19500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.492172599330836
+I1129 08:21:46.002784 137274321021824 utils.py:1231] [19500] core_hours = 34.492172599330836
+I1129 08:21:46.002864 137274321021824 train.py:125] NOTE: Steps:19500/112603 [17.3%]
+Walltime:1d10h31m (0s eval)
+ETA:6d20h42m
+Total train time:8d7h11m
+I1129 08:26:45.497014 137274321021824 utils.py:1231] [19550] l2_params = 324.89276296846606
+I1129 08:26:45.497281 137274321021824 utils.py:1231] [19550] train/loss = 5.584177136421204
+I1129 08:26:45.497396 137274321021824 utils.py:1231] [19550] l2_grads = 1.0224679708480835
+I1129 08:26:45.497462 137274321021824 utils.py:1231] [19550] lr = 0.0009787802935941373
+I1129 08:26:45.497536 137274321021824 utils.py:1231] [19550] uptime = 124594.85989793001
+I1129 08:26:45.497594 137274321021824 utils.py:1231] [19550] examples_seen = 20019200.0
+I1129 08:26:45.497654 137274321021824 utils.py:1231] [19550] progress = 0.17361882010248395
+I1129 08:26:45.497713 137274321021824 utils.py:1231] [19550] epoch = 15.625753707362115
+I1129 08:26:45.497778 137274321021824 utils.py:1231] [19550] img/sec/core = 170.95434187763254
+I1129 08:26:45.497840 137274321021824 utils.py:1231] [19550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.57536570262389
+I1129 08:26:45.497901 137274321021824 utils.py:1231] [19550] core_hours = 34.57536570262389
+I1129 08:26:45.497970 137274321021824 train.py:125] NOTE: Steps:19550/112603 [17.4%]
+Walltime:1d10h36m (0s eval)
+ETA:6d20h35m
+Total train time:8d7h9m
+I1129 08:31:46.882005 137274321021824 utils.py:1231] [19600] l2_params = 324.9983172526261
+I1129 08:31:46.882266 137274321021824 utils.py:1231] [19600] train/loss = 3.1762249767780304
+I1129 08:31:46.882365 137274321021824 utils.py:1231] [19600] l2_grads = 1.3672906160354614
+I1129 08:31:46.882433 137274321021824 utils.py:1231] [19600] lr = 0.0009785590987651492
+I1129 08:31:46.882490 137274321021824 utils.py:1231] [19600] uptime = 124896.24485191402
+I1129 08:31:46.882548 137274321021824 utils.py:1231] [19600] examples_seen = 20070400.0
+I1129 08:31:46.882603 137274321021824 utils.py:1231] [19600] progress = 0.17406285800555935
+I1129 08:31:46.882657 137274321021824 utils.py:1231] [19600] epoch = 15.665717271831072
+I1129 08:31:46.882712 137274321021824 utils.py:1231] [19600] img/sec/core = 169.88240230040785
+I1129 08:31:46.882773 137274321021824 utils.py:1231] [19600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.659083745397226
+I1129 08:31:46.882828 137274321021824 utils.py:1231] [19600] core_hours = 34.659083745397226
+I1129 08:31:46.882903 137274321021824 train.py:125] NOTE: Steps:19600/112603 [17.4%]
+Walltime:1d10h41m (0s eval)
+ETA:6d20h28m
+Total train time:8d7h8m
+I1129 08:36:43.950269 137274321021824 utils.py:1231] [19650] l2_params = 325.07347898346734
+I1129 08:36:43.950476 137274321021824 utils.py:1231] [19650] train/loss = 5.6044246554374695
+I1129 08:36:43.950561 137274321021824 utils.py:1231] [19650] l2_grads = 1.0456479787826538
+I1129 08:36:43.950617 137274321021824 utils.py:1231] [19650] lr = 0.0009783367822919261
+I1129 08:36:43.950676 137274321021824 utils.py:1231] [19650] uptime = 125193.31303920301
+I1129 08:36:43.950725 137274321021824 utils.py:1231] [19650] examples_seen = 20121600.0
+I1129 08:36:43.950769 137274321021824 utils.py:1231] [19650] progress = 0.17450689590863477
+I1129 08:36:43.950812 137274321021824 utils.py:1231] [19650] epoch = 15.70568083630003
+I1129 08:36:43.950858 137274321021824 utils.py:1231] [19650] img/sec/core = 172.35100286989288
+I1129 08:36:43.950915 137274321021824 utils.py:1231] [19650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.741602686310834
+I1129 08:36:43.950960 137274321021824 utils.py:1231] [19650] core_hours = 34.741602686310834
+I1129 08:36:43.951015 137274321021824 train.py:125] NOTE: Steps:19650/112603 [17.5%]
+Walltime:1d10h46m (0s eval)
+ETA:6d20h21m
+Total train time:8d7h6m
+I1129 08:41:52.291933 137274321021824 utils.py:1231] [19700] l2_params = 325.1453103501469
+I1129 08:41:52.292149 137274321021824 utils.py:1231] [19700] train/loss = 3.8603623509407043
+I1129 08:41:52.292244 137274321021824 utils.py:1231] [19700] l2_grads = 1.2337967157363892
+I1129 08:41:52.292312 137274321021824 utils.py:1231] [19700] lr = 0.0009781133446955328
+I1129 08:41:52.292362 137274321021824 utils.py:1231] [19700] uptime = 125501.65472408001
+I1129 08:41:52.292412 137274321021824 utils.py:1231] [19700] examples_seen = 20172800.0
+I1129 08:41:52.292459 137274321021824 utils.py:1231] [19700] progress = 0.17495093381171017
+I1129 08:41:52.292506 137274321021824 utils.py:1231] [19700] epoch = 15.745644400768986
+I1129 08:41:52.292555 137274321021824 utils.py:1231] [19700] img/sec/core = 166.04955642122863
+I1129 08:41:52.292621 137274321021824 utils.py:1231] [19700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.82725315433222
+I1129 08:41:52.292670 137274321021824 utils.py:1231] [19700] core_hours = 34.82725315433222
+I1129 08:41:52.292747 137274321021824 train.py:125] NOTE: Steps:19700/112603 [17.5%]
+Walltime:1d10h51m (0s eval)
+ETA:6d20h15m
+Total train time:8d7h5m
+I1129 08:46:50.740402 137274321021824 utils.py:1231] [19750] l2_params = 325.22127352369887
+I1129 08:46:50.740647 137274321021824 utils.py:1231] [19750] train/loss = 4.997040212154388
+I1129 08:46:50.740748 137274321021824 utils.py:1231] [19750] l2_grads = 0.9376548528671265
+I1129 08:46:50.740807 137274321021824 utils.py:1231] [19750] lr = 0.0009778887864996602
+I1129 08:46:50.740861 137274321021824 utils.py:1231] [19750] uptime = 125800.103223283
+I1129 08:46:50.740924 137274321021824 utils.py:1231] [19750] examples_seen = 20224000.0
+I1129 08:46:50.740974 137274321021824 utils.py:1231] [19750] progress = 0.17539497171478557
+I1129 08:46:50.741021 137274321021824 utils.py:1231] [19750] epoch = 15.785607965237944
+I1129 08:46:50.741070 137274321021824 utils.py:1231] [19750] img/sec/core = 171.55388663950438
+I1129 08:46:50.741124 137274321021824 utils.py:1231] [19750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.910155515221945
+I1129 08:46:50.741173 137274321021824 utils.py:1231] [19750] core_hours = 34.910155515221945
+I1129 08:46:50.741232 137274321021824 train.py:125] NOTE: Steps:19750/112603 [17.5%]
+Walltime:1d10h56m (0s eval)
+ETA:6d20h8m
+Total train time:8d7h3m
+I1129 08:51:49.808925 137274321021824 utils.py:1231] [19800] l2_params = 325.32646009653
+I1129 08:51:49.809178 137274321021824 utils.py:1231] [19800] train/loss = 3.2873929142951965
+I1129 08:51:49.809338 137274321021824 utils.py:1231] [19800] l2_grads = 1.282878041267395
+I1129 08:51:49.809432 137274321021824 utils.py:1231] [19800] lr = 0.0009776631082306275
+I1129 08:51:49.809481 137274321021824 utils.py:1231] [19800] uptime = 126099.171844024
+I1129 08:51:49.809529 137274321021824 utils.py:1231] [19800] examples_seen = 20275200.0
+I1129 08:51:49.809574 137274321021824 utils.py:1231] [19800] progress = 0.17583900961786098
+I1129 08:51:49.809619 137274321021824 utils.py:1231] [19800] epoch = 15.8255715297069
+I1129 08:51:49.809665 137274321021824 utils.py:1231] [19800] img/sec/core = 171.19816807642354
+I1129 08:51:49.809715 137274321021824 utils.py:1231] [19800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 34.99323013209444
+I1129 08:51:49.809762 137274321021824 utils.py:1231] [19800] core_hours = 34.99323013209444
+I1129 08:51:49.809816 137274321021824 train.py:125] NOTE: Steps:19800/112603 [17.6%]
+Walltime:1d11h1m (0s eval)
+ETA:6d20h1m
+Total train time:8d7h1m
+I1129 08:56:54.094223 137274321021824 utils.py:1231] [19850] l2_params = 325.46142312930726
+I1129 08:56:54.094443 137274321021824 utils.py:1231] [19850] train/loss = 3.3655734062194824
+I1129 08:56:54.094544 137274321021824 utils.py:1231] [19850] l2_grads = 1.4125709533691406
+I1129 08:56:54.094606 137274321021824 utils.py:1231] [19850] lr = 0.0009774363104173775
+I1129 08:56:54.094657 137274321021824 utils.py:1231] [19850] uptime = 126403.457018753
+I1129 08:56:54.094708 137274321021824 utils.py:1231] [19850] examples_seen = 20326400.0
+I1129 08:56:54.094758 137274321021824 utils.py:1231] [19850] progress = 0.17628304752093638
+I1129 08:56:54.094805 137274321021824 utils.py:1231] [19850] epoch = 15.865535094175858
+I1129 08:56:54.094856 137274321021824 utils.py:1231] [19850] img/sec/core = 168.26320916093727
+I1129 08:56:54.094915 137274321021824 utils.py:1231] [19850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.07775379174139
+I1129 08:56:54.094972 137274321021824 utils.py:1231] [19850] core_hours = 35.07775379174139
+I1129 08:56:54.095034 137274321021824 train.py:125] NOTE: Steps:19850/112603 [17.6%]
+Walltime:1d11h6m (0s eval)
+ETA:6d19h55m
+Total train time:8d7h0m
+I1129 09:01:55.220163 137274321021824 utils.py:1231] [19900] l2_params = 325.58220143980736
+I1129 09:01:55.220419 137274321021824 utils.py:1231] [19900] train/loss = 4.535114943981171
+I1129 09:01:55.220553 137274321021824 utils.py:1231] [19900] l2_grads = 0.996752917766571
+I1129 09:01:55.220636 137274321021824 utils.py:1231] [19900] lr = 0.0009772083935914765
+I1129 09:01:55.220721 137274321021824 utils.py:1231] [19900] uptime = 126704.583078404
+I1129 09:01:55.220780 137274321021824 utils.py:1231] [19900] examples_seen = 20377600.0
+I1129 09:01:55.220836 137274321021824 utils.py:1231] [19900] progress = 0.1767270854240118
+I1129 09:01:55.220897 137274321021824 utils.py:1231] [19900] epoch = 15.905498658644813
+I1129 09:01:55.220962 137274321021824 utils.py:1231] [19900] img/sec/core = 170.0284593745906
+I1129 09:01:55.221051 137274321021824 utils.py:1231] [19900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.16139991942222
+I1129 09:01:55.221105 137274321021824 utils.py:1231] [19900] core_hours = 35.16139991942222
+I1129 09:01:55.221174 137274321021824 train.py:125] NOTE: Steps:19900/112603 [17.7%]
+Walltime:1d11h11m (0s eval)
+ETA:6d19h48m
+Total train time:8d6h58m
+I1129 09:07:04.691133 137274321021824 utils.py:1231] [19950] l2_params = 325.6752904629962
+I1129 09:07:04.691379 137274321021824 utils.py:1231] [19950] train/loss = 3.360542893409729
+I1129 09:07:04.691514 137274321021824 utils.py:1231] [19950] l2_grads = 1.1938540935516357
+I1129 09:07:04.691644 137274321021824 utils.py:1231] [19950] lr = 0.000976979358287116
+I1129 09:07:04.691744 137274321021824 utils.py:1231] [19950] uptime = 127014.05410203601
+I1129 09:07:04.691822 137274321021824 utils.py:1231] [19950] examples_seen = 20428800.0
+I1129 09:07:04.691915 137274321021824 utils.py:1231] [19950] progress = 0.1771711233270872
+I1129 09:07:04.692006 137274321021824 utils.py:1231] [19950] epoch = 15.94546222311377
+I1129 09:07:04.692118 137274321021824 utils.py:1231] [19950] img/sec/core = 165.44359920714786
+I1129 09:07:04.692249 137274321021824 utils.py:1231] [19950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.247364092653335
+I1129 09:07:04.692350 137274321021824 utils.py:1231] [19950] core_hours = 35.247364092653335
+I1129 09:07:04.692456 137274321021824 train.py:125] NOTE: Steps:19950/112603 [17.7%]
+Walltime:1d11h16m (0s eval)
+ETA:6d19h42m
+Total train time:8d6h57m
+I1129 09:12:02.873437 137274321021824 utils.py:1231] [20000] l2_params = 325.77300481274216
+I1129 09:12:02.873666 137274321021824 utils.py:1231] [20000] train/loss = 3.2563186585903168
+I1129 09:12:02.873791 137274321021824 utils.py:1231] [20000] l2_grads = 1.3203916549682617
+I1129 09:12:02.873872 137274321021824 utils.py:1231] [20000] lr = 0.0009767492050411083
+I1129 09:12:02.873947 137274321021824 utils.py:1231] [20000] uptime = 127312.236308865
+I1129 09:12:02.874006 137274321021824 utils.py:1231] [20000] examples_seen = 20480000.0
+I1129 09:12:02.874061 137274321021824 utils.py:1231] [20000] progress = 0.1776151612301626
+I1129 09:12:02.874112 137274321021824 utils.py:1231] [20000] epoch = 15.985425787582727
+I1129 09:12:02.874168 137274321021824 utils.py:1231] [20000] img/sec/core = 171.70709327187797
+I1129 09:12:02.874229 137274321021824 utils.py:1231] [20000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.33019248343916
+I1129 09:12:02.874284 137274321021824 utils.py:1231] [20000] core_hours = 35.33019248343916
+I1129 09:12:02.874355 137274321021824 train.py:125] NOTE: Steps:20000/112603 [17.8%]
+Walltime:1d11h21m (0s eval)
+ETA:6d19h36m
+Total train time:8d6h56m
+I1129 09:12:03.229342 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:20000/112603 [17.8%]
+Walltime:1d11h21m (0s eval)
+ETA:6d19h36m
+Total train time:8d6h56m
+I1129 09:13:34.775761 137274321021824 utils.py:1231] [20000] val/acc@1 = 0.5157445790816326
+I1129 09:13:34.776089 137274321021824 utils.py:1231] [20000] val/loss = 2.1216095281498775
+I1129 09:13:34.776291 137274321021824 utils.py:1231] [20000] z/secs/eval/val = 91.54664248798508
+I1129 09:13:34.776387 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 91.54664248798508
+I1129 09:18:40.971693 137274321021824 utils.py:1231] [20050] l2_params = 325.85503898210607
+I1129 09:18:40.971976 137274321021824 utils.py:1231] [20050] train/loss = 4.995117604732513
+I1129 09:18:40.972094 137274321021824 utils.py:1231] [20050] l2_grads = 0.9273712038993835
+I1129 09:18:40.972164 137274321021824 utils.py:1231] [20050] lr = 0.0009765179343928833
+I1129 09:18:40.972223 137274321021824 utils.py:1231] [20050] uptime = 127710.334579959
+I1129 09:18:40.972286 137274321021824 utils.py:1231] [20050] examples_seen = 20531200.0
+I1129 09:18:40.972335 137274321021824 utils.py:1231] [20050] progress = 0.178059199133238
+I1129 09:18:40.972386 137274321021824 utils.py:1231] [20050] epoch = 16.025389352051683
+I1129 09:18:40.972435 137274321021824 utils.py:1231] [20050] img/sec/core = 128.6114603293779
+I1129 09:18:40.972496 137274321021824 utils.py:1231] [20050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.44077533652083
+I1129 09:18:40.972558 137274321021824 utils.py:1231] [20050] core_hours = 35.44077533652083
+I1129 09:18:40.972616 137274321021824 train.py:125] NOTE: Steps:20050/112603 [17.8%]
+Walltime:1d11h28m (0s eval)
+ETA:6d19h36m
+Total train time:8d7h3m
+I1129 09:23:46.560374 137274321021824 utils.py:1231] [20100] l2_params = 325.9683227531691
+I1129 09:23:46.560597 137274321021824 utils.py:1231] [20100] train/loss = 5.352097690105438
+I1129 09:23:46.560697 137274321021824 utils.py:1231] [20100] l2_grads = 0.9392507076263428
+I1129 09:23:46.560766 137274321021824 utils.py:1231] [20100] lr = 0.0009762855468844938
+I1129 09:23:46.560828 137274321021824 utils.py:1231] [20100] uptime = 128015.923189946
+I1129 09:23:46.560891 137274321021824 utils.py:1231] [20100] examples_seen = 20582400.0
+I1129 09:23:46.560947 137274321021824 utils.py:1231] [20100] progress = 0.17850323703631343
+I1129 09:23:46.561001 137274321021824 utils.py:1231] [20100] epoch = 16.06535291652064
+I1129 09:23:46.561057 137274321021824 utils.py:1231] [20100] img/sec/core = 167.54551160194956
+I1129 09:23:46.561117 137274321021824 utils.py:1231] [20100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.52566106151722
+I1129 09:23:46.561171 137274321021824 utils.py:1231] [20100] core_hours = 35.52566106151722
+I1129 09:23:46.561232 137274321021824 train.py:125] NOTE: Steps:20100/112603 [17.9%]
+Walltime:1d11h33m (0s eval)
+ETA:6d19h30m
+Total train time:8d7h2m
+I1129 09:28:49.761919 137274321021824 utils.py:1231] [20150] l2_params = 326.0561586453529
+I1129 09:28:49.762122 137274321021824 utils.py:1231] [20150] train/loss = 3.9534584283828735
+I1129 09:28:49.762220 137274321021824 utils.py:1231] [20150] l2_grads = 1.0681617259979248
+I1129 09:28:49.762281 137274321021824 utils.py:1231] [20150] lr = 0.0009760520430606079
+I1129 09:28:49.762331 137274321021824 utils.py:1231] [20150] uptime = 128319.12469368301
+I1129 09:28:49.762383 137274321021824 utils.py:1231] [20150] examples_seen = 20633600.0
+I1129 09:28:49.762433 137274321021824 utils.py:1231] [20150] progress = 0.17894727493938883
+I1129 09:28:49.762487 137274321021824 utils.py:1231] [20150] epoch = 16.105316480989597
+I1129 09:28:49.762537 137274321021824 utils.py:1231] [20150] img/sec/core = 168.864597862974
+I1129 09:28:49.762591 137274321021824 utils.py:1231] [20150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.60988370144416
+I1129 09:28:49.762644 137274321021824 utils.py:1231] [20150] core_hours = 35.60988370144416
+I1129 09:28:49.762702 137274321021824 train.py:125] NOTE: Steps:20150/112603 [17.9%]
+Walltime:1d11h38m (0s eval)
+ETA:6d19h24m
+Total train time:8d7h0m
+I1129 09:33:53.774667 137274321021824 utils.py:1231] [20200] l2_params = 326.13899091948946
+I1129 09:33:53.774951 137274321021824 utils.py:1231] [20200] train/loss = 5.0084228515625
+I1129 09:33:53.775087 137274321021824 utils.py:1231] [20200] l2_grads = 1.1525312662124634
+I1129 09:33:53.775163 137274321021824 utils.py:1231] [20200] lr = 0.0009758174234685112
+I1129 09:33:53.775214 137274321021824 utils.py:1231] [20200] uptime = 128623.1375766
+I1129 09:33:53.775275 137274321021824 utils.py:1231] [20200] examples_seen = 20684800.0
+I1129 09:33:53.775323 137274321021824 utils.py:1231] [20200] progress = 0.17939131284246423
+I1129 09:33:53.775371 137274321021824 utils.py:1231] [20200] epoch = 16.145280045458556
+I1129 09:33:53.775419 137274321021824 utils.py:1231] [20200] img/sec/core = 168.41391558390302
+I1129 09:33:53.775506 137274321021824 utils.py:1231] [20200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.69433172447666
+I1129 09:33:53.775556 137274321021824 utils.py:1231] [20200] core_hours = 35.69433172447666
+I1129 09:33:53.775619 137274321021824 train.py:125] NOTE: Steps:20200/112603 [17.9%]
+Walltime:1d11h43m (0s eval)
+ETA:6d19h17m
+Total train time:8d6h59m
+I1129 09:38:50.710108 137274321021824 utils.py:1231] [20250] l2_params = 326.2415105307557
+I1129 09:38:50.710355 137274321021824 utils.py:1231] [20250] train/loss = 3.1088889241218567
+I1129 09:38:50.710461 137274321021824 utils.py:1231] [20250] l2_grads = 1.2428712844848633
+I1129 09:38:50.710522 137274321021824 utils.py:1231] [20250] lr = 0.0009755816886581035
+I1129 09:38:50.710596 137274321021824 utils.py:1231] [20250] uptime = 128920.07295280501
+I1129 09:38:50.710664 137274321021824 utils.py:1231] [20250] examples_seen = 20736000.0
+I1129 09:38:50.710718 137274321021824 utils.py:1231] [20250] progress = 0.17983535074553963
+I1129 09:38:50.710767 137274321021824 utils.py:1231] [20250] epoch = 16.18524360992751
+I1129 09:38:50.710819 137274321021824 utils.py:1231] [20250] img/sec/core = 172.4280907662868
+I1129 09:38:50.710876 137274321021824 utils.py:1231] [20250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.776813773422504
+I1129 09:38:50.710939 137274321021824 utils.py:1231] [20250] core_hours = 35.776813773422504
+I1129 09:38:50.710999 137274321021824 train.py:125] NOTE: Steps:20250/112603 [18.0%]
+Walltime:1d11h48m (0s eval)
+ETA:6d19h10m
+Total train time:8d6h57m
+I1129 09:43:47.775827 137274321021824 utils.py:1231] [20300] l2_params = 326.3448678808181
+I1129 09:43:47.776035 137274321021824 utils.py:1231] [20300] train/loss = 2.9702707529067993
+I1129 09:43:47.776148 137274321021824 utils.py:1231] [20300] l2_grads = 1.2513604164123535
+I1129 09:43:47.776218 137274321021824 utils.py:1231] [20300] lr = 0.0009753448391818979
+I1129 09:43:47.776277 137274321021824 utils.py:1231] [20300] uptime = 129217.138638172
+I1129 09:43:47.776335 137274321021824 utils.py:1231] [20300] examples_seen = 20787200.0
+I1129 09:43:47.776390 137274321021824 utils.py:1231] [20300] progress = 0.18027938864861504
+I1129 09:43:47.776443 137274321021824 utils.py:1231] [20300] epoch = 16.22520717439647
+I1129 09:43:47.776499 137274321021824 utils.py:1231] [20300] img/sec/core = 172.3524544302392
+I1129 09:43:47.776559 137274321021824 utils.py:1231] [20300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.85933201935777
+I1129 09:43:47.776614 137274321021824 utils.py:1231] [20300] core_hours = 35.85933201935777
+I1129 09:43:47.776679 137274321021824 train.py:125] NOTE: Steps:20300/112603 [18.0%]
+Walltime:1d11h53m (0s eval)
+ETA:6d19h3m
+Total train time:8d6h55m
+I1129 09:48:45.095438 137274321021824 utils.py:1231] [20350] l2_params = 326.4802402136839
+I1129 09:48:45.095725 137274321021824 utils.py:1231] [20350] train/loss = 2.971688538789749
+I1129 09:48:45.095914 137274321021824 utils.py:1231] [20350] l2_grads = 1.3296124935150146
+I1129 09:48:45.095990 137274321021824 utils.py:1231] [20350] lr = 0.0009751068755950216
+I1129 09:48:45.096048 137274321021824 utils.py:1231] [20350] uptime = 129514.458409715
+I1129 09:48:45.096109 137274321021824 utils.py:1231] [20350] examples_seen = 20838400.0
+I1129 09:48:45.096165 137274321021824 utils.py:1231] [20350] progress = 0.18072342655169046
+I1129 09:48:45.096221 137274321021824 utils.py:1231] [20350] epoch = 16.265170738865425
+I1129 09:48:45.096286 137274321021824 utils.py:1231] [20350] img/sec/core = 172.20516393607963
+I1129 09:48:45.096373 137274321021824 utils.py:1231] [20350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 35.94192084478639
+I1129 09:48:45.096438 137274321021824 utils.py:1231] [20350] core_hours = 35.94192084478639
+I1129 09:48:45.096503 137274321021824 train.py:125] NOTE: Steps:20350/112603 [18.1%]
+Walltime:1d11h58m (0s eval)
+ETA:6d18h57m
+Total train time:8d6h53m
+I1129 09:53:51.484006 137274321021824 utils.py:1231] [20400] l2_params = 326.5869910333575
+I1129 09:53:51.484198 137274321021824 utils.py:1231] [20400] train/loss = 5.434939205646515
+I1129 09:53:51.484283 137274321021824 utils.py:1231] [20400] l2_grads = 1.2309684753417969
+I1129 09:53:51.484342 137274321021824 utils.py:1231] [20400] lr = 0.0009748677984552128
+I1129 09:53:51.484390 137274321021824 utils.py:1231] [20400] uptime = 129820.846752711
+I1129 09:53:51.484448 137274321021824 utils.py:1231] [20400] examples_seen = 20889600.0
+I1129 09:53:51.484493 137274321021824 utils.py:1231] [20400] progress = 0.18116746445476586
+I1129 09:53:51.484536 137274321021824 utils.py:1231] [20400] epoch = 16.30513430333438
+I1129 09:53:51.484582 137274321021824 utils.py:1231] [20400] img/sec/core = 167.10818531587057
+I1129 09:53:51.484634 137274321021824 utils.py:1231] [20400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.027028717840835
+I1129 09:53:51.484680 137274321021824 utils.py:1231] [20400] core_hours = 36.027028717840835
+I1129 09:53:51.484735 137274321021824 train.py:125] NOTE: Steps:20400/112603 [18.1%]
+Walltime:1d12h3m (0s eval)
+ETA:6d18h50m
+Total train time:8d6h52m
+I1129 09:58:48.548771 137274321021824 utils.py:1231] [20450] l2_params = 326.6745848947679
+I1129 09:58:48.549035 137274321021824 utils.py:1231] [20450] train/loss = 3.1879460215568542
+I1129 09:58:48.549162 137274321021824 utils.py:1231] [20450] l2_grads = 1.2961355447769165
+I1129 09:58:48.549285 137274321021824 utils.py:1231] [20450] lr = 0.000974627608322818
+I1129 09:58:48.549407 137274321021824 utils.py:1231] [20450] uptime = 130117.91175685001
+I1129 09:58:48.549490 137274321021824 utils.py:1231] [20450] examples_seen = 20940800.0
+I1129 09:58:48.549567 137274321021824 utils.py:1231] [20450] progress = 0.18161150235784126
+I1129 09:58:48.549641 137274321021824 utils.py:1231] [20450] epoch = 16.34509786780334
+I1129 09:58:48.549705 137274321021824 utils.py:1231] [20450] img/sec/core = 172.35284966802195
+I1129 09:58:48.549794 137274321021824 utils.py:1231] [20450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.10954677454611
+I1129 09:58:48.549911 137274321021824 utils.py:1231] [20450] core_hours = 36.10954677454611
+I1129 09:58:48.549984 137274321021824 train.py:125] NOTE: Steps:20450/112603 [18.2%]
+Walltime:1d12h8m (0s eval)
+ETA:6d18h44m
+Total train time:8d6h50m
+I1129 10:03:49.398775 137274321021824 utils.py:1231] [20500] l2_params = 326.746852271574
+I1129 10:03:49.398985 137274321021824 utils.py:1231] [20500] train/loss = 3.0533712208271027
+I1129 10:03:49.399082 137274321021824 utils.py:1231] [20500] l2_grads = 1.3350504636764526
+I1129 10:03:49.399142 137274321021824 utils.py:1231] [20500] lr = 0.0009743863057607944
+I1129 10:03:49.399193 137274321021824 utils.py:1231] [20500] uptime = 130418.76155537301
+I1129 10:03:49.399245 137274321021824 utils.py:1231] [20500] examples_seen = 20992000.0
+I1129 10:03:49.399295 137274321021824 utils.py:1231] [20500] progress = 0.18205554026091667
+I1129 10:03:49.399342 137274321021824 utils.py:1231] [20500] epoch = 16.385061432272295
+I1129 10:03:49.399392 137274321021824 utils.py:1231] [20500] img/sec/core = 170.18459128562364
+I1129 10:03:49.399447 137274321021824 utils.py:1231] [20500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.19311616302472
+I1129 10:03:49.399495 137274321021824 utils.py:1231] [20500] core_hours = 36.19311616302472
+I1129 10:03:49.399556 137274321021824 train.py:125] NOTE: Steps:20500/112603 [18.2%]
+Walltime:1d12h13m (0s eval)
+ETA:6d18h37m
+Total train time:8d6h49m
+I1129 10:08:57.279676 137274321021824 utils.py:1231] [20550] l2_params = 326.8130679454362
+I1129 10:08:57.279930 137274321021824 utils.py:1231] [20550] train/loss = 3.082298755645752
+I1129 10:08:57.280061 137274321021824 utils.py:1231] [20550] l2_grads = 1.3314608335494995
+I1129 10:08:57.280147 137274321021824 utils.py:1231] [20550] lr = 0.0009741438913347054
+I1129 10:08:57.280214 137274321021824 utils.py:1231] [20550] uptime = 130726.642575494
+I1129 10:08:57.280277 137274321021824 utils.py:1231] [20550] examples_seen = 21043200.0
+I1129 10:08:57.280345 137274321021824 utils.py:1231] [20550] progress = 0.18249957816399207
+I1129 10:08:57.280400 137274321021824 utils.py:1231] [20550] epoch = 16.425024996741254
+I1129 10:08:57.280456 137274321021824 utils.py:1231] [20550] img/sec/core = 166.2980068725262
+I1129 10:08:57.280517 137274321021824 utils.py:1231] [20550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.27863866861389
+I1129 10:08:57.280585 137274321021824 utils.py:1231] [20550] core_hours = 36.27863866861389
+I1129 10:08:57.280658 137274321021824 train.py:125] NOTE: Steps:20550/112603 [18.2%]
+Walltime:1d12h18m (0s eval)
+ETA:6d18h31m
+Total train time:8d6h48m
+I1129 10:14:00.494434 137274321021824 utils.py:1231] [20600] l2_params = 326.82595546967264
+I1129 10:14:00.494664 137274321021824 utils.py:1231] [20600] train/loss = 3.0026644468307495
+I1129 10:14:00.494760 137274321021824 utils.py:1231] [20600] l2_grads = 1.3667805194854736
+I1129 10:14:00.494823 137274321021824 utils.py:1231] [20600] lr = 0.0009739003656127207
+I1129 10:14:00.494880 137274321021824 utils.py:1231] [20600] uptime = 131029.85724186101
+I1129 10:14:00.494937 137274321021824 utils.py:1231] [20600] examples_seen = 21094400.0
+I1129 10:14:00.494986 137274321021824 utils.py:1231] [20600] progress = 0.1829436160670675
+I1129 10:14:00.495034 137274321021824 utils.py:1231] [20600] epoch = 16.46498856121021
+I1129 10:14:00.495086 137274321021824 utils.py:1231] [20600] img/sec/core = 168.85726740548583
+I1129 10:14:00.495142 137274321021824 utils.py:1231] [20600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.36286496482695
+I1129 10:14:00.495194 137274321021824 utils.py:1231] [20600] core_hours = 36.36286496482695
+I1129 10:14:00.495254 137274321021824 train.py:125] NOTE: Steps:20600/112603 [18.3%]
+Walltime:1d12h23m (0s eval)
+ETA:6d18h25m
+Total train time:8d6h47m
+I1129 10:19:06.504762 137274321021824 utils.py:1231] [20650] l2_params = 326.8353819772972
+I1129 10:19:06.505031 137274321021824 utils.py:1231] [20650] train/loss = 4.618876338005066
+I1129 10:19:06.505145 137274321021824 utils.py:1231] [20650] l2_grads = 1.0269134044647217
+I1129 10:19:06.505234 137274321021824 utils.py:1231] [20650] lr = 0.0009736557291656145
+I1129 10:19:06.505307 137274321021824 utils.py:1231] [20650] uptime = 131335.867664075
+I1129 10:19:06.505374 137274321021824 utils.py:1231] [20650] examples_seen = 21145600.0
+I1129 10:19:06.505439 137274321021824 utils.py:1231] [20650] progress = 0.1833876539701429
+I1129 10:19:06.505494 137274321021824 utils.py:1231] [20650] epoch = 16.504952125679164
+I1129 10:19:06.505549 137274321021824 utils.py:1231] [20650] img/sec/core = 167.31456278373398
+I1129 10:19:06.505610 137274321021824 utils.py:1231] [20650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.44786785988639
+I1129 10:19:06.505669 137274321021824 utils.py:1231] [20650] core_hours = 36.44786785988639
+I1129 10:19:06.505743 137274321021824 train.py:125] NOTE: Steps:20650/112603 [18.3%]
+Walltime:1d12h28m (0s eval)
+ETA:6d18h18m
+Total train time:8d6h46m
+I1129 10:24:15.495129 137274321021824 utils.py:1231] [20700] l2_params = 326.90447789966225
+I1129 10:24:15.495332 137274321021824 utils.py:1231] [20700] train/loss = 3.281924247741699
+I1129 10:24:15.495445 137274321021824 utils.py:1231] [20700] l2_grads = 1.2520108222961426
+I1129 10:24:15.495515 137274321021824 utils.py:1231] [20700] lr = 0.0009734099825667643
+I1129 10:24:15.495585 137274321021824 utils.py:1231] [20700] uptime = 131644.85794395
+I1129 10:24:15.495651 137274321021824 utils.py:1231] [20700] examples_seen = 21196800.0
+I1129 10:24:15.495713 137274321021824 utils.py:1231] [20700] progress = 0.1838316918732183
+I1129 10:24:15.495769 137274321021824 utils.py:1231] [20700] epoch = 16.544915690148123
+I1129 10:24:15.495827 137274321021824 utils.py:1231] [20700] img/sec/core = 165.70100528959387
+I1129 10:24:15.495905 137274321021824 utils.py:1231] [20700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.53369849318499
+I1129 10:24:15.495975 137274321021824 utils.py:1231] [20700] core_hours = 36.53369849318499
+I1129 10:24:15.496050 137274321021824 train.py:125] NOTE: Steps:20700/112603 [18.4%]
+Walltime:1d12h34m (0s eval)
+ETA:6d18h12m
+Total train time:8d6h45m
+I1129 10:29:20.511218 137274321021824 utils.py:1231] [20750] l2_params = 327.0516802130094
+I1129 10:29:20.511427 137274321021824 utils.py:1231] [20750] train/loss = 3.3527342677116394
+I1129 10:29:20.511531 137274321021824 utils.py:1231] [20750] l2_grads = 1.4677541255950928
+I1129 10:29:20.511593 137274321021824 utils.py:1231] [20750] lr = 0.0009731631263921505
+I1129 10:29:20.511645 137274321021824 utils.py:1231] [20750] uptime = 131949.874007212
+I1129 10:29:20.511701 137274321021824 utils.py:1231] [20750] examples_seen = 21248000.0
+I1129 10:29:20.511751 137274321021824 utils.py:1231] [20750] progress = 0.1842757297762937
+I1129 10:29:20.511800 137274321021824 utils.py:1231] [20750] epoch = 16.58487925461708
+I1129 10:29:20.511852 137274321021824 utils.py:1231] [20750] img/sec/core = 167.8600118709781
+I1129 10:29:20.511916 137274321021824 utils.py:1231] [20750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.61842517742444
+I1129 10:29:20.511970 137274321021824 utils.py:1231] [20750] core_hours = 36.61842517742444
+I1129 10:29:20.512032 137274321021824 train.py:125] NOTE: Steps:20750/112603 [18.4%]
+Walltime:1d12h39m (0s eval)
+ETA:6d18h6m
+Total train time:8d6h44m
+I1129 10:34:25.171757 137274321021824 utils.py:1231] [20800] l2_params = 327.16775104728845
+I1129 10:34:25.172027 137274321021824 utils.py:1231] [20800] train/loss = 3.389565348625183
+I1129 10:34:25.172255 137274321021824 utils.py:1231] [20800] l2_grads = 1.1330680847167969
+I1129 10:34:25.172387 137274321021824 utils.py:1231] [20800] lr = 0.0009729151612203525
+I1129 10:34:25.172506 137274321021824 utils.py:1231] [20800] uptime = 132254.534857094
+I1129 10:34:25.172605 137274321021824 utils.py:1231] [20800] examples_seen = 21299200.0
+I1129 10:34:25.172714 137274321021824 utils.py:1231] [20800] progress = 0.18471976767936912
+I1129 10:34:25.172818 137274321021824 utils.py:1231] [20800] epoch = 16.624842819086037
+I1129 10:34:25.172935 137274321021824 utils.py:1231] [20800] img/sec/core = 168.05572498018958
+I1129 10:34:25.173063 137274321021824 utils.py:1231] [20800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.70305319128055
+I1129 10:34:25.173164 137274321021824 utils.py:1231] [20800] core_hours = 36.70305319128055
+I1129 10:34:25.173308 137274321021824 train.py:125] NOTE: Steps:20800/112603 [18.5%]
+Walltime:1d12h44m (0s eval)
+ETA:6d18h0m
+Total train time:8d6h42m
+I1129 10:39:30.226620 137274321021824 utils.py:1231] [20850] l2_params = 327.22454751623724
+I1129 10:39:30.226816 137274321021824 utils.py:1231] [20850] train/loss = 3.2432928383350372
+I1129 10:39:30.226920 137274321021824 utils.py:1231] [20850] l2_grads = 1.383533000946045
+I1129 10:39:30.226980 137274321021824 utils.py:1231] [20850] lr = 0.0009726660876325496
+I1129 10:39:30.227038 137274321021824 utils.py:1231] [20850] uptime = 132559.58940039
+I1129 10:39:30.227094 137274321021824 utils.py:1231] [20850] examples_seen = 21350400.0
+I1129 10:39:30.227139 137274321021824 utils.py:1231] [20850] progress = 0.18516380558244452
+I1129 10:39:30.227185 137274321021824 utils.py:1231] [20850] epoch = 16.664806383554993
+I1129 10:39:30.227232 137274321021824 utils.py:1231] [20850] img/sec/core = 167.838837759322
+I1129 10:39:30.227284 137274321021824 utils.py:1231] [20850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.78779056441832
+I1129 10:39:30.227335 137274321021824 utils.py:1231] [20850] core_hours = 36.78779056441832
+I1129 10:39:30.227391 137274321021824 train.py:125] NOTE: Steps:20850/112603 [18.5%]
+Walltime:1d12h49m (0s eval)
+ETA:6d17h54m
+Total train time:8d6h41m
+I1129 10:44:30.546635 137274321021824 utils.py:1231] [20900] l2_params = 327.21794891368654
+I1129 10:44:30.546832 137274321021824 utils.py:1231] [20900] train/loss = 3.1489337682724
+I1129 10:44:30.546929 137274321021824 utils.py:1231] [20900] l2_grads = 1.271670937538147
+I1129 10:44:30.547009 137274321021824 utils.py:1231] [20900] lr = 0.0009724159062125193
+I1129 10:44:30.547070 137274321021824 utils.py:1231] [20900] uptime = 132859.90943184902
+I1129 10:44:30.547120 137274321021824 utils.py:1231] [20900] examples_seen = 21401600.0
+I1129 10:44:30.547168 137274321021824 utils.py:1231] [20900] progress = 0.18560784348551992
+I1129 10:44:30.547216 137274321021824 utils.py:1231] [20900] epoch = 16.70476994802395
+I1129 10:44:30.547265 137274321021824 utils.py:1231] [20900] img/sec/core = 170.4847983374986
+I1129 10:44:30.547320 137274321021824 utils.py:1231] [20900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.871212795379165
+I1129 10:44:30.547377 137274321021824 utils.py:1231] [20900] core_hours = 36.871212795379165
+I1129 10:44:30.547438 137274321021824 train.py:125] NOTE: Steps:20900/112603 [18.6%]
+Walltime:1d12h54m (0s eval)
+ETA:6d17h47m
+Total train time:8d6h40m
+I1129 10:49:30.932767 137274321021824 utils.py:1231] [20950] l2_params = 327.29675032327873
+I1129 10:49:30.932967 137274321021824 utils.py:1231] [20950] train/loss = 5.50791335105896
+I1129 10:49:30.933059 137274321021824 utils.py:1231] [20950] l2_grads = 1.1902928352355957
+I1129 10:49:30.933116 137274321021824 utils.py:1231] [20950] lr = 0.0009721646175466357
+I1129 10:49:30.933179 137274321021824 utils.py:1231] [20950] uptime = 133160.295539458
+I1129 10:49:30.933243 137274321021824 utils.py:1231] [20950] examples_seen = 21452800.0
+I1129 10:49:30.933292 137274321021824 utils.py:1231] [20950] progress = 0.18605188138859532
+I1129 10:49:30.933337 137274321021824 utils.py:1231] [20950] epoch = 16.744733512492907
+I1129 10:49:30.933384 137274321021824 utils.py:1231] [20950] img/sec/core = 170.44729667273947
+I1129 10:49:30.933436 137274321021824 utils.py:1231] [20950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 36.95465338082611
+I1129 10:49:30.933481 137274321021824 utils.py:1231] [20950] core_hours = 36.95465338082611
+I1129 10:49:30.933538 137274321021824 train.py:125] NOTE: Steps:20950/112603 [18.6%]
+Walltime:1d12h59m (0s eval)
+ETA:6d17h41m
+Total train time:8d6h38m
+I1129 10:54:31.009585 137274321021824 utils.py:1231] [21000] l2_params = 327.4009616653098
+I1129 10:54:31.009816 137274321021824 utils.py:1231] [21000] train/loss = 5.097145617008209
+I1129 10:54:31.009922 137274321021824 utils.py:1231] [21000] l2_grads = 0.9201270937919617
+I1129 10:54:31.009981 137274321021824 utils.py:1231] [21000] lr = 0.0009719122222238677
+I1129 10:54:31.010045 137274321021824 utils.py:1231] [21000] uptime = 133460.372407121
+I1129 10:54:31.010095 137274321021824 utils.py:1231] [21000] examples_seen = 21504000.0
+I1129 10:54:31.010141 137274321021824 utils.py:1231] [21000] progress = 0.18649591929167073
+I1129 10:54:31.010186 137274321021824 utils.py:1231] [21000] epoch = 16.784697076961862
+I1129 10:54:31.010234 137274321021824 utils.py:1231] [21000] img/sec/core = 170.62294870892262
+I1129 10:54:31.010288 137274321021824 utils.py:1231] [21000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.03800806628805
+I1129 10:54:31.010335 137274321021824 utils.py:1231] [21000] core_hours = 37.03800806628805
+I1129 10:54:31.010392 137274321021824 train.py:125] NOTE: Steps:21000/112603 [18.6%]
+Walltime:1d13h4m (0s eval)
+ETA:6d17h34m
+Total train time:8d6h37m
+I1129 10:59:34.708603 137274321021824 utils.py:1231] [21050] l2_params = 327.48265871380596
+I1129 10:59:34.708870 137274321021824 utils.py:1231] [21050] train/loss = 3.1479936838150024
+I1129 10:59:34.708981 137274321021824 utils.py:1231] [21050] l2_grads = 1.3093698024749756
+I1129 10:59:34.709078 137274321021824 utils.py:1231] [21050] lr = 0.0009716587208357784
+I1129 10:59:34.709162 137274321021824 utils.py:1231] [21050] uptime = 133764.07152360503
+I1129 10:59:34.709234 137274321021824 utils.py:1231] [21050] examples_seen = 21555200.0
+I1129 10:59:34.709295 137274321021824 utils.py:1231] [21050] progress = 0.18693995719474615
+I1129 10:59:34.709369 137274321021824 utils.py:1231] [21050] epoch = 16.82466064143082
+I1129 10:59:34.709431 137274321021824 utils.py:1231] [21050] img/sec/core = 168.58791224931795
+I1129 10:59:34.709502 137274321021824 utils.py:1231] [21050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.12236893197805
+I1129 10:59:34.709556 137274321021824 utils.py:1231] [21050] core_hours = 37.12236893197805
+I1129 10:59:34.709622 137274321021824 train.py:125] NOTE: Steps:21050/112603 [18.7%]
+Walltime:1d13h9m (0s eval)
+ETA:6d17h28m
+Total train time:8d6h35m
+I1129 11:04:38.917604 137274321021824 utils.py:1231] [21100] l2_params = 327.5259278992584
+I1129 11:04:38.917828 137274321021824 utils.py:1231] [21100] train/loss = 3.00521582365036
+I1129 11:04:38.917958 137274321021824 utils.py:1231] [21100] l2_grads = 1.2843466997146606
+I1129 11:04:38.918046 137274321021824 utils.py:1231] [21100] lr = 0.000971404113976522
+I1129 11:04:38.918142 137274321021824 utils.py:1231] [21100] uptime = 134068.28049355902
+I1129 11:04:38.918239 137274321021824 utils.py:1231] [21100] examples_seen = 21606400.0
+I1129 11:04:38.918306 137274321021824 utils.py:1231] [21100] progress = 0.18738399509782155
+I1129 11:04:38.918360 137274321021824 utils.py:1231] [21100] epoch = 16.864624205899776
+I1129 11:04:38.918414 137274321021824 utils.py:1231] [21100] img/sec/core = 168.3053593315907
+I1129 11:04:38.918479 137274321021824 utils.py:1231] [21100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.20687142363194
+I1129 11:04:38.918543 137274321021824 utils.py:1231] [21100] core_hours = 37.20687142363194
+I1129 11:04:38.918607 137274321021824 train.py:125] NOTE: Steps:21100/112603 [18.7%]
+Walltime:1d13h14m (0s eval)
+ETA:6d17h22m
+Total train time:8d6h34m
+I1129 11:09:47.674260 137274321021824 utils.py:1231] [21150] l2_params = 327.56555172685614
+I1129 11:09:47.674515 137274321021824 utils.py:1231] [21150] train/loss = 3.09657022356987
+I1129 11:09:47.674629 137274321021824 utils.py:1231] [21150] l2_grads = 1.2198295593261719
+I1129 11:09:47.674709 137274321021824 utils.py:1231] [21150] lr = 0.0009711484022428453
+I1129 11:09:47.674787 137274321021824 utils.py:1231] [21150] uptime = 134377.037142687
+I1129 11:09:47.674869 137274321021824 utils.py:1231] [21150] examples_seen = 21657600.0
+I1129 11:09:47.674973 137274321021824 utils.py:1231] [21150] progress = 0.18782803300089695
+I1129 11:09:47.675036 137274321021824 utils.py:1231] [21150] epoch = 16.904587770368735
+I1129 11:09:47.675105 137274321021824 utils.py:1231] [21150] img/sec/core = 165.82638833723195
+I1129 11:09:47.675194 137274321021824 utils.py:1231] [21150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.29263715950083
+I1129 11:09:47.675266 137274321021824 utils.py:1231] [21150] core_hours = 37.29263715950083
+I1129 11:09:47.675347 137274321021824 train.py:125] NOTE: Steps:21150/112603 [18.8%]
+Walltime:1d13h19m (0s eval)
+ETA:6d17h16m
+Total train time:8d6h33m
+I1129 11:14:53.454901 137274321021824 utils.py:1231] [21200] l2_params = 327.6665127474937
+I1129 11:14:53.455189 137274321021824 utils.py:1231] [21200] train/loss = 3.080221474170685
+I1129 11:14:53.455304 137274321021824 utils.py:1231] [21200] l2_grads = 1.3186315298080444
+I1129 11:14:53.455377 137274321021824 utils.py:1231] [21200] lr = 0.0009708915862340844
+I1129 11:14:53.455449 137274321021824 utils.py:1231] [21200] uptime = 134682.817810298
+I1129 11:14:53.455501 137274321021824 utils.py:1231] [21200] examples_seen = 21708800.0
+I1129 11:14:53.455552 137274321021824 utils.py:1231] [21200] progress = 0.18827207090397236
+I1129 11:14:53.455599 137274321021824 utils.py:1231] [21200] epoch = 16.94455133483769
+I1129 11:14:53.455649 137274321021824 utils.py:1231] [21200] img/sec/core = 167.44027802676857
+I1129 11:14:53.455704 137274321021824 utils.py:1231] [21200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.37757623383722
+I1129 11:14:53.455751 137274321021824 utils.py:1231] [21200] core_hours = 37.37757623383722
+I1129 11:14:53.455810 137274321021824 train.py:125] NOTE: Steps:21200/112603 [18.8%]
+Walltime:1d13h24m (0s eval)
+ETA:6d17h10m
+Total train time:8d6h32m
+I1129 11:20:03.110377 137274321021824 utils.py:1231] [21250] l2_params = 327.7582986911849
+I1129 11:20:03.110621 137274321021824 utils.py:1231] [21250] train/loss = 3.180527627468109
+I1129 11:20:03.110747 137274321021824 utils.py:1231] [21250] l2_grads = 1.3202357292175293
+I1129 11:20:03.110816 137274321021824 utils.py:1231] [21250] lr = 0.0009706336665521626
+I1129 11:20:03.110876 137274321021824 utils.py:1231] [21250] uptime = 134992.473237661
+I1129 11:20:03.110941 137274321021824 utils.py:1231] [21250] examples_seen = 21760000.0
+I1129 11:20:03.110997 137274321021824 utils.py:1231] [21250] progress = 0.18871610880704778
+I1129 11:20:03.111054 137274321021824 utils.py:1231] [21250] epoch = 16.98451489930665
+I1129 11:20:03.111110 137274321021824 utils.py:1231] [21250] img/sec/core = 165.34507544731036
+I1129 11:20:03.111166 137274321021824 utils.py:1231] [21250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.46359163032694
+I1129 11:20:03.111220 137274321021824 utils.py:1231] [21250] core_hours = 37.46359163032694
+I1129 11:20:03.111285 137274321021824 train.py:125] NOTE: Steps:21250/112603 [18.9%]
+Walltime:1d13h29m (0s eval)
+ETA:6d17h4m
+Total train time:8d6h32m
+I1129 11:25:07.792234 137274321021824 utils.py:1231] [21300] l2_params = 327.84915237054474
+I1129 11:25:07.792512 137274321021824 utils.py:1231] [21300] train/loss = 4.316700220108032
+I1129 11:25:07.792697 137274321021824 utils.py:1231] [21300] l2_grads = 1.081154704093933
+I1129 11:25:07.792799 137274321021824 utils.py:1231] [21300] lr = 0.000970374643801591
+I1129 11:25:07.792866 137274321021824 utils.py:1231] [21300] uptime = 135297.15522657498
+I1129 11:25:07.792956 137274321021824 utils.py:1231] [21300] examples_seen = 21811200.0
+I1129 11:25:07.793050 137274321021824 utils.py:1231] [21300] progress = 0.18916014671012318
+I1129 11:25:07.793132 137274321021824 utils.py:1231] [21300] epoch = 17.024478463775605
+I1129 11:25:07.793199 137274321021824 utils.py:1231] [21300] img/sec/core = 168.0440651660948
+I1129 11:25:07.793273 137274321021824 utils.py:1231] [21300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.548225516136384
+I1129 11:25:07.793364 137274321021824 utils.py:1231] [21300] core_hours = 37.548225516136384
+I1129 11:25:07.793456 137274321021824 train.py:125] NOTE: Steps:21300/112603 [18.9%]
+Walltime:1d13h34m (0s eval)
+ETA:6d16h57m
+Total train time:8d6h31m
+I1129 11:30:13.633627 137274321021824 utils.py:1231] [21350] l2_params = 327.9063885953969
+I1129 11:30:13.633908 137274321021824 utils.py:1231] [21350] train/loss = 3.017299175262451
+I1129 11:30:13.634028 137274321021824 utils.py:1231] [21350] l2_grads = 1.2172774076461792
+I1129 11:30:13.634110 137274321021824 utils.py:1231] [21350] lr = 0.0009701145185894661
+I1129 11:30:13.634197 137274321021824 utils.py:1231] [21350] uptime = 135602.996553462
+I1129 11:30:13.634253 137274321021824 utils.py:1231] [21350] examples_seen = 21862400.0
+I1129 11:30:13.634304 137274321021824 utils.py:1231] [21350] progress = 0.18960418461319858
+I1129 11:30:13.634356 137274321021824 utils.py:1231] [21350] epoch = 17.06444202824456
+I1129 11:30:13.634404 137274321021824 utils.py:1231] [21350] img/sec/core = 167.407068629785
+I1129 11:30:13.634478 137274321021824 utils.py:1231] [21350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.63318144027166
+I1129 11:30:13.634540 137274321021824 utils.py:1231] [21350] core_hours = 37.63318144027166
+I1129 11:30:13.634620 137274321021824 train.py:125] NOTE: Steps:21350/112603 [19.0%]
+Walltime:1d13h40m (0s eval)
+ETA:6d16h51m
+Total train time:8d6h30m
+I1129 11:35:23.600520 137274321021824 utils.py:1231] [21400] l2_params = 327.96844396781364
+I1129 11:35:23.600730 137274321021824 utils.py:1231] [21400] train/loss = 5.092306613922119
+I1129 11:35:23.600816 137274321021824 utils.py:1231] [21400] l2_grads = 0.8990407586097717
+I1129 11:35:23.600870 137274321021824 utils.py:1231] [21400] lr = 0.0009698532915254676
+I1129 11:35:23.600942 137274321021824 utils.py:1231] [21400] uptime = 135912.96329890803
+I1129 11:35:23.601007 137274321021824 utils.py:1231] [21400] examples_seen = 21913600.0
+I1129 11:35:23.601056 137274321021824 utils.py:1231] [21400] progress = 0.19004822251627398
+I1129 11:35:23.601099 137274321021824 utils.py:1231] [21400] epoch = 17.10440559271352
+I1129 11:35:23.601152 137274321021824 utils.py:1231] [21400] img/sec/core = 165.1790095299747
+I1129 11:35:23.601202 137274321021824 utils.py:1231] [21400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.71928331400667
+I1129 11:35:23.601245 137274321021824 utils.py:1231] [21400] core_hours = 37.71928331400667
+I1129 11:35:23.601300 137274321021824 train.py:125] NOTE: Steps:21400/112603 [19.0%]
+Walltime:1d13h45m (0s eval)
+ETA:6d16h46m
+Total train time:8d6h29m
+I1129 11:40:35.367863 137274321021824 utils.py:1231] [21450] l2_params = 328.0015992204311
+I1129 11:40:35.368075 137274321021824 utils.py:1231] [21450] train/loss = 2.9221816062927246
+I1129 11:40:35.368181 137274321021824 utils.py:1231] [21450] l2_grads = 1.2845510244369507
+I1129 11:40:35.368248 137274321021824 utils.py:1231] [21450] lr = 0.0009695909632218582
+I1129 11:40:35.368304 137274321021824 utils.py:1231] [21450] uptime = 136224.73066604603
+I1129 11:40:35.368361 137274321021824 utils.py:1231] [21450] examples_seen = 21964800.0
+I1129 11:40:35.368417 137274321021824 utils.py:1231] [21450] progress = 0.19049226041934939
+I1129 11:40:35.368474 137274321021824 utils.py:1231] [21450] epoch = 17.144369157182474
+I1129 11:40:35.368528 137274321021824 utils.py:1231] [21450] img/sec/core = 164.2250132527064
+I1129 11:40:35.368582 137274321021824 utils.py:1231] [21450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.805885360433884
+I1129 11:40:35.368638 137274321021824 utils.py:1231] [21450] core_hours = 37.805885360433884
+I1129 11:40:35.368721 137274321021824 train.py:125] NOTE: Steps:21450/112603 [19.0%]
+Walltime:1d13h50m (0s eval)
+ETA:6d16h40m
+Total train time:8d6h28m
+I1129 11:45:44.740988 137274321021824 utils.py:1231] [21500] l2_params = 328.08351824986113
+I1129 11:45:44.741226 137274321021824 utils.py:1231] [21500] train/loss = 3.1223250925540924
+I1129 11:45:44.741357 137274321021824 utils.py:1231] [21500] l2_grads = 1.3771657943725586
+I1129 11:45:44.741448 137274321021824 utils.py:1231] [21500] lr = 0.0009693275342934815
+I1129 11:45:44.741536 137274321021824 utils.py:1231] [21500] uptime = 136534.103893732
+I1129 11:45:44.741594 137274321021824 utils.py:1231] [21500] examples_seen = 22016000.0
+I1129 11:45:44.741653 137274321021824 utils.py:1231] [21500] progress = 0.1909362983224248
+I1129 11:45:44.741707 137274321021824 utils.py:1231] [21500] epoch = 17.184332721651433
+I1129 11:45:44.741761 137274321021824 utils.py:1231] [21500] img/sec/core = 165.4958975699373
+I1129 11:45:44.741832 137274321021824 utils.py:1231] [21500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.89182236812444
+I1129 11:45:44.741886 137274321021824 utils.py:1231] [21500] core_hours = 37.89182236812444
+I1129 11:45:44.741953 137274321021824 train.py:125] NOTE: Steps:21500/112603 [19.1%]
+Walltime:1d13h55m (0s eval)
+ETA:6d16h34m
+Total train time:8d6h28m
+I1129 11:50:51.428095 137274321021824 utils.py:1231] [21550] l2_params = 328.1386394595966
+I1129 11:50:51.428330 137274321021824 utils.py:1231] [21550] train/loss = 5.467374801635742
+I1129 11:50:51.428446 137274321021824 utils.py:1231] [21550] l2_grads = 1.001287817955017
+I1129 11:50:51.428521 137274321021824 utils.py:1231] [21550] lr = 0.0009690630053577604
+I1129 11:50:51.428575 137274321021824 utils.py:1231] [21550] uptime = 136840.79093800002
+I1129 11:50:51.428641 137274321021824 utils.py:1231] [21550] examples_seen = 22067200.0
+I1129 11:50:51.428688 137274321021824 utils.py:1231] [21550] progress = 0.1913803362255002
+I1129 11:50:51.428734 137274321021824 utils.py:1231] [21550] epoch = 17.22429628612039
+I1129 11:50:51.428792 137274321021824 utils.py:1231] [21550] img/sec/core = 166.94542843243562
+I1129 11:50:51.428846 137274321021824 utils.py:1231] [21550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 37.97701321375444
+I1129 11:50:51.428904 137274321021824 utils.py:1231] [21550] core_hours = 37.97701321375444
+I1129 11:50:51.428989 137274321021824 train.py:125] NOTE: Steps:21550/112603 [19.1%]
+Walltime:1d14h0m (0s eval)
+ETA:6d16h28m
+Total train time:8d6h27m
+I1129 11:55:57.923291 137274321021824 utils.py:1231] [21600] l2_params = 328.2298852344528
+I1129 11:55:57.923494 137274321021824 utils.py:1231] [21600] train/loss = 3.083576500415802
+I1129 11:55:57.923605 137274321021824 utils.py:1231] [21600] l2_grads = 1.2520439624786377
+I1129 11:55:57.923694 137274321021824 utils.py:1231] [21600] lr = 0.0009687973770346977
+I1129 11:55:57.923756 137274321021824 utils.py:1231] [21600] uptime = 137147.286117999
+I1129 11:55:57.923834 137274321021824 utils.py:1231] [21600] examples_seen = 22118400.0
+I1129 11:55:57.923899 137274321021824 utils.py:1231] [21600] progress = 0.19182437412857561
+I1129 11:55:57.923957 137274321021824 utils.py:1231] [21600] epoch = 17.264259850589344
+I1129 11:55:57.924023 137274321021824 utils.py:1231] [21600] img/sec/core = 167.04993533720702
+I1129 11:55:57.924100 137274321021824 utils.py:1231] [21600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.06215076375416
+I1129 11:55:57.924155 137274321021824 utils.py:1231] [21600] core_hours = 38.06215076375416
+I1129 11:55:57.924225 137274321021824 train.py:125] NOTE: Steps:21600/112603 [19.2%]
+Walltime:1d14h5m (0s eval)
+ETA:6d16h22m
+Total train time:8d6h26m
+I1129 12:01:09.704353 137274321021824 utils.py:1231] [21650] l2_params = 328.310578599609
+I1129 12:01:09.704590 137274321021824 utils.py:1231] [21650] train/loss = 3.0513616800308228
+I1129 12:01:09.704707 137274321021824 utils.py:1231] [21650] l2_grads = 1.3177412748336792
+I1129 12:01:09.704787 137274321021824 utils.py:1231] [21650] lr = 0.0009685306499468692
+I1129 12:01:09.704856 137274321021824 utils.py:1231] [21650] uptime = 137459.06721811998
+I1129 12:01:09.704920 137274321021824 utils.py:1231] [21650] examples_seen = 22169600.0
+I1129 12:01:09.704971 137274321021824 utils.py:1231] [21650] progress = 0.19226841203165101
+I1129 12:01:09.705020 137274321021824 utils.py:1231] [21650] epoch = 17.304223415058303
+I1129 12:01:09.705070 137274321021824 utils.py:1231] [21650] img/sec/core = 164.21777965416553
+I1129 12:01:09.705126 137274321021824 utils.py:1231] [21650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.14875662489888
+I1129 12:01:09.705176 137274321021824 utils.py:1231] [21650] core_hours = 38.14875662489888
+I1129 12:01:09.705235 137274321021824 train.py:125] NOTE: Steps:21650/112603 [19.2%]
+Walltime:1d14h10m (0s eval)
+ETA:6d16h16m
+Total train time:8d6h25m
+I1129 12:06:15.297497 137274321021824 utils.py:1231] [21700] l2_params = 328.37609379178343
+I1129 12:06:15.297758 137274321021824 utils.py:1231] [21700] train/loss = 5.160000562667847
+I1129 12:06:15.297894 137274321021824 utils.py:1231] [21700] l2_grads = 0.9486634135246277
+I1129 12:06:15.297968 137274321021824 utils.py:1231] [21700] lr = 0.0009682628247194309
+I1129 12:06:15.298028 137274321021824 utils.py:1231] [21700] uptime = 137764.66038973298
+I1129 12:06:15.298087 137274321021824 utils.py:1231] [21700] examples_seen = 22220800.0
+I1129 12:06:15.298142 137274321021824 utils.py:1231] [21700] progress = 0.19271244993472642
+I1129 12:06:15.298209 137274321021824 utils.py:1231] [21700] epoch = 17.344186979527258
+I1129 12:06:15.298271 137274321021824 utils.py:1231] [21700] img/sec/core = 167.54301062995992
+I1129 12:06:15.298344 137274321021824 utils.py:1231] [21700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.233643617013605
+I1129 12:06:15.298419 137274321021824 utils.py:1231] [21700] core_hours = 38.233643617013605
+I1129 12:06:15.298500 137274321021824 train.py:125] NOTE: Steps:21700/112603 [19.3%]
+Walltime:1d14h16m (0s eval)
+ETA:6d16h10m
+Total train time:8d6h24m
+I1129 12:11:23.578044 137274321021824 utils.py:1231] [21750] l2_params = 328.42503503866004
+I1129 12:11:23.578274 137274321021824 utils.py:1231] [21750] train/loss = 3.0278473794460297
+I1129 12:11:23.578370 137274321021824 utils.py:1231] [21750] l2_grads = 1.450813889503479
+I1129 12:11:23.578433 137274321021824 utils.py:1231] [21750] lr = 0.0009679939019801087
+I1129 12:11:23.578488 137274321021824 utils.py:1231] [21750] uptime = 138072.940849958
+I1129 12:11:23.578539 137274321021824 utils.py:1231] [21750] examples_seen = 22272000.0
+I1129 12:11:23.578619 137274321021824 utils.py:1231] [21750] progress = 0.19315648783780184
+I1129 12:11:23.578671 137274321021824 utils.py:1231] [21750] epoch = 17.384150543996217
+I1129 12:11:23.578719 137274321021824 utils.py:1231] [21750] img/sec/core = 166.08253394532252
+I1129 12:11:23.578773 137274321021824 utils.py:1231] [21750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.31927707818722
+I1129 12:11:23.578821 137274321021824 utils.py:1231] [21750] core_hours = 38.31927707818722
+I1129 12:11:23.578879 137274321021824 train.py:125] NOTE: Steps:21750/112603 [19.3%]
+Walltime:1d14h21m (0s eval)
+ETA:6d16h4m
+Total train time:8d6h24m
+I1129 12:16:30.565980 137274321021824 utils.py:1231] [21800] l2_params = 328.49195082433664
+I1129 12:16:30.566182 137274321021824 utils.py:1231] [21800] train/loss = 3.043752670288086
+I1129 12:16:30.566280 137274321021824 utils.py:1231] [21800] l2_grads = 1.3129953145980835
+I1129 12:16:30.566338 137274321021824 utils.py:1231] [21800] lr = 0.0009677238823592026
+I1129 12:16:30.566389 137274321021824 utils.py:1231] [21800] uptime = 138379.92875066597
+I1129 12:16:30.566441 137274321021824 utils.py:1231] [21800] examples_seen = 22323200.0
+I1129 12:16:30.566489 137274321021824 utils.py:1231] [21800] progress = 0.19360052574087724
+I1129 12:16:30.566536 137274321021824 utils.py:1231] [21800] epoch = 17.424114108465172
+I1129 12:16:30.566585 137274321021824 utils.py:1231] [21800] img/sec/core = 166.7818174003621
+I1129 12:16:30.566638 137274321021824 utils.py:1231] [21800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.40455149505055
+I1129 12:16:30.566686 137274321021824 utils.py:1231] [21800] core_hours = 38.40455149505055
+I1129 12:16:30.566744 137274321021824 train.py:125] NOTE: Steps:21800/112603 [19.4%]
+Walltime:1d14h26m (0s eval)
+ETA:6d15h58m
+Total train time:8d6h23m
+I1129 12:21:40.615221 137274321021824 utils.py:1231] [21850] l2_params = 328.59158961283805
+I1129 12:21:40.615416 137274321021824 utils.py:1231] [21850] train/loss = 3.174329310655594
+I1129 12:21:40.615505 137274321021824 utils.py:1231] [21850] l2_grads = 1.3976562023162842
+I1129 12:21:40.615561 137274321021824 utils.py:1231] [21850] lr = 0.0009674527664895842
+I1129 12:21:40.615610 137274321021824 utils.py:1231] [21850] uptime = 138689.977972787
+I1129 12:21:40.615660 137274321021824 utils.py:1231] [21850] examples_seen = 22374400.0
+I1129 12:21:40.615706 137274321021824 utils.py:1231] [21850] progress = 0.19404456364395264
+I1129 12:21:40.615751 137274321021824 utils.py:1231] [21850] epoch = 17.46407767293413
+I1129 12:21:40.615798 137274321021824 utils.py:1231] [21850] img/sec/core = 165.13507000516242
+I1129 12:21:40.615853 137274321021824 utils.py:1231] [21850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.49067627897305
+I1129 12:21:40.615904 137274321021824 utils.py:1231] [21850] core_hours = 38.49067627897305
+I1129 12:21:40.615962 137274321021824 train.py:125] NOTE: Steps:21850/112603 [19.4%]
+Walltime:1d14h31m (0s eval)
+ETA:6d15h53m
+Total train time:8d6h22m
+I1129 12:26:52.396602 137274321021824 utils.py:1231] [21900] l2_params = 328.66521674167984
+I1129 12:26:52.396840 137274321021824 utils.py:1231] [21900] train/loss = 5.175489604473114
+I1129 12:26:52.396957 137274321021824 utils.py:1231] [21900] l2_grads = 0.8827363848686218
+I1129 12:26:52.397023 137274321021824 utils.py:1231] [21900] lr = 0.0009671805550066915
+I1129 12:26:52.397084 137274321021824 utils.py:1231] [21900] uptime = 139001.75944314903
+I1129 12:26:52.397143 137274321021824 utils.py:1231] [21900] examples_seen = 22425600.0
+I1129 12:26:52.397190 137274321021824 utils.py:1231] [21900] progress = 0.19448860154702804
+I1129 12:26:52.397240 137274321021824 utils.py:1231] [21900] epoch = 17.504041237403086
+I1129 12:26:52.397304 137274321021824 utils.py:1231] [21900] img/sec/core = 164.21758464526977
+I1129 12:26:52.397358 137274321021824 utils.py:1231] [21900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.5772822429625
+I1129 12:26:52.397406 137274321021824 utils.py:1231] [21900] core_hours = 38.5772822429625
+I1129 12:26:52.397465 137274321021824 train.py:125] NOTE: Steps:21900/112603 [19.4%]
+Walltime:1d14h36m (0s eval)
+ETA:6d15h47m
+Total train time:8d6h22m
+I1129 12:32:04.190390 137274321021824 utils.py:1231] [21950] l2_params = 328.70058187364185
+I1129 12:32:04.190646 137274321021824 utils.py:1231] [21950] train/loss = 2.9398085474967957
+I1129 12:32:04.190788 137274321021824 utils.py:1231] [21950] l2_grads = 1.3231555223464966
+I1129 12:32:04.190878 137274321021824 utils.py:1231] [21950] lr = 0.0009669072485485337
+I1129 12:32:04.190970 137274321021824 utils.py:1231] [21950] uptime = 139313.55333114398
+I1129 12:32:04.191057 137274321021824 utils.py:1231] [21950] examples_seen = 22476800.0
+I1129 12:32:04.191133 137274321021824 utils.py:1231] [21950] progress = 0.19493263945010347
+I1129 12:32:04.191201 137274321021824 utils.py:1231] [21950] epoch = 17.544004801872042
+I1129 12:32:04.191256 137274321021824 utils.py:1231] [21950] img/sec/core = 164.21104444749213
+I1129 12:32:04.191315 137274321021824 utils.py:1231] [21950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.66389165629444
+I1129 12:32:04.191370 137274321021824 utils.py:1231] [21950] core_hours = 38.66389165629444
+I1129 12:32:04.191434 137274321021824 train.py:125] NOTE: Steps:21950/112603 [19.5%]
+Walltime:1d14h41m (0s eval)
+ETA:6d15h41m
+Total train time:8d6h21m
+I1129 12:37:13.403490 137274321021824 utils.py:1231] [22000] l2_params = 328.7468319645505
+I1129 12:37:13.403783 137274321021824 utils.py:1231] [22000] train/loss = 3.0623638927936554
+I1129 12:37:13.403987 137274321021824 utils.py:1231] [22000] l2_grads = 1.2562675476074219
+I1129 12:37:13.404081 137274321021824 utils.py:1231] [22000] lr = 0.0009666328477556844
+I1129 12:37:13.404153 137274321021824 utils.py:1231] [22000] uptime = 139622.766512412
+I1129 12:37:13.404242 137274321021824 utils.py:1231] [22000] examples_seen = 22528000.0
+I1129 12:37:13.404341 137274321021824 utils.py:1231] [22000] progress = 0.19537667735317887
+I1129 12:37:13.404424 137274321021824 utils.py:1231] [22000] epoch = 17.583968366341
+I1129 12:37:13.404508 137274321021824 utils.py:1231] [22000] img/sec/core = 165.58155700232402
+I1129 12:37:13.404589 137274321021824 utils.py:1231] [22000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.74978420664666
+I1129 12:37:13.404670 137274321021824 utils.py:1231] [22000] core_hours = 38.74978420664666
+I1129 12:37:13.404760 137274321021824 train.py:125] NOTE: Steps:22000/112603 [19.5%]
+Walltime:1d14h47m (0s eval)
+ETA:6d15h35m
+Total train time:8d6h21m
+I1129 12:42:21.969575 137274321021824 utils.py:1231] [22050] l2_params = 328.77005817233095
+I1129 12:42:21.969853 137274321021824 utils.py:1231] [22050] train/loss = 3.1333001852035522
+I1129 12:42:21.969981 137274321021824 utils.py:1231] [22050] l2_grads = 1.3831939697265625
+I1129 12:42:21.970062 137274321021824 utils.py:1231] [22050] lr = 0.000966357353271283
+I1129 12:42:21.970122 137274321021824 utils.py:1231] [22050] uptime = 139931.33248448203
+I1129 12:42:21.970174 137274321021824 utils.py:1231] [22050] examples_seen = 22579200.0
+I1129 12:42:21.970230 137274321021824 utils.py:1231] [22050] progress = 0.19582071525625427
+I1129 12:42:21.970291 137274321021824 utils.py:1231] [22050] epoch = 17.623931930809956
+I1129 12:42:21.970341 137274321021824 utils.py:1231] [22050] img/sec/core = 165.92886006362357
+I1129 12:42:21.970397 137274321021824 utils.py:1231] [22050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.835496976666114
+I1129 12:42:21.970448 137274321021824 utils.py:1231] [22050] core_hours = 38.835496976666114
+I1129 12:42:21.970517 137274321021824 train.py:125] NOTE: Steps:22050/112603 [19.6%]
+Walltime:1d14h52m (0s eval)
+ETA:6d15h30m
+Total train time:8d6h20m
+I1129 12:47:33.757106 137274321021824 utils.py:1231] [22100] l2_params = 328.8501553283662
+I1129 12:47:33.757329 137274321021824 utils.py:1231] [22100] train/loss = 5.280042886734009
+I1129 12:47:33.757439 137274321021824 utils.py:1231] [22100] l2_grads = 1.3492155075073242
+I1129 12:47:33.757509 137274321021824 utils.py:1231] [22100] lr = 0.000966080765741032
+I1129 12:47:33.757567 137274321021824 utils.py:1231] [22100] uptime = 140243.119929038
+I1129 12:47:33.757642 137274321021824 utils.py:1231] [22100] examples_seen = 22630400.0
+I1129 12:47:33.757714 137274321021824 utils.py:1231] [22100] progress = 0.19626475315932967
+I1129 12:47:33.757769 137274321021824 utils.py:1231] [22100] epoch = 17.663895495278915
+I1129 12:47:33.757843 137274321021824 utils.py:1231] [22100] img/sec/core = 164.21443805382683
+I1129 12:47:33.757913 137274321021824 utils.py:1231] [22100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 38.92210460015388
+I1129 12:47:33.757982 137274321021824 utils.py:1231] [22100] core_hours = 38.92210460015388
+I1129 12:47:33.758058 137274321021824 train.py:125] NOTE: Steps:22100/112603 [19.6%]
+Walltime:1d14h57m (0s eval)
+ETA:6d15h24m
+Total train time:8d6h19m
+I1129 12:52:45.526420 137274321021824 utils.py:1231] [22150] l2_params = 328.91774470963594
+I1129 12:52:45.526619 137274321021824 utils.py:1231] [22150] train/loss = 3.0084464251995087
+I1129 12:52:45.526725 137274321021824 utils.py:1231] [22150] l2_grads = 1.393335223197937
+I1129 12:52:45.526800 137274321021824 utils.py:1231] [22150] lr = 0.0009658030858131953
+I1129 12:52:45.526859 137274321021824 utils.py:1231] [22150] uptime = 140554.88922068902
+I1129 12:52:45.526932 137274321021824 utils.py:1231] [22150] examples_seen = 22681600.0
+I1129 12:52:45.526985 137274321021824 utils.py:1231] [22150] progress = 0.19670879106240508
+I1129 12:52:45.527036 137274321021824 utils.py:1231] [22150] epoch = 17.70385905974787
+I1129 12:52:45.527088 137274321021824 utils.py:1231] [22150] img/sec/core = 164.22399951215976
+I1129 12:52:45.527146 137274321021824 utils.py:1231] [22150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.00870718116805
+I1129 12:52:45.527204 137274321021824 utils.py:1231] [22150] core_hours = 39.00870718116805
+I1129 12:52:45.527263 137274321021824 train.py:125] NOTE: Steps:22150/112603 [19.7%]
+Walltime:1d15h2m (0s eval)
+ETA:6d15h18m
+Total train time:8d6h19m
+I1129 12:57:57.298424 137274321021824 utils.py:1231] [22200] l2_params = 328.96518799604627
+I1129 12:57:57.298650 137274321021824 utils.py:1231] [22200] train/loss = 5.003807902336121
+I1129 12:57:57.298748 137274321021824 utils.py:1231] [22200] l2_grads = 1.1649739742279053
+I1129 12:57:57.298816 137274321021824 utils.py:1231] [22200] lr = 0.0009655243141385982
+I1129 12:57:57.298891 137274321021824 utils.py:1231] [22200] uptime = 140866.66124644998
+I1129 12:57:57.298966 137274321021824 utils.py:1231] [22200] examples_seen = 22732800.0
+I1129 12:57:57.299022 137274321021824 utils.py:1231] [22200] progress = 0.1971528289654805
+I1129 12:57:57.299077 137274321021824 utils.py:1231] [22200] epoch = 17.743822624216826
+I1129 12:57:57.299132 137274321021824 utils.py:1231] [22200] img/sec/core = 164.22255933651564
+I1129 12:57:57.299189 137274321021824 utils.py:1231] [22200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.09531052165721
+I1129 12:57:57.299242 137274321021824 utils.py:1231] [22200] core_hours = 39.09531052165721
+I1129 12:57:57.299321 137274321021824 train.py:125] NOTE: Steps:22200/112603 [19.7%]
+Walltime:1d15h7m (0s eval)
+ETA:6d15h13m
+Total train time:8d6h19m
+I1129 13:03:09.050596 137274321021824 utils.py:1231] [22250] l2_params = 329.0427161357456
+I1129 13:03:09.050810 137274321021824 utils.py:1231] [22250] train/loss = 5.318931937217712
+I1129 13:03:09.050915 137274321021824 utils.py:1231] [22250] l2_grads = 1.0858615636825562
+I1129 13:03:09.050998 137274321021824 utils.py:1231] [22250] lr = 0.0009652444513706237
+I1129 13:03:09.051069 137274321021824 utils.py:1231] [22250] uptime = 141178.41343023803
+I1129 13:03:09.051126 137274321021824 utils.py:1231] [22250] examples_seen = 22784000.0
+I1129 13:03:09.051186 137274321021824 utils.py:1231] [22250] progress = 0.1975968668685559
+I1129 13:03:09.051239 137274321021824 utils.py:1231] [22250] epoch = 17.783786188685784
+I1129 13:03:09.051293 137274321021824 utils.py:1231] [22250] img/sec/core = 164.23301154744823
+I1129 13:03:09.051350 137274321021824 utils.py:1231] [22250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.18190835048722
+I1129 13:03:09.051398 137274321021824 utils.py:1231] [22250] core_hours = 39.18190835048722
+I1129 13:03:09.051488 137274321021824 train.py:125] NOTE: Steps:22250/112603 [19.8%]
+Walltime:1d15h12m (0s eval)
+ETA:6d15h7m
+Total train time:8d6h18m
+I1129 13:08:20.831611 137274321021824 utils.py:1231] [22300] l2_params = 329.0789428311535
+I1129 13:08:20.831871 137274321021824 utils.py:1231] [22300] train/loss = 3.0355974435806274
+I1129 13:08:20.832011 137274321021824 utils.py:1231] [22300] l2_grads = 1.4243907928466797
+I1129 13:08:20.832077 137274321021824 utils.py:1231] [22300] lr = 0.000964963498165213
+I1129 13:08:20.832129 137274321021824 utils.py:1231] [22300] uptime = 141490.19449067203
+I1129 13:08:20.832180 137274321021824 utils.py:1231] [22300] examples_seen = 22835200.0
+I1129 13:08:20.832227 137274321021824 utils.py:1231] [22300] progress = 0.1980409047716313
+I1129 13:08:20.832274 137274321021824 utils.py:1231] [22300] epoch = 17.82374975315474
+I1129 13:08:20.832323 137274321021824 utils.py:1231] [22300] img/sec/core = 164.21780055763713
+I1129 13:08:20.832379 137274321021824 utils.py:1231] [22300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.268514200607775
+I1129 13:08:20.832429 137274321021824 utils.py:1231] [22300] core_hours = 39.268514200607775
+I1129 13:08:20.832488 137274321021824 train.py:125] NOTE: Steps:22300/112603 [19.8%]
+Walltime:1d15h18m (0s eval)
+ETA:6d15h1m
+Total train time:8d6h18m
+I1129 13:13:32.644946 137274321021824 utils.py:1231] [22350] l2_params = 329.127900821722
+I1129 13:13:32.645163 137274321021824 utils.py:1231] [22350] train/loss = 3.4080249965190887
+I1129 13:13:32.645264 137274321021824 utils.py:1231] [22350] l2_grads = 1.2728983163833618
+I1129 13:13:32.645337 137274321021824 utils.py:1231] [22350] lr = 0.0009646814551808627
+I1129 13:13:32.645388 137274321021824 utils.py:1231] [22350] uptime = 141802.007749899
+I1129 13:13:32.645438 137274321021824 utils.py:1231] [22350] examples_seen = 22886400.0
+I1129 13:13:32.645487 137274321021824 utils.py:1231] [22350] progress = 0.1984849426747067
+I1129 13:13:32.645535 137274321021824 utils.py:1231] [22350] epoch = 17.8637133176237
+I1129 13:13:32.645585 137274321021824 utils.py:1231] [22350] img/sec/core = 164.2008429241516
+I1129 13:13:32.645639 137274321021824 utils.py:1231] [22350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.3551289948375
+I1129 13:13:32.645688 137274321021824 utils.py:1231] [22350] core_hours = 39.3551289948375
+I1129 13:13:32.645746 137274321021824 train.py:125] NOTE: Steps:22350/112603 [19.8%]
+Walltime:1d15h23m (0s eval)
+ETA:6d14h56m
+Total train time:8d6h17m
+I1129 13:18:44.428951 137274321021824 utils.py:1231] [22400] l2_params = 329.2311873200868
+I1129 13:18:44.429207 137274321021824 utils.py:1231] [22400] train/loss = 3.4833216965198517
+I1129 13:18:44.429397 137274321021824 utils.py:1231] [22400] l2_grads = 1.1407510042190552
+I1129 13:18:44.429501 137274321021824 utils.py:1231] [22400] lr = 0.0009643983230786233
+I1129 13:18:44.429596 137274321021824 utils.py:1231] [22400] uptime = 142113.79195747402
+I1129 13:18:44.429666 137274321021824 utils.py:1231] [22400] examples_seen = 22937600.0
+I1129 13:18:44.429718 137274321021824 utils.py:1231] [22400] progress = 0.19892898057778213
+I1129 13:18:44.429778 137274321021824 utils.py:1231] [22400] epoch = 17.903676882092654
+I1129 13:18:44.429840 137274321021824 utils.py:1231] [22400] img/sec/core = 164.2161429477943
+I1129 13:18:44.429905 137274321021824 utils.py:1231] [22400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.44173571916388
+I1129 13:18:44.429972 137274321021824 utils.py:1231] [22400] core_hours = 39.44173571916388
+I1129 13:18:44.430031 137274321021824 train.py:125] NOTE: Steps:22400/112603 [19.9%]
+Walltime:1d15h28m (0s eval)
+ETA:6d14h50m
+Total train time:8d6h17m
+I1129 13:23:52.569507 137274321021824 utils.py:1231] [22450] l2_params = 329.28443007025555
+I1129 13:23:52.569763 137274321021824 utils.py:1231] [22450] train/loss = 4.704208254814148
+I1129 13:23:52.569996 137274321021824 utils.py:1231] [22450] l2_grads = 1.0013866424560547
+I1129 13:23:52.570095 137274321021824 utils.py:1231] [22450] lr = 0.0009641141025220983
+I1129 13:23:52.570158 137274321021824 utils.py:1231] [22450] uptime = 142421.932519847
+I1129 13:23:52.570229 137274321021824 utils.py:1231] [22450] examples_seen = 22988800.0
+I1129 13:23:52.570285 137274321021824 utils.py:1231] [22450] progress = 0.19937301848085753
+I1129 13:23:52.570338 137274321021824 utils.py:1231] [22450] epoch = 17.943640446561613
+I1129 13:23:52.570387 137274321021824 utils.py:1231] [22450] img/sec/core = 166.15793651347164
+I1129 13:23:52.570441 137274321021824 utils.py:1231] [22450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.52733031982305
+I1129 13:23:52.570489 137274321021824 utils.py:1231] [22450] core_hours = 39.52733031982305
+I1129 13:23:52.570546 137274321021824 train.py:125] NOTE: Steps:22450/112603 [19.9%]
+Walltime:1d15h33m (0s eval)
+ETA:6d14h44m
+Total train time:8d6h16m
+I1129 13:29:04.424374 137274321021824 utils.py:1231] [22500] l2_params = 329.3252577085612
+I1129 13:29:04.424631 137274321021824 utils.py:1231] [22500] train/loss = 3.1354830265045166
+I1129 13:29:04.424747 137274321021824 utils.py:1231] [22500] l2_grads = 1.2638496160507202
+I1129 13:29:04.424821 137274321021824 utils.py:1231] [22500] lr = 0.0009638287941774426
+I1129 13:29:04.424889 137274321021824 utils.py:1231] [22500] uptime = 142733.787244518
+I1129 13:29:04.424953 137274321021824 utils.py:1231] [22500] examples_seen = 23040000.0
+I1129 13:29:04.425010 137274321021824 utils.py:1231] [22500] progress = 0.19981705638393293
+I1129 13:29:04.425066 137274321021824 utils.py:1231] [22500] epoch = 17.983604011030568
+I1129 13:29:04.425124 137274321021824 utils.py:1231] [22500] img/sec/core = 164.17901012728288
+I1129 13:29:04.425187 137274321021824 utils.py:1231] [22500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.61395663223166
+I1129 13:29:04.425244 137274321021824 utils.py:1231] [22500] core_hours = 39.61395663223166
+I1129 13:29:04.425309 137274321021824 train.py:125] NOTE: Steps:22500/112603 [20.0%]
+Walltime:1d15h38m (0s eval)
+ETA:6d14h39m
+Total train time:8d6h16m
+I1129 13:29:04.425419 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:22500/112603 [20.0%]
+Walltime:1d15h38m (0s eval)
+ETA:6d14h39m
+Total train time:8d6h16m
+I1129 13:30:43.845673 137274321021824 utils.py:1231] [22500] val/acc@1 = 0.5332629145408163
+I1129 13:30:43.845937 137274321021824 utils.py:1231] [22500] val/loss = 1.9990800156885264
+I1129 13:30:43.846127 137274321021824 utils.py:1231] [22500] z/secs/eval/val = 99.42062883399194
+I1129 13:30:43.846192 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 99.42062883399194
+I1129 13:35:55.677264 137274321021824 utils.py:1231] [22550] l2_params = 329.4530139542072
+I1129 13:35:55.677522 137274321021824 utils.py:1231] [22550] train/loss = 2.941831797361374
+I1129 13:35:55.677651 137274321021824 utils.py:1231] [22550] l2_grads = 1.3445862531661987
+I1129 13:35:55.677742 137274321021824 utils.py:1231] [22550] lr = 0.0009635423987133605
+I1129 13:35:55.677819 137274321021824 utils.py:1231] [22550] uptime = 143145.04017303698
+I1129 13:35:55.677903 137274321021824 utils.py:1231] [22550] examples_seen = 23091200.0
+I1129 13:35:55.677962 137274321021824 utils.py:1231] [22550] progress = 0.20026109428700833
+I1129 13:35:55.678018 137274321021824 utils.py:1231] [22550] epoch = 18.023567575499523
+I1129 13:35:55.678076 137274321021824 utils.py:1231] [22550] img/sec/core = 124.49759369345594
+I1129 13:35:55.678138 137274321021824 utils.py:1231] [22550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.72819355682027
+I1129 13:35:55.678195 137274321021824 utils.py:1231] [22550] core_hours = 39.72819355682027
+I1129 13:35:55.678261 137274321021824 train.py:125] NOTE: Steps:22550/112603 [20.0%]
+Walltime:1d15h45m (0s eval)
+ETA:6d14h40m
+Total train time:8d6h23m
+I1129 13:41:07.517313 137274321021824 utils.py:1231] [22600] l2_params = 329.4915782928714
+I1129 13:41:07.517639 137274321021824 utils.py:1231] [22600] train/loss = 5.206822574138641
+I1129 13:41:07.517818 137274321021824 utils.py:1231] [22600] l2_grads = 1.0258712768554688
+I1129 13:41:07.517910 137274321021824 utils.py:1231] [22600] lr = 0.0009632549168011038
+I1129 13:41:07.517972 137274321021824 utils.py:1231] [22600] uptime = 143456.880332998
+I1129 13:41:07.518033 137274321021824 utils.py:1231] [22600] examples_seen = 23142400.0
+I1129 13:41:07.518090 137274321021824 utils.py:1231] [22600] progress = 0.20070513219008373
+I1129 13:41:07.518145 137274321021824 utils.py:1231] [22600] epoch = 18.063531139968482
+I1129 13:41:07.518203 137274321021824 utils.py:1231] [22600] img/sec/core = 164.18667822130286
+I1129 13:41:07.518267 137274321021824 utils.py:1231] [22600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.81481582347611
+I1129 13:41:07.518326 137274321021824 utils.py:1231] [22600] core_hours = 39.81481582347611
+I1129 13:41:07.518396 137274321021824 train.py:125] NOTE: Steps:22600/112603 [20.1%]
+Walltime:1d15h50m (0s eval)
+ETA:6d14h34m
+Total train time:8d6h23m
+I1129 13:46:19.347205 137274321021824 utils.py:1231] [22650] l2_params = 329.5165929383051
+I1129 13:46:19.347416 137274321021824 utils.py:1231] [22650] train/loss = 3.0248874723911285
+I1129 13:46:19.347505 137274321021824 utils.py:1231] [22650] l2_grads = 1.4074065685272217
+I1129 13:46:19.347563 137274321021824 utils.py:1231] [22650] lr = 0.0009629663491144714
+I1129 13:46:19.347615 137274321021824 utils.py:1231] [22650] uptime = 143768.70997671603
+I1129 13:46:19.347666 137274321021824 utils.py:1231] [22650] examples_seen = 23193600.0
+I1129 13:46:19.347716 137274321021824 utils.py:1231] [22650] progress = 0.20114917009315916
+I1129 13:46:19.347764 137274321021824 utils.py:1231] [22650] epoch = 18.103494704437438
+I1129 13:46:19.347817 137274321021824 utils.py:1231] [22650] img/sec/core = 164.1922153055559
+I1129 13:46:19.347875 137274321021824 utils.py:1231] [22650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.90143516895333
+I1129 13:46:19.347931 137274321021824 utils.py:1231] [22650] core_hours = 39.90143516895333
+I1129 13:46:19.347993 137274321021824 train.py:125] NOTE: Steps:22650/112603 [20.1%]
+Walltime:1d15h56m (0s eval)
+ETA:6d14h28m
+Total train time:8d6h23m
+I1129 13:51:31.147781 137274321021824 utils.py:1231] [22700] l2_params = 329.58881585261804
+I1129 13:51:31.147999 137274321021824 utils.py:1231] [22700] train/loss = 3.0917303562164307
+I1129 13:51:31.148096 137274321021824 utils.py:1231] [22700] l2_grads = 1.2840121984481812
+I1129 13:51:31.148156 137274321021824 utils.py:1231] [22700] lr = 0.0009626766963298065
+I1129 13:51:31.148211 137274321021824 utils.py:1231] [22700] uptime = 144080.51057230303
+I1129 13:51:31.148267 137274321021824 utils.py:1231] [22700] examples_seen = 23244800.0
+I1129 13:51:31.148319 137274321021824 utils.py:1231] [22700] progress = 0.20159320799623456
+I1129 13:51:31.148371 137274321021824 utils.py:1231] [22700] epoch = 18.143458268906397
+I1129 13:51:31.148425 137274321021824 utils.py:1231] [22700] img/sec/core = 164.20751186703058
+I1129 13:51:31.148483 137274321021824 utils.py:1231] [22700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 39.98804644550527
+I1129 13:51:31.148559 137274321021824 utils.py:1231] [22700] core_hours = 39.98804644550527
+I1129 13:51:31.148626 137274321021824 train.py:125] NOTE: Steps:22700/112603 [20.2%]
+Walltime:1d16h1m (0s eval)
+ETA:6d14h23m
+Total train time:8d6h22m
+I1129 13:56:42.976732 137274321021824 utils.py:1231] [22750] l2_params = 329.642222933068
+I1129 13:56:42.976962 137274321021824 utils.py:1231] [22750] train/loss = 2.9898690283298492
+I1129 13:56:42.977067 137274321021824 utils.py:1231] [22750] l2_grads = 1.3470057249069214
+I1129 13:56:42.977128 137274321021824 utils.py:1231] [22750] lr = 0.0009623859591259959
+I1129 13:56:42.977180 137274321021824 utils.py:1231] [22750] uptime = 144392.33954159298
+I1129 13:56:42.977233 137274321021824 utils.py:1231] [22750] examples_seen = 23296000.0
+I1129 13:56:42.977283 137274321021824 utils.py:1231] [22750] progress = 0.20203724589930996
+I1129 13:56:42.977333 137274321021824 utils.py:1231] [22750] epoch = 18.183421833375352
+I1129 13:56:42.977383 137274321021824 utils.py:1231] [22750] img/sec/core = 164.19257042275177
+I1129 13:56:42.977439 137274321021824 utils.py:1231] [22750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.074665603641385
+I1129 13:56:42.977491 137274321021824 utils.py:1231] [22750] core_hours = 40.074665603641385
+I1129 13:56:42.977552 137274321021824 train.py:125] NOTE: Steps:22750/112603 [20.2%]
+Walltime:1d16h6m (0s eval)
+ETA:6d14h17m
+Total train time:8d6h22m
+I1129 14:01:54.814134 137274321021824 utils.py:1231] [22800] l2_params = 329.7097956286
+I1129 14:01:54.814437 137274321021824 utils.py:1231] [22800] train/loss = 3.2685332596302032
+I1129 14:01:54.814604 137274321021824 utils.py:1231] [22800] l2_grads = 1.1757797002792358
+I1129 14:01:54.814680 137274321021824 utils.py:1231] [22800] lr = 0.0009620941381844675
+I1129 14:01:54.814734 137274321021824 utils.py:1231] [22800] uptime = 144704.177095782
+I1129 14:01:54.814799 137274321021824 utils.py:1231] [22800] examples_seen = 23347200.0
+I1129 14:01:54.814847 137274321021824 utils.py:1231] [22800] progress = 0.20248128380238536
+I1129 14:01:54.814907 137274321021824 utils.py:1231] [22800] epoch = 18.22338539784431
+I1129 14:01:54.814967 137274321021824 utils.py:1231] [22800] img/sec/core = 164.18805019541924
+I1129 14:01:54.815025 137274321021824 utils.py:1231] [22800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.16128714647166
+I1129 14:01:54.815091 137274321021824 utils.py:1231] [22800] core_hours = 40.16128714647166
+I1129 14:01:54.815156 137274321021824 train.py:125] NOTE: Steps:22800/112603 [20.2%]
+Walltime:1d16h11m (0s eval)
+ETA:6d14h11m
+Total train time:8d6h21m
+I1129 14:07:06.668031 137274321021824 utils.py:1231] [22850] l2_params = 329.7530317157661
+I1129 14:07:06.668346 137274321021824 utils.py:1231] [22850] train/loss = 5.3940218687057495
+I1129 14:07:06.668596 137274321021824 utils.py:1231] [22850] l2_grads = 0.9669804573059082
+I1129 14:07:06.668702 137274321021824 utils.py:1231] [22850] lr = 0.0009618012341891903
+I1129 14:07:06.668784 137274321021824 utils.py:1231] [22850] uptime = 145016.031142415
+I1129 14:07:06.668853 137274321021824 utils.py:1231] [22850] examples_seen = 23398400.0
+I1129 14:07:06.668926 137274321021824 utils.py:1231] [22850] progress = 0.20292532170546076
+I1129 14:07:06.669012 137274321021824 utils.py:1231] [22850] epoch = 18.263348962313266
+I1129 14:07:06.669078 137274321021824 utils.py:1231] [22850] img/sec/core = 164.17936708787875
+I1129 14:07:06.669149 137274321021824 utils.py:1231] [22850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.24791327053638
+I1129 14:07:06.669215 137274321021824 utils.py:1231] [22850] core_hours = 40.24791327053638
+I1129 14:07:06.669292 137274321021824 train.py:125] NOTE: Steps:22850/112603 [20.3%]
+Walltime:1d16h16m (0s eval)
+ETA:6d14h6m
+Total train time:8d6h21m
+I1129 14:12:18.499949 137274321021824 utils.py:1231] [22900] l2_params = 329.78488831820687
+I1129 14:12:18.500157 137274321021824 utils.py:1231] [22900] train/loss = 3.1172974705696106
+I1129 14:12:18.500272 137274321021824 utils.py:1231] [22900] l2_grads = 1.2357784509658813
+I1129 14:12:18.500344 137274321021824 utils.py:1231] [22900] lr = 0.00096150724782667
+I1129 14:12:18.500406 137274321021824 utils.py:1231] [22900] uptime = 145327.862767027
+I1129 14:12:18.500468 137274321021824 utils.py:1231] [22900] examples_seen = 23449600.0
+I1129 14:12:18.500538 137274321021824 utils.py:1231] [22900] progress = 0.2033693596085362
+I1129 14:12:18.500597 137274321021824 utils.py:1231] [22900] epoch = 18.30331252678222
+I1129 14:12:18.500655 137274321021824 utils.py:1231] [22900] img/sec/core = 164.19117228313448
+I1129 14:12:18.500720 137274321021824 utils.py:1231] [22900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.33453316626194
+I1129 14:12:18.500777 137274321021824 utils.py:1231] [22900] core_hours = 40.33453316626194
+I1129 14:12:18.500850 137274321021824 train.py:125] NOTE: Steps:22900/112603 [20.3%]
+Walltime:1d16h22m (0s eval)
+ETA:6d14h0m
+Total train time:8d6h20m
+I1129 14:17:30.380308 137274321021824 utils.py:1231] [22950] l2_params = 329.83980438017505
+I1129 14:17:30.380566 137274321021824 utils.py:1231] [22950] train/loss = 3.7692970037460327
+I1129 14:17:30.380700 137274321021824 utils.py:1231] [22950] l2_grads = 1.112181305885315
+I1129 14:17:30.380783 137274321021824 utils.py:1231] [22950] lr = 0.0009612121797859512
+I1129 14:17:30.380838 137274321021824 utils.py:1231] [22950] uptime = 145639.74319973303
+I1129 14:17:30.380902 137274321021824 utils.py:1231] [22950] examples_seen = 23500800.0
+I1129 14:17:30.380954 137274321021824 utils.py:1231] [22950] progress = 0.2038133975116116
+I1129 14:17:30.381002 137274321021824 utils.py:1231] [22950] epoch = 18.34327609125118
+I1129 14:17:30.381056 137274321021824 utils.py:1231] [22950] img/sec/core = 164.1654769931131
+I1129 14:17:30.381114 137274321021824 utils.py:1231] [22950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.421166619791386
+I1129 14:17:30.381164 137274321021824 utils.py:1231] [22950] core_hours = 40.421166619791386
+I1129 14:17:30.381227 137274321021824 train.py:125] NOTE: Steps:22950/112603 [20.4%]
+Walltime:1d16h27m (0s eval)
+ETA:6d13h55m
+Total train time:8d6h20m
+I1129 14:22:42.259449 137274321021824 utils.py:1231] [23000] l2_params = 329.87460106214456
+I1129 14:22:42.259758 137274321021824 utils.py:1231] [23000] train/loss = 4.263831079006195
+I1129 14:22:42.259927 137274321021824 utils.py:1231] [23000] l2_grads = 1.1432368755340576
+I1129 14:22:42.259991 137274321021824 utils.py:1231] [23000] lr = 0.0009609160307586124
+I1129 14:22:42.260043 137274321021824 utils.py:1231] [23000] uptime = 145951.62240467098
+I1129 14:22:42.260095 137274321021824 utils.py:1231] [23000] examples_seen = 23552000.0
+I1129 14:22:42.260144 137274321021824 utils.py:1231] [23000] progress = 0.204257435414687
+I1129 14:22:42.260193 137274321021824 utils.py:1231] [23000] epoch = 18.383239655720136
+I1129 14:22:42.260244 137274321021824 utils.py:1231] [23000] img/sec/core = 164.1661232597436
+I1129 14:22:42.260300 137274321021824 utils.py:1231] [23000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.50779973227416
+I1129 14:22:42.260355 137274321021824 utils.py:1231] [23000] core_hours = 40.50779973227416
+I1129 14:22:42.260416 137274321021824 train.py:125] NOTE: Steps:23000/112603 [20.4%]
+Walltime:1d16h32m (0s eval)
+ETA:6d13h49m
+Total train time:8d6h20m
+I1129 14:27:54.473349 137274321021824 utils.py:1231] [23050] l2_params = 329.9275897234582
+I1129 14:27:54.473566 137274321021824 utils.py:1231] [23050] train/loss = 4.593002021312714
+I1129 14:27:54.473679 137274321021824 utils.py:1231] [23050] l2_grads = 1.0379414558410645
+I1129 14:27:54.473753 137274321021824 utils.py:1231] [23050] lr = 0.0009606188014387664
+I1129 14:27:54.473815 137274321021824 utils.py:1231] [23050] uptime = 146263.836175951
+I1129 14:27:54.473944 137274321021824 utils.py:1231] [23050] examples_seen = 23603200.0
+I1129 14:27:54.474026 137274321021824 utils.py:1231] [23050] progress = 0.2047014733177624
+I1129 14:27:54.474099 137274321021824 utils.py:1231] [23050] epoch = 18.423203220189095
+I1129 14:27:54.474192 137274321021824 utils.py:1231] [23050] img/sec/core = 163.99020385964923
+I1129 14:27:54.474262 137274321021824 utils.py:1231] [23050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.59452577985194
+I1129 14:27:54.474321 137274321021824 utils.py:1231] [23050] core_hours = 40.59452577985194
+I1129 14:27:54.474391 137274321021824 train.py:125] NOTE: Steps:23050/112603 [20.5%]
+Walltime:1d16h37m (0s eval)
+ETA:6d13h43m
+Total train time:8d6h19m
+I1129 14:33:06.305380 137274321021824 utils.py:1231] [23100] l2_params = 329.9385791249846
+I1129 14:33:06.305669 137274321021824 utils.py:1231] [23100] train/loss = 3.4746838808059692
+I1129 14:33:06.305845 137274321021824 utils.py:1231] [23100] l2_grads = 1.280544400215149
+I1129 14:33:06.305938 137274321021824 utils.py:1231] [23100] lr = 0.0009603204925230576
+I1129 14:33:06.306004 137274321021824 utils.py:1231] [23100] uptime = 146575.66836518602
+I1129 14:33:06.306065 137274321021824 utils.py:1231] [23100] examples_seen = 23654400.0
+I1129 14:33:06.306124 137274321021824 utils.py:1231] [23100] progress = 0.20514551122083782
+I1129 14:33:06.306181 137274321021824 utils.py:1231] [23100] epoch = 18.46316678465805
+I1129 14:33:06.306240 137274321021824 utils.py:1231] [23100] img/sec/core = 164.19087498826624
+I1129 14:33:06.306303 137274321021824 utils.py:1231] [23100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.681145832417215
+I1129 14:33:06.306359 137274321021824 utils.py:1231] [23100] core_hours = 40.681145832417215
+I1129 14:33:06.306425 137274321021824 train.py:125] NOTE: Steps:23100/112603 [20.5%]
+Walltime:1d16h42m (0s eval)
+ETA:6d13h38m
+Total train time:8d6h19m
+I1129 14:38:18.104261 137274321021824 utils.py:1231] [23150] l2_params = 329.95843757765317
+I1129 14:38:18.104493 137274321021824 utils.py:1231] [23150] train/loss = 4.37756609916687
+I1129 14:38:18.104621 137274321021824 utils.py:1231] [23150] l2_grads = 1.1917158365249634
+I1129 14:38:18.104724 137274321021824 utils.py:1231] [23150] lr = 0.000960021104710661
+I1129 14:38:18.104792 137274321021824 utils.py:1231] [23150] uptime = 146887.46715406002
+I1129 14:38:18.104856 137274321021824 utils.py:1231] [23150] examples_seen = 23705600.0
+I1129 14:38:18.104922 137274321021824 utils.py:1231] [23150] progress = 0.20558954912391322
+I1129 14:38:18.104982 137274321021824 utils.py:1231] [23150] epoch = 18.503130349127005
+I1129 14:38:18.105039 137274321021824 utils.py:1231] [23150] img/sec/core = 164.2084633647818
+I1129 14:38:18.105099 137274321021824 utils.py:1231] [23150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.76775660710444
+I1129 14:38:18.105172 137274321021824 utils.py:1231] [23150] core_hours = 40.76775660710444
+I1129 14:38:18.105273 137274321021824 train.py:125] NOTE: Steps:23150/112603 [20.6%]
+Walltime:1d16h48m (0s eval)
+ETA:6d13h32m
+Total train time:8d6h18m
+I1129 14:43:29.971602 137274321021824 utils.py:1231] [23200] l2_params = 330.0016300759912
+I1129 14:43:29.971858 137274321021824 utils.py:1231] [23200] train/loss = 3.6522574424743652
+I1129 14:43:29.971981 137274321021824 utils.py:1231] [23200] l2_grads = 1.098816990852356
+I1129 14:43:29.972060 137274321021824 utils.py:1231] [23200] lr = 0.0009597206387032788
+I1129 14:43:29.972125 137274321021824 utils.py:1231] [23200] uptime = 147199.33448701398
+I1129 14:43:29.972185 137274321021824 utils.py:1231] [23200] examples_seen = 23756800.0
+I1129 14:43:29.972241 137274321021824 utils.py:1231] [23200] progress = 0.20603358702698862
+I1129 14:43:29.972294 137274321021824 utils.py:1231] [23200] epoch = 18.543093913595964
+I1129 14:43:29.972346 137274321021824 utils.py:1231] [23200] img/sec/core = 164.17237264011158
+I1129 14:43:29.972410 137274321021824 utils.py:1231] [23200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.85438642181388
+I1129 14:43:29.972464 137274321021824 utils.py:1231] [23200] core_hours = 40.85438642181388
+I1129 14:43:29.972537 137274321021824 train.py:125] NOTE: Steps:23200/112603 [20.6%]
+Walltime:1d16h53m (0s eval)
+ETA:6d13h26m
+Total train time:8d6h18m
+I1129 14:48:41.852563 137274321021824 utils.py:1231] [23250] l2_params = 330.1074723478546
+I1129 14:48:41.852783 137274321021824 utils.py:1231] [23250] train/loss = 3.1172560453414917
+I1129 14:48:41.852906 137274321021824 utils.py:1231] [23250] l2_grads = 1.3245468139648438
+I1129 14:48:41.852982 137274321021824 utils.py:1231] [23250] lr = 0.0009594190952051431
+I1129 14:48:41.853037 137274321021824 utils.py:1231] [23250] uptime = 147511.21539745003
+I1129 14:48:41.853093 137274321021824 utils.py:1231] [23250] examples_seen = 23808000.0
+I1129 14:48:41.853153 137274321021824 utils.py:1231] [23250] progress = 0.20647762493006402
+I1129 14:48:41.853206 137274321021824 utils.py:1231] [23250] epoch = 18.58305747806492
+I1129 14:48:41.853257 137274321021824 utils.py:1231] [23250] img/sec/core = 164.1652255292639
+I1129 14:48:41.853321 137274321021824 utils.py:1231] [23250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 40.9410200080461
+I1129 14:48:41.853375 137274321021824 utils.py:1231] [23250] core_hours = 40.9410200080461
+I1129 14:48:41.853444 137274321021824 train.py:125] NOTE: Steps:23250/112603 [20.6%]
+Walltime:1d16h58m (0s eval)
+ETA:6d13h21m
+Total train time:8d6h18m
+I1129 14:53:53.736351 137274321021824 utils.py:1231] [23300] l2_params = 330.1680521106228
+I1129 14:53:53.736659 137274321021824 utils.py:1231] [23300] train/loss = 3.100735455751419
+I1129 14:53:53.736825 137274321021824 utils.py:1231] [23300] l2_grads = 1.344480037689209
+I1129 14:53:53.736906 137274321021824 utils.py:1231] [23300] lr = 0.0009591164749230098
+I1129 14:53:53.736961 137274321021824 utils.py:1231] [23300] uptime = 147823.09932270402
+I1129 14:53:53.737013 137274321021824 utils.py:1231] [23300] examples_seen = 23859200.0
+I1129 14:53:53.737061 137274321021824 utils.py:1231] [23300] progress = 0.20692166283313942
+I1129 14:53:53.737109 137274321021824 utils.py:1231] [23300] epoch = 18.62302104253388
+I1129 14:53:53.737165 137274321021824 utils.py:1231] [23300] img/sec/core = 164.16363863031384
+I1129 14:53:53.737234 137274321021824 utils.py:1231] [23300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.027654431727775
+I1129 14:53:53.737293 137274321021824 utils.py:1231] [23300] core_hours = 41.027654431727775
+I1129 14:53:53.737364 137274321021824 train.py:125] NOTE: Steps:23300/112603 [20.7%]
+Walltime:1d17h3m (0s eval)
+ETA:6d13h15m
+Total train time:8d6h17m
+I1129 14:59:05.621179 137274321021824 utils.py:1231] [23350] l2_params = 330.248697168941
+I1129 14:59:05.621432 137274321021824 utils.py:1231] [23350] train/loss = 3.696235477924347
+I1129 14:59:05.621538 137274321021824 utils.py:1231] [23350] l2_grads = 1.261072039604187
+I1129 14:59:05.621618 137274321021824 utils.py:1231] [23350] lr = 0.000958812778566158
+I1129 14:59:05.621679 137274321021824 utils.py:1231] [23350] uptime = 148134.984041097
+I1129 14:59:05.621740 137274321021824 utils.py:1231] [23350] examples_seen = 23910400.0
+I1129 14:59:05.621796 137274321021824 utils.py:1231] [23350] progress = 0.20736570073621485
+I1129 14:59:05.621856 137274321021824 utils.py:1231] [23350] epoch = 18.662984607002834
+I1129 14:59:05.621922 137274321021824 utils.py:1231] [23350] img/sec/core = 164.16322115366702
+I1129 14:59:05.621986 137274321021824 utils.py:1231] [23350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.11428907572583
+I1129 14:59:05.622053 137274321021824 utils.py:1231] [23350] core_hours = 41.11428907572583
+I1129 14:59:05.622125 137274321021824 train.py:125] NOTE: Steps:23350/112603 [20.7%]
+Walltime:1d17h8m (0s eval)
+ETA:6d13h10m
+Total train time:8d6h17m
+I1129 15:04:17.490170 137274321021824 utils.py:1231] [23400] l2_params = 330.2994092938529
+I1129 15:04:17.490467 137274321021824 utils.py:1231] [23400] train/loss = 3.0194165110588074
+I1129 15:04:17.490649 137274321021824 utils.py:1231] [23400] l2_grads = 1.1934700012207031
+I1129 15:04:17.490742 137274321021824 utils.py:1231] [23400] lr = 0.0009585080068463899
+I1129 15:04:17.490825 137274321021824 utils.py:1231] [23400] uptime = 148446.85318632203
+I1129 15:04:17.490913 137274321021824 utils.py:1231] [23400] examples_seen = 23961600.0
+I1129 15:04:17.490993 137274321021824 utils.py:1231] [23400] progress = 0.20780973863929025
+I1129 15:04:17.491066 137274321021824 utils.py:1231] [23400] epoch = 18.702948171471792
+I1129 15:04:17.491149 137274321021824 utils.py:1231] [23400] img/sec/core = 164.17141863476255
+I1129 15:04:17.491244 137274321021824 utils.py:1231] [23400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.20091939384388
+I1129 15:04:17.491316 137274321021824 utils.py:1231] [23400] core_hours = 41.20091939384388
+I1129 15:04:17.491403 137274321021824 train.py:125] NOTE: Steps:23400/112603 [20.8%]
+Walltime:1d17h14m (0s eval)
+ETA:6d13h4m
+Total train time:8d6h16m
+I1129 15:09:29.363151 137274321021824 utils.py:1231] [23450] l2_params = 330.2980302060245
+I1129 15:09:29.363370 137274321021824 utils.py:1231] [23450] train/loss = 3.3913366496562958
+I1129 15:09:29.363467 137274321021824 utils.py:1231] [23450] l2_grads = 1.2386361360549927
+I1129 15:09:29.363530 137274321021824 utils.py:1231] [23450] lr = 0.0009582021604780281
+I1129 15:09:29.363586 137274321021824 utils.py:1231] [23450] uptime = 148758.725948425
+I1129 15:09:29.363642 137274321021824 utils.py:1231] [23450] examples_seen = 24012800.0
+I1129 15:09:29.363695 137274321021824 utils.py:1231] [23450] progress = 0.20825377654236565
+I1129 15:09:29.363747 137274321021824 utils.py:1231] [23450] epoch = 18.742911735940748
+I1129 15:09:29.363800 137274321021824 utils.py:1231] [23450] img/sec/core = 164.16951469167844
+I1129 15:09:29.363862 137274321021824 utils.py:1231] [23450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.287550716650266
+I1129 15:09:29.363921 137274321021824 utils.py:1231] [23450] core_hours = 41.287550716650266
+I1129 15:09:29.363988 137274321021824 train.py:125] NOTE: Steps:23450/112603 [20.8%]
+Walltime:1d17h19m (0s eval)
+ETA:6d12h58m
+Total train time:8d6h16m
+I1129 15:14:41.246075 137274321021824 utils.py:1231] [23500] l2_params = 330.3487544279273
+I1129 15:14:41.246360 137274321021824 utils.py:1231] [23500] train/loss = 4.716654121875763
+I1129 15:14:41.246542 137274321021824 utils.py:1231] [23500] l2_grads = 0.9935318231582642
+I1129 15:14:41.246613 137274321021824 utils.py:1231] [23500] lr = 0.0009578952401779134
+I1129 15:14:41.246675 137274321021824 utils.py:1231] [23500] uptime = 149070.609036871
+I1129 15:14:41.246728 137274321021824 utils.py:1231] [23500] examples_seen = 24064000.0
+I1129 15:14:41.246777 137274321021824 utils.py:1231] [23500] progress = 0.20869781444544105
+I1129 15:14:41.246825 137274321021824 utils.py:1231] [23500] epoch = 18.782875300409703
+I1129 15:14:41.246875 137274321021824 utils.py:1231] [23500] img/sec/core = 164.1640790948555
+I1129 15:14:41.246955 137274321021824 utils.py:1231] [23500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.37418490788527
+I1129 15:14:41.247006 137274321021824 utils.py:1231] [23500] core_hours = 41.37418490788527
+I1129 15:14:41.247067 137274321021824 train.py:125] NOTE: Steps:23500/112603 [20.9%]
+Walltime:1d17h24m (0s eval)
+ETA:6d12h53m
+Total train time:8d6h15m
+I1129 15:19:53.135940 137274321021824 utils.py:1231] [23550] l2_params = 330.36037616797245
+I1129 15:19:53.136238 137274321021824 utils.py:1231] [23550] train/loss = 2.9064529836177826
+I1129 15:19:53.136475 137274321021824 utils.py:1231] [23550] l2_grads = 1.230272889137268
+I1129 15:19:53.136564 137274321021824 utils.py:1231] [23550] lr = 0.0009575872466654046
+I1129 15:19:53.136626 137274321021824 utils.py:1231] [23550] uptime = 149382.49898748
+I1129 15:19:53.136687 137274321021824 utils.py:1231] [23550] examples_seen = 24115200.0
+I1129 15:19:53.136748 137274321021824 utils.py:1231] [23550] progress = 0.20914185234851648
+I1129 15:19:53.136803 137274321021824 utils.py:1231] [23550] epoch = 18.822838864878662
+I1129 15:19:53.136858 137274321021824 utils.py:1231] [23550] img/sec/core = 164.16046717770013
+I1129 15:19:53.136926 137274321021824 utils.py:1231] [23550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.46082100527666
+I1129 15:19:53.136991 137274321021824 utils.py:1231] [23550] core_hours = 41.46082100527666
+I1129 15:19:53.137059 137274321021824 train.py:125] NOTE: Steps:23550/112603 [20.9%]
+Walltime:1d17h29m (0s eval)
+ETA:6d12h47m
+Total train time:8d6h15m
+I1129 15:25:05.026386 137274321021824 utils.py:1231] [23600] l2_params = 330.4119035653314
+I1129 15:25:05.026746 137274321021824 utils.py:1231] [23600] train/loss = 4.7958919405937195
+I1129 15:25:05.026935 137274321021824 utils.py:1231] [23600] l2_grads = 0.9594590663909912
+I1129 15:25:05.027016 137274321021824 utils.py:1231] [23600] lr = 0.0009572781806623737
+I1129 15:25:05.027090 137274321021824 utils.py:1231] [23600] uptime = 149694.389442068
+I1129 15:25:05.027159 137274321021824 utils.py:1231] [23600] examples_seen = 24166400.0
+I1129 15:25:05.027231 137274321021824 utils.py:1231] [23600] progress = 0.20958589025159188
+I1129 15:25:05.027299 137274321021824 utils.py:1231] [23600] epoch = 18.862802429347617
+I1129 15:25:05.027362 137274321021824 utils.py:1231] [23600] img/sec/core = 164.16020191329835
+I1129 15:25:05.027434 137274321021824 utils.py:1231] [23600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.54745724266222
+I1129 15:25:05.027493 137274321021824 utils.py:1231] [23600] core_hours = 41.54745724266222
+I1129 15:25:05.027575 137274321021824 train.py:125] NOTE: Steps:23600/112603 [21.0%]
+Walltime:1d17h34m (0s eval)
+ETA:6d12h42m
+Total train time:8d6h15m
+I1129 15:30:16.900851 137274321021824 utils.py:1231] [23650] l2_params = 330.457041403587
+I1129 15:30:16.901077 137274321021824 utils.py:1231] [23650] train/loss = 3.1302425265312195
+I1129 15:30:16.901181 137274321021824 utils.py:1231] [23650] l2_grads = 1.300582766532898
+I1129 15:30:16.901260 137274321021824 utils.py:1231] [23650] lr = 0.0009569680428932088
+I1129 15:30:16.901317 137274321021824 utils.py:1231] [23650] uptime = 150006.26367905602
+I1129 15:30:16.901386 137274321021824 utils.py:1231] [23650] examples_seen = 24217600.0
+I1129 15:30:16.901441 137274321021824 utils.py:1231] [23650] progress = 0.21002992815466728
+I1129 15:30:16.901510 137274321021824 utils.py:1231] [23650] epoch = 18.902765993816576
+I1129 15:30:16.901598 137274321021824 utils.py:1231] [23650] img/sec/core = 164.16873831733494
+I1129 15:30:16.901666 137274321021824 utils.py:1231] [23650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.63408897515888
+I1129 15:30:16.901755 137274321021824 utils.py:1231] [23650] core_hours = 41.63408897515888
+I1129 15:30:16.901825 137274321021824 train.py:125] NOTE: Steps:23650/112603 [21.0%]
+Walltime:1d17h40m (0s eval)
+ETA:6d12h36m
+Total train time:8d6h14m
+I1129 15:35:28.770959 137274321021824 utils.py:1231] [23700] l2_params = 330.4830302546161
+I1129 15:35:28.771250 137274321021824 utils.py:1231] [23700] train/loss = 3.098418951034546
+I1129 15:35:28.771429 137274321021824 utils.py:1231] [23700] l2_grads = 1.2943758964538574
+I1129 15:35:28.771523 137274321021824 utils.py:1231] [23700] lr = 0.0009566568340848092
+I1129 15:35:28.771598 137274321021824 utils.py:1231] [23700] uptime = 150318.13395840698
+I1129 15:35:28.771671 137274321021824 utils.py:1231] [23700] examples_seen = 24268800.0
+I1129 15:35:28.771735 137274321021824 utils.py:1231] [23700] progress = 0.21047396605774268
+I1129 15:35:28.771800 137274321021824 utils.py:1231] [23700] epoch = 18.94272955828553
+I1129 15:35:28.771859 137274321021824 utils.py:1231] [23700] img/sec/core = 164.17082162028692
+I1129 15:35:28.771939 137274321021824 utils.py:1231] [23700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.72071960831194
+I1129 15:35:28.771997 137274321021824 utils.py:1231] [23700] core_hours = 41.72071960831194
+I1129 15:35:28.772063 137274321021824 train.py:125] NOTE: Steps:23700/112603 [21.0%]
+Walltime:1d17h45m (0s eval)
+ETA:6d12h30m
+Total train time:8d6h14m
+I1129 15:40:40.649318 137274321021824 utils.py:1231] [23750] l2_params = 330.5363873632242
+I1129 15:40:40.649537 137274321021824 utils.py:1231] [23750] train/loss = 2.984461188316345
+I1129 15:40:40.649647 137274321021824 utils.py:1231] [23750] l2_grads = 1.3250240087509155
+I1129 15:40:40.649709 137274321021824 utils.py:1231] [23750] lr = 0.0009563445549665844
+I1129 15:40:40.649761 137274321021824 utils.py:1231] [23750] uptime = 150630.01212325203
+I1129 15:40:40.649815 137274321021824 utils.py:1231] [23750] examples_seen = 24320000.0
+I1129 15:40:40.649868 137274321021824 utils.py:1231] [23750] progress = 0.21091800396081808
+I1129 15:40:40.649923 137274321021824 utils.py:1231] [23750] epoch = 18.98269312275449
+I1129 15:40:40.649976 137274321021824 utils.py:1231] [23750] img/sec/core = 164.16667074285849
+I1129 15:40:40.650040 137274321021824 utils.py:1231] [23750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.80735243188
+I1129 15:40:40.650092 137274321021824 utils.py:1231] [23750] core_hours = 41.80735243188
+I1129 15:40:40.650163 137274321021824 train.py:125] NOTE: Steps:23750/112603 [21.1%]
+Walltime:1d17h50m (0s eval)
+ETA:6d12h25m
+Total train time:8d6h13m
+I1129 15:45:57.622714 137274321021824 utils.py:1231] [23800] l2_params = 330.5443997630572
+I1129 15:45:57.622982 137274321021824 utils.py:1231] [23800] train/loss = 3.1156474947929382
+I1129 15:45:57.623106 137274321021824 utils.py:1231] [23800] l2_grads = 1.3423781394958496
+I1129 15:45:57.623181 137274321021824 utils.py:1231] [23800] lr = 0.0009560312062704532
+I1129 15:45:57.623269 137274321021824 utils.py:1231] [23800] uptime = 150946.985624257
+I1129 15:45:57.623355 137274321021824 utils.py:1231] [23800] examples_seen = 24371200.0
+I1129 15:45:57.623416 137274321021824 utils.py:1231] [23800] progress = 0.2113620418638935
+I1129 15:45:57.623476 137274321021824 utils.py:1231] [23800] epoch = 19.022656687223446
+I1129 15:45:57.623537 137274321021824 utils.py:1231] [23800] img/sec/core = 161.52769817561517
+I1129 15:45:57.623610 137274321021824 utils.py:1231] [23800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.8954006266036
+I1129 15:45:57.623678 137274321021824 utils.py:1231] [23800] core_hours = 41.8954006266036
+I1129 15:45:57.623746 137274321021824 train.py:125] NOTE: Steps:23800/112603 [21.1%]
+Walltime:1d17h55m (0s eval)
+ETA:6d12h20m
+Total train time:8d6h13m
+I1129 15:51:09.503017 137274321021824 utils.py:1231] [23850] l2_params = 330.54264890917017
+I1129 15:51:09.503281 137274321021824 utils.py:1231] [23850] train/loss = 3.1077904403209686
+I1129 15:51:09.503422 137274321021824 utils.py:1231] [23850] l2_grads = 1.2043360471725464
+I1129 15:51:09.503506 137274321021824 utils.py:1231] [23850] lr = 0.0009557167887308391
+I1129 15:51:09.503571 137274321021824 utils.py:1231] [23850] uptime = 151258.865929401
+I1129 15:51:09.503629 137274321021824 utils.py:1231] [23850] examples_seen = 24422400.0
+I1129 15:51:09.503681 137274321021824 utils.py:1231] [23850] progress = 0.2118060797669689
+I1129 15:51:09.503735 137274321021824 utils.py:1231] [23850] epoch = 19.0626202516924
+I1129 15:51:09.503791 137274321021824 utils.py:1231] [23850] img/sec/core = 164.16554413834956
+I1129 15:51:09.503855 137274321021824 utils.py:1231] [23850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 41.98203404469916
+I1129 15:51:09.503925 137274321021824 utils.py:1231] [23850] core_hours = 41.98203404469916
+I1129 15:51:09.504011 137274321021824 train.py:125] NOTE: Steps:23850/112603 [21.2%]
+Walltime:1d18h0m (0s eval)
+ETA:6d12h14m
+Total train time:8d6h13m
+I1129 15:56:21.416761 137274321021824 utils.py:1231] [23900] l2_params = 330.5760139215457
+I1129 15:56:21.417037 137274321021824 utils.py:1231] [23900] train/loss = 5.411723613739014
+I1129 15:56:21.417212 137274321021824 utils.py:1231] [23900] l2_grads = 1.1277661323547363
+I1129 15:56:21.417289 137274321021824 utils.py:1231] [23900] lr = 0.0009554013030846728
+I1129 15:56:21.417355 137274321021824 utils.py:1231] [23900] uptime = 151570.779716351
+I1129 15:56:21.417409 137274321021824 utils.py:1231] [23900] examples_seen = 24473600.0
+I1129 15:56:21.417459 137274321021824 utils.py:1231] [23900] progress = 0.2122501176700443
+I1129 15:56:21.417515 137274321021824 utils.py:1231] [23900] epoch = 19.10258381616136
+I1129 15:56:21.417587 137274321021824 utils.py:1231] [23900] img/sec/core = 164.14792209299756
+I1129 15:56:21.417650 137274321021824 utils.py:1231] [23900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.06867676329638
+I1129 15:56:21.417704 137274321021824 utils.py:1231] [23900] core_hours = 42.06867676329638
+I1129 15:56:21.417767 137274321021824 train.py:125] NOTE: Steps:23900/112603 [21.2%]
+Walltime:1d18h6m (0s eval)
+ETA:6d12h8m
+Total train time:8d6h13m
+I1129 16:01:37.555547 137274321021824 utils.py:1231] [23950] l2_params = 330.62954931734856
+I1129 16:01:37.555849 137274321021824 utils.py:1231] [23950] train/loss = 2.919912040233612
+I1129 16:01:37.556050 137274321021824 utils.py:1231] [23950] l2_grads = 1.3742376565933228
+I1129 16:01:37.556129 137274321021824 utils.py:1231] [23950] lr = 0.0009550847500713886
+I1129 16:01:37.556191 137274321021824 utils.py:1231] [23950] uptime = 151886.918552508
+I1129 16:01:37.556251 137274321021824 utils.py:1231] [23950] examples_seen = 24524800.0
+I1129 16:01:37.556309 137274321021824 utils.py:1231] [23950] progress = 0.21269415557311971
+I1129 16:01:37.556365 137274321021824 utils.py:1231] [23950] epoch = 19.142547380630315
+I1129 16:01:37.556429 137274321021824 utils.py:1231] [23950] img/sec/core = 161.95416109702833
+I1129 16:01:37.556497 137274321021824 utils.py:1231] [23950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.15649310667333
+I1129 16:01:37.556553 137274321021824 utils.py:1231] [23950] core_hours = 42.15649310667333
+I1129 16:01:37.556626 137274321021824 train.py:125] NOTE: Steps:23950/112603 [21.3%]
+Walltime:1d18h11m (0s eval)
+ETA:6d12h3m
+Total train time:8d6h13m
+I1129 16:07:07.544481 137274321021824 utils.py:1231] [24000] l2_params = 330.6891373605699
+I1129 16:07:07.544775 137274321021824 utils.py:1231] [24000] train/loss = 4.576291143894196
+I1129 16:07:07.544961 137274321021824 utils.py:1231] [24000] l2_grads = 1.0018764734268188
+I1129 16:07:07.545099 137274321021824 utils.py:1231] [24000] lr = 0.0009547671304329212
+I1129 16:07:07.545217 137274321021824 utils.py:1231] [24000] uptime = 152216.90757633
+I1129 16:07:07.545305 137274321021824 utils.py:1231] [24000] examples_seen = 24576000.0
+I1129 16:07:07.545373 137274321021824 utils.py:1231] [24000] progress = 0.21313819347619511
+I1129 16:07:07.545473 137274321021824 utils.py:1231] [24000] epoch = 19.182510945099274
+I1129 16:07:07.545559 137274321021824 utils.py:1231] [24000] img/sec/core = 155.15667584028728
+I1129 16:07:07.545640 137274321021824 utils.py:1231] [24000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.24815672440165
+I1129 16:07:07.545708 137274321021824 utils.py:1231] [24000] core_hours = 42.24815672440165
+I1129 16:07:07.545824 137274321021824 train.py:125] NOTE: Steps:24000/112603 [21.3%]
+Walltime:1d18h16m (0s eval)
+ETA:6d11h59m
+Total train time:8d6h14m
+I1129 16:12:37.307825 137274321021824 utils.py:1231] [24050] l2_params = 330.76580675444404
+I1129 16:12:37.308114 137274321021824 utils.py:1231] [24050] train/loss = 3.1766662895679474
+I1129 16:12:37.308282 137274321021824 utils.py:1231] [24050] l2_grads = 1.2469961643218994
+I1129 16:12:37.308366 137274321021824 utils.py:1231] [24050] lr = 0.0009544484449137056
+I1129 16:12:37.308429 137274321021824 utils.py:1231] [24050] uptime = 152546.67079027498
+I1129 16:12:37.308494 137274321021824 utils.py:1231] [24050] examples_seen = 24627200.0
+I1129 16:12:37.308556 137274321021824 utils.py:1231] [24050] progress = 0.21358223137927054
+I1129 16:12:37.308620 137274321021824 utils.py:1231] [24050] epoch = 19.22247450956823
+I1129 16:12:37.308681 137274321021824 utils.py:1231] [24050] img/sec/core = 155.26292149899868
+I1129 16:12:37.308761 137274321021824 utils.py:1231] [24050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.33975761716416
+I1129 16:12:37.308834 137274321021824 utils.py:1231] [24050] core_hours = 42.33975761716416
+I1129 16:12:37.308908 137274321021824 train.py:125] NOTE: Steps:24050/112603 [21.4%]
+Walltime:1d18h22m (0s eval)
+ETA:6d11h54m
+Total train time:8d6h15m
+I1129 16:17:58.534765 137274321021824 utils.py:1231] [24100] l2_params = 330.7751329409585
+I1129 16:17:58.535117 137274321021824 utils.py:1231] [24100] train/loss = 2.895594537258148
+I1129 16:17:58.535295 137274321021824 utils.py:1231] [24100] l2_grads = 1.4064724445343018
+I1129 16:17:58.535384 137274321021824 utils.py:1231] [24100] lr = 0.0009541286942606756
+I1129 16:17:58.535457 137274321021824 utils.py:1231] [24100] uptime = 152867.89781834703
+I1129 16:17:58.535532 137274321021824 utils.py:1231] [24100] examples_seen = 24678400.0
+I1129 16:17:58.535592 137274321021824 utils.py:1231] [24100] progress = 0.21402626928234594
+I1129 16:17:58.535671 137274321021824 utils.py:1231] [24100] epoch = 19.262438074037185
+I1129 16:17:58.535734 137274321021824 utils.py:1231] [24100] img/sec/core = 159.38882947458544
+I1129 16:17:58.535801 137274321021824 utils.py:1231] [24100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.42898734718416
+I1129 16:17:58.535873 137274321021824 utils.py:1231] [24100] core_hours = 42.42898734718416
+I1129 16:17:58.535961 137274321021824 train.py:125] NOTE: Steps:24100/112603 [21.4%]
+Walltime:1d18h27m (0s eval)
+ETA:6d11h49m
+Total train time:8d6h15m
+I1129 16:23:15.865863 137274321021824 utils.py:1231] [24150] l2_params = 330.83172662565636
+I1129 16:23:15.866132 137274321021824 utils.py:1231] [24150] train/loss = 2.9054976999759674
+I1129 16:23:15.866304 137274321021824 utils.py:1231] [24150] l2_grads = 1.3667678833007812
+I1129 16:23:15.866375 137274321021824 utils.py:1231] [24150] lr = 0.000953807879223261
+I1129 16:23:15.866438 137274321021824 utils.py:1231] [24150] uptime = 153185.22879902698
+I1129 16:23:15.866499 137274321021824 utils.py:1231] [24150] examples_seen = 24729600.0
+I1129 16:23:15.866559 137274321021824 utils.py:1231] [24150] progress = 0.21447030718542134
+I1129 16:23:15.866619 137274321021824 utils.py:1231] [24150] epoch = 19.302401638506144
+I1129 16:23:15.866675 137274321021824 utils.py:1231] [24150] img/sec/core = 161.34573400393498
+I1129 16:23:15.866739 137274321021824 utils.py:1231] [24150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.51713484181749
+I1129 16:23:15.866792 137274321021824 utils.py:1231] [24150] core_hours = 42.51713484181749
+I1129 16:23:15.866863 137274321021824 train.py:125] NOTE: Steps:24150/112603 [21.4%]
+Walltime:1d18h33m (0s eval)
+ETA:6d11h44m
+Total train time:8d6h15m
+I1129 16:28:27.744934 137274321021824 utils.py:1231] [24200] l2_params = 330.8477966592224
+I1129 16:28:27.745297 137274321021824 utils.py:1231] [24200] train/loss = 3.0586827993392944
+I1129 16:28:27.745538 137274321021824 utils.py:1231] [24200] l2_grads = 1.3930139541625977
+I1129 16:28:27.745686 137274321021824 utils.py:1231] [24200] lr = 0.000953486000553386
+I1129 16:28:27.745795 137274321021824 utils.py:1231] [24200] uptime = 153497.10814851598
+I1129 16:28:27.745877 137274321021824 utils.py:1231] [24200] examples_seen = 24780800.0
+I1129 16:28:27.745977 137274321021824 utils.py:1231] [24200] progress = 0.21491434508849674
+I1129 16:28:27.746062 137274321021824 utils.py:1231] [24200] epoch = 19.3423652029751
+I1129 16:28:27.746142 137274321021824 utils.py:1231] [24200] img/sec/core = 164.16604717141166
+I1129 16:28:27.746244 137274321021824 utils.py:1231] [24200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.60376799445332
+I1129 16:28:27.746332 137274321021824 utils.py:1231] [24200] core_hours = 42.60376799445332
+I1129 16:28:27.746422 137274321021824 train.py:125] NOTE: Steps:24200/112603 [21.5%]
+Walltime:1d18h38m (0s eval)
+ETA:6d11h38m
+Total train time:8d6h15m
+I1129 16:33:43.856824 137274321021824 utils.py:1231] [24250] l2_params = 330.85583980621783
+I1129 16:33:43.857068 137274321021824 utils.py:1231] [24250] train/loss = 5.387768626213074
+I1129 16:33:43.857164 137274321021824 utils.py:1231] [24250] l2_grads = 0.9482086896896362
+I1129 16:33:43.857223 137274321021824 utils.py:1231] [24250] lr = 0.0009531630590054684
+I1129 16:33:43.857281 137274321021824 utils.py:1231] [24250] uptime = 153813.21964330302
+I1129 16:33:43.857330 137274321021824 utils.py:1231] [24250] examples_seen = 24832000.0
+I1129 16:33:43.857376 137274321021824 utils.py:1231] [24250] progress = 0.21535838299157217
+I1129 16:33:43.857422 137274321021824 utils.py:1231] [24250] epoch = 19.382328767444058
+I1129 16:33:43.857470 137274321021824 utils.py:1231] [24250] img/sec/core = 161.96816896676876
+I1129 16:33:43.857524 137274321021824 utils.py:1231] [24250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.69157674300527
+I1129 16:33:43.857574 137274321021824 utils.py:1231] [24250] core_hours = 42.69157674300527
+I1129 16:33:43.857636 137274321021824 train.py:125] NOTE: Steps:24250/112603 [21.5%]
+Walltime:1d18h43m (0s eval)
+ETA:6d11h33m
+Total train time:8d6h15m
+I1129 16:38:59.725420 137274321021824 utils.py:1231] [24300] l2_params = 330.93218262176515
+I1129 16:38:59.725664 137274321021824 utils.py:1231] [24300] train/loss = 3.111878126859665
+I1129 16:38:59.725787 137274321021824 utils.py:1231] [24300] l2_grads = 1.3551028966903687
+I1129 16:38:59.725860 137274321021824 utils.py:1231] [24300] lr = 0.0009528390553364168
+I1129 16:38:59.725927 137274321021824 utils.py:1231] [24300] uptime = 154129.088289025
+I1129 16:38:59.725982 137274321021824 utils.py:1231] [24300] examples_seen = 24883200.0
+I1129 16:38:59.726033 137274321021824 utils.py:1231] [24300] progress = 0.21580242089464757
+I1129 16:38:59.726082 137274321021824 utils.py:1231] [24300] epoch = 19.422292331913013
+I1129 16:38:59.726135 137274321021824 utils.py:1231] [24300] img/sec/core = 162.09269483829576
+I1129 16:38:59.726194 137274321021824 utils.py:1231] [24300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.779318033483605
+I1129 16:38:59.726248 137274321021824 utils.py:1231] [24300] core_hours = 42.779318033483605
+I1129 16:38:59.726309 137274321021824 train.py:125] NOTE: Steps:24300/112603 [21.6%]
+Walltime:1d18h48m (0s eval)
+ETA:6d11h28m
+Total train time:8d6h15m
+I1129 16:44:11.615307 137274321021824 utils.py:1231] [24350] l2_params = 330.95652380541105
+I1129 16:44:11.615595 137274321021824 utils.py:1231] [24350] train/loss = 3.6558629274368286
+I1129 16:44:11.615767 137274321021824 utils.py:1231] [24350] l2_grads = 1.1238374710083008
+I1129 16:44:11.615841 137274321021824 utils.py:1231] [24350] lr = 0.0009525139903056294
+I1129 16:44:11.615906 137274321021824 utils.py:1231] [24350] uptime = 154440.978267522
+I1129 16:44:11.615962 137274321021824 utils.py:1231] [24350] examples_seen = 24934400.0
+I1129 16:44:11.616013 137274321021824 utils.py:1231] [24350] progress = 0.21624645879772297
+I1129 16:44:11.616064 137274321021824 utils.py:1231] [24350] epoch = 19.462255896381972
+I1129 16:44:11.616116 137274321021824 utils.py:1231] [24350] img/sec/core = 164.1604524990943
+I1129 16:44:11.616182 137274321021824 utils.py:1231] [24350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.86595413862165
+I1129 16:44:11.616239 137274321021824 utils.py:1231] [24350] core_hours = 42.86595413862165
+I1129 16:44:11.616302 137274321021824 train.py:125] NOTE: Steps:24350/112603 [21.6%]
+Walltime:1d18h54m (0s eval)
+ETA:6d11h22m
+Total train time:8d6h14m
+I1129 16:49:34.735078 137274321021824 utils.py:1231] [24400] l2_params = 330.99482067887743
+I1129 16:49:34.735299 137274321021824 utils.py:1231] [24400] train/loss = 4.421006798744202
+I1129 16:49:34.735406 137274321021824 utils.py:1231] [24400] l2_grads = 1.20015549659729
+I1129 16:49:34.735478 137274321021824 utils.py:1231] [24400] lr = 0.000952187864674992
+I1129 16:49:34.735543 137274321021824 utils.py:1231] [24400] uptime = 154764.097904424
+I1129 16:49:34.735609 137274321021824 utils.py:1231] [24400] examples_seen = 24985600.0
+I1129 16:49:34.735670 137274321021824 utils.py:1231] [24400] progress = 0.21669049670079837
+I1129 16:49:34.735731 137274321021824 utils.py:1231] [24400] epoch = 19.502219460850927
+I1129 16:49:34.735793 137274321021824 utils.py:1231] [24400] img/sec/core = 158.4552411945426
+I1129 16:49:34.735859 137274321021824 utils.py:1231] [24400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 42.95570959331666
+I1129 16:49:34.735930 137274321021824 utils.py:1231] [24400] core_hours = 42.95570959331666
+I1129 16:49:34.735998 137274321021824 train.py:125] NOTE: Steps:24400/112603 [21.7%]
+Walltime:1d18h59m (0s eval)
+ETA:6d11h17m
+Total train time:8d6h15m
+I1129 16:54:50.721081 137274321021824 utils.py:1231] [24450] l2_params = 331.02943454285304
+I1129 16:54:50.721298 137274321021824 utils.py:1231] [24450] train/loss = 3.0846266746520996
+I1129 16:54:50.721399 137274321021824 utils.py:1231] [24450] l2_grads = 1.394758701324463
+I1129 16:54:50.721465 137274321021824 utils.py:1231] [24450] lr = 0.0009518606792088751
+I1129 16:54:50.721519 137274321021824 utils.py:1231] [24450] uptime = 155080.08388043003
+I1129 16:54:50.721573 137274321021824 utils.py:1231] [24450] examples_seen = 25036800.0
+I1129 16:54:50.721624 137274321021824 utils.py:1231] [24450] progress = 0.21713453460387377
+I1129 16:54:50.721674 137274321021824 utils.py:1231] [24450] epoch = 19.542183025319883
+I1129 16:54:50.721728 137274321021824 utils.py:1231] [24450] img/sec/core = 162.03250741426714
+I1129 16:54:50.721786 137274321021824 utils.py:1231] [24450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.043483475540555
+I1129 16:54:50.721839 137274321021824 utils.py:1231] [24450] core_hours = 43.043483475540555
+I1129 16:54:50.721909 137274321021824 train.py:125] NOTE: Steps:24450/112603 [21.7%]
+Walltime:1d19h4m (0s eval)
+ETA:6d11h12m
+Total train time:8d6h15m
+I1129 17:00:02.549617 137274321021824 utils.py:1231] [24500] l2_params = 331.0493251301877
+I1129 17:00:02.549866 137274321021824 utils.py:1231] [24500] train/loss = 2.9376578330993652
+I1129 17:00:02.549967 137274321021824 utils.py:1231] [24500] l2_grads = 1.4513424634933472
+I1129 17:00:02.550029 137274321021824 utils.py:1231] [24500] lr = 0.0009515324346741358
+I1129 17:00:02.550082 137274321021824 utils.py:1231] [24500] uptime = 155391.912444268
+I1129 17:00:02.550136 137274321021824 utils.py:1231] [24500] examples_seen = 25088000.0
+I1129 17:00:02.550187 137274321021824 utils.py:1231] [24500] progress = 0.2175785725069492
+I1129 17:00:02.550235 137274321021824 utils.py:1231] [24500] epoch = 19.58214658978884
+I1129 17:00:02.550288 137274321021824 utils.py:1231] [24500] img/sec/core = 164.19278391251987
+I1129 17:00:02.550347 137274321021824 utils.py:1231] [24500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.1301025210511
+I1129 17:00:02.550400 137274321021824 utils.py:1231] [24500] core_hours = 43.1301025210511
+I1129 17:00:02.550461 137274321021824 train.py:125] NOTE: Steps:24500/112603 [21.8%]
+Walltime:1d19h9m (0s eval)
+ETA:6d11h6m
+Total train time:8d6h14m
+I1129 17:05:14.398814 137274321021824 utils.py:1231] [24550] l2_params = 331.04322881748146
+I1129 17:05:14.399098 137274321021824 utils.py:1231] [24550] train/loss = 3.0132977664470673
+I1129 17:05:14.399200 137274321021824 utils.py:1231] [24550] l2_grads = 1.3230255842208862
+I1129 17:05:14.399265 137274321021824 utils.py:1231] [24550] lr = 0.0009512031318401106
+I1129 17:05:14.399317 137274321021824 utils.py:1231] [24550] uptime = 155703.761679638
+I1129 17:05:14.399372 137274321021824 utils.py:1231] [24550] examples_seen = 25139200.0
+I1129 17:05:14.399422 137274321021824 utils.py:1231] [24550] progress = 0.2180226104100246
+I1129 17:05:14.399472 137274321021824 utils.py:1231] [24550] epoch = 19.622110154257797
+I1129 17:05:14.399523 137274321021824 utils.py:1231] [24550] img/sec/core = 164.1819000750416
+I1129 17:05:14.399583 137274321021824 utils.py:1231] [24550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.21672730865389
+I1129 17:05:14.399634 137274321021824 utils.py:1231] [24550] core_hours = 43.21672730865389
+I1129 17:05:14.399697 137274321021824 train.py:125] NOTE: Steps:24550/112603 [21.8%]
+Walltime:1d19h15m (0s eval)
+ETA:6d11h1m
+Total train time:8d6h14m
+I1129 17:10:26.281306 137274321021824 utils.py:1231] [24600] l2_params = 331.0923182850334
+I1129 17:10:26.281564 137274321021824 utils.py:1231] [24600] train/loss = 3.096827208995819
+I1129 17:10:26.281728 137274321021824 utils.py:1231] [24600] l2_grads = 1.3081220388412476
+I1129 17:10:26.281839 137274321021824 utils.py:1231] [24600] lr = 0.000950872771478618
+I1129 17:10:26.281953 137274321021824 utils.py:1231] [24600] uptime = 156015.64430371998
+I1129 17:10:26.282036 137274321021824 utils.py:1231] [24600] examples_seen = 25190400.0
+I1129 17:10:26.282103 137274321021824 utils.py:1231] [24600] progress = 0.2184666483131
+I1129 17:10:26.282180 137274321021824 utils.py:1231] [24600] epoch = 19.662073718726756
+I1129 17:10:26.282252 137274321021824 utils.py:1231] [24600] img/sec/core = 164.16432351980734
+I1129 17:10:26.282351 137274321021824 utils.py:1231] [24600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.30336137089888
+I1129 17:10:26.282422 137274321021824 utils.py:1231] [24600] core_hours = 43.30336137089888
+I1129 17:10:26.282516 137274321021824 train.py:125] NOTE: Steps:24600/112603 [21.8%]
+Walltime:1d19h20m (0s eval)
+ETA:6d10h55m
+Total train time:8d6h13m
+I1129 17:15:43.628568 137274321021824 utils.py:1231] [24650] l2_params = 331.11269967784085
+I1129 17:15:43.628870 137274321021824 utils.py:1231] [24650] train/loss = 2.905965745449066
+I1129 17:15:43.629027 137274321021824 utils.py:1231] [24650] l2_grads = 1.4472721815109253
+I1129 17:15:43.629093 137274321021824 utils.py:1231] [24650] lr = 0.0009505413543639546
+I1129 17:15:43.629152 137274321021824 utils.py:1231] [24650] uptime = 156332.991509852
+I1129 17:15:43.629221 137274321021824 utils.py:1231] [24650] examples_seen = 25241600.0
+I1129 17:15:43.629276 137274321021824 utils.py:1231] [24650] progress = 0.2189106862161754
+I1129 17:15:43.629333 137274321021824 utils.py:1231] [24650] epoch = 19.70203728319571
+I1129 17:15:43.629404 137274321021824 utils.py:1231] [24650] img/sec/core = 161.33748465615952
+I1129 17:15:43.629470 137274321021824 utils.py:1231] [24650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.391513372602226
+I1129 17:15:43.629524 137274321021824 utils.py:1231] [24650] core_hours = 43.391513372602226
+I1129 17:15:43.629590 137274321021824 train.py:125] NOTE: Steps:24650/112603 [21.9%]
+Walltime:1d19h25m (0s eval)
+ETA:6d10h50m
+Total train time:8d6h13m
+I1129 17:20:55.501527 137274321021824 utils.py:1231] [24700] l2_params = 331.1208599064144
+I1129 17:20:55.501816 137274321021824 utils.py:1231] [24700] train/loss = 4.59053909778595
+I1129 17:20:55.501999 137274321021824 utils.py:1231] [24700] l2_grads = 1.0541400909423828
+I1129 17:20:55.502081 137274321021824 utils.py:1231] [24700] lr = 0.0009502088812728944
+I1129 17:20:55.502143 137274321021824 utils.py:1231] [24700] uptime = 156644.86450406298
+I1129 17:20:55.502205 137274321021824 utils.py:1231] [24700] examples_seen = 25292800.0
+I1129 17:20:55.502263 137274321021824 utils.py:1231] [24700] progress = 0.21935472411925083
+I1129 17:20:55.502322 137274321021824 utils.py:1231] [24700] epoch = 19.742000847664666
+I1129 17:20:55.502388 137274321021824 utils.py:1231] [24700] img/sec/core = 164.1693925103492
+I1129 17:20:55.502459 137274321021824 utils.py:1231] [24700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.47814475988305
+I1129 17:20:55.502537 137274321021824 utils.py:1231] [24700] core_hours = 43.47814475988305
+I1129 17:20:55.502609 137274321021824 train.py:125] NOTE: Steps:24700/112603 [21.9%]
+Walltime:1d19h30m (0s eval)
+ETA:6d10h44m
+Total train time:8d6h13m
+I1129 17:26:10.730340 137274321021824 utils.py:1231] [24750] l2_params = 331.1537814252149
+I1129 17:26:10.730568 137274321021824 utils.py:1231] [24750] train/loss = 2.994306296110153
+I1129 17:26:10.730660 137274321021824 utils.py:1231] [24750] l2_grads = 1.412888765335083
+I1129 17:26:10.730719 137274321021824 utils.py:1231] [24750] lr = 0.0009498753529846866
+I1129 17:26:10.730776 137274321021824 utils.py:1231] [24750] uptime = 156960.0931392
+I1129 17:26:10.730845 137274321021824 utils.py:1231] [24750] examples_seen = 25344000.0
+I1129 17:26:10.730898 137274321021824 utils.py:1231] [24750] progress = 0.21979876202232623
+I1129 17:26:10.730946 137274321021824 utils.py:1231] [24750] epoch = 19.781964412133625
+I1129 17:26:10.730994 137274321021824 utils.py:1231] [24750] img/sec/core = 162.42179260697068
+I1129 17:26:10.731046 137274321021824 utils.py:1231] [24750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.56570826964332
+I1129 17:26:10.731102 137274321021824 utils.py:1231] [24750] core_hours = 43.56570826964332
+I1129 17:26:10.731160 137274321021824 train.py:125] NOTE: Steps:24750/112603 [22.0%]
+Walltime:1d19h36m (0s eval)
+ETA:6d10h39m
+Total train time:8d6h13m
+I1129 17:31:22.531612 137274321021824 utils.py:1231] [24800] l2_params = 331.17443735706684
+I1129 17:31:22.531845 137274321021824 utils.py:1231] [24800] train/loss = 2.967255026102066
+I1129 17:31:22.531961 137274321021824 utils.py:1231] [24800] l2_grads = 1.3478565216064453
+I1129 17:31:22.532027 137274321021824 utils.py:1231] [24800] lr = 0.0009495407702810517
+I1129 17:31:22.532079 137274321021824 utils.py:1231] [24800] uptime = 157271.894441013
+I1129 17:31:22.532133 137274321021824 utils.py:1231] [24800] examples_seen = 25395200.0
+I1129 17:31:22.532188 137274321021824 utils.py:1231] [24800] progress = 0.22024279992540163
+I1129 17:31:22.532238 137274321021824 utils.py:1231] [24800] epoch = 19.82192797660258
+I1129 17:31:22.532291 137274321021824 utils.py:1231] [24800] img/sec/core = 164.2071399390894
+I1129 17:31:22.532350 137274321021824 utils.py:1231] [24800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.652319742369166
+I1129 17:31:22.532402 137274321021824 utils.py:1231] [24800] core_hours = 43.652319742369166
+I1129 17:31:22.532467 137274321021824 train.py:125] NOTE: Steps:24800/112603 [22.0%]
+Walltime:1d19h41m (0s eval)
+ETA:6d10h33m
+Total train time:8d6h13m
+I1129 17:36:34.369178 137274321021824 utils.py:1231] [24850] l2_params = 331.1657227365194
+I1129 17:36:34.369487 137274321021824 utils.py:1231] [24850] train/loss = 2.875247836112976
+I1129 17:36:34.369732 137274321021824 utils.py:1231] [24850] l2_grads = 1.4301950931549072
+I1129 17:36:34.369863 137274321021824 utils.py:1231] [24850] lr = 0.0009492051339461843
+I1129 17:36:34.369957 137274321021824 utils.py:1231] [24850] uptime = 157583.73231395398
+I1129 17:36:34.370038 137274321021824 utils.py:1231] [24850] examples_seen = 25446400.0
+I1129 17:36:34.370107 137274321021824 utils.py:1231] [24850] progress = 0.22068683782847703
+I1129 17:36:34.370197 137274321021824 utils.py:1231] [24850] epoch = 19.86189154107154
+I1129 17:36:34.370288 137274321021824 utils.py:1231] [24850] img/sec/core = 164.18788236696616
+I1129 17:36:34.370380 137274321021824 utils.py:1231] [24850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.73894137374166
+I1129 17:36:34.370469 137274321021824 utils.py:1231] [24850] core_hours = 43.73894137374166
+I1129 17:36:34.370556 137274321021824 train.py:125] NOTE: Steps:24850/112603 [22.1%]
+Walltime:1d19h46m (0s eval)
+ETA:6d10h28m
+Total train time:8d6h12m
+I1129 17:41:46.189003 137274321021824 utils.py:1231] [24900] l2_params = 331.1622866186273
+I1129 17:41:46.189269 137274321021824 utils.py:1231] [24900] train/loss = 3.3909180760383606
+I1129 17:41:46.189470 137274321021824 utils.py:1231] [24900] l2_grads = 1.2009197473526
+I1129 17:41:46.189546 137274321021824 utils.py:1231] [24900] lr = 0.0009488684447667458
+I1129 17:41:46.189606 137274321021824 utils.py:1231] [24900] uptime = 157895.55196734902
+I1129 17:41:46.189666 137274321021824 utils.py:1231] [24900] examples_seen = 25497600.0
+I1129 17:41:46.189734 137274321021824 utils.py:1231] [24900] progress = 0.22113087573155243
+I1129 17:41:46.189797 137274321021824 utils.py:1231] [24900] epoch = 19.901855105540495
+I1129 17:41:46.189861 137274321021824 utils.py:1231] [24900] img/sec/core = 164.19747582472067
+I1129 17:41:46.189932 137274321021824 utils.py:1231] [24900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.82555794412916
+I1129 17:41:46.189998 137274321021824 utils.py:1231] [24900] core_hours = 43.82555794412916
+I1129 17:41:46.190070 137274321021824 train.py:125] NOTE: Steps:24900/112603 [22.1%]
+Walltime:1d19h51m (0s eval)
+ETA:6d10h22m
+Total train time:8d6h12m
+I1129 17:46:57.974116 137274321021824 utils.py:1231] [24950] l2_params = 331.1586829364569
+I1129 17:46:57.974349 137274321021824 utils.py:1231] [24950] train/loss = 3.1302131712436676
+I1129 17:46:57.974467 137274321021824 utils.py:1231] [24950] l2_grads = 1.2913132905960083
+I1129 17:46:57.974538 137274321021824 utils.py:1231] [24950] lr = 0.000948530703531867
+I1129 17:46:57.974597 137274321021824 utils.py:1231] [24950] uptime = 158207.33695842698
+I1129 17:46:57.974657 137274321021824 utils.py:1231] [24950] examples_seen = 25548800.0
+I1129 17:46:57.974721 137274321021824 utils.py:1231] [24950] progress = 0.22157491363462786
+I1129 17:46:57.974778 137274321021824 utils.py:1231] [24950] epoch = 19.941818670009454
+I1129 17:46:57.974836 137274321021824 utils.py:1231] [24950] img/sec/core = 164.21573027930833
+I1129 17:46:57.974906 137274321021824 utils.py:1231] [24950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.91216488609527
+I1129 17:46:57.974962 137274321021824 utils.py:1231] [24950] core_hours = 43.91216488609527
+I1129 17:46:57.975029 137274321021824 train.py:125] NOTE: Steps:24950/112603 [22.2%]
+Walltime:1d19h56m (0s eval)
+ETA:6d10h16m
+Total train time:8d6h11m
+I1129 17:52:09.743652 137274321021824 utils.py:1231] [25000] l2_params = 331.19439235105284
+I1129 17:52:09.743867 137274321021824 utils.py:1231] [25000] train/loss = 3.212635815143585
+I1129 17:52:09.743987 137274321021824 utils.py:1231] [25000] l2_grads = 1.308467984199524
+I1129 17:52:09.744050 137274321021824 utils.py:1231] [25000] lr = 0.0009481919110331444
+I1129 17:52:09.744103 137274321021824 utils.py:1231] [25000] uptime = 158519.106464562
+I1129 17:52:09.744155 137274321021824 utils.py:1231] [25000] examples_seen = 25600000.0
+I1129 17:52:09.744214 137274321021824 utils.py:1231] [25000] progress = 0.22201895153770326
+I1129 17:52:09.744263 137274321021824 utils.py:1231] [25000] epoch = 19.98178223447841
+I1129 17:52:09.744321 137274321021824 utils.py:1231] [25000] img/sec/core = 164.2238865331191
+I1129 17:52:09.744378 137274321021824 utils.py:1231] [25000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 43.998767526688326
+I1129 17:52:09.744429 137274321021824 utils.py:1231] [25000] core_hours = 43.998767526688326
+I1129 17:52:09.744489 137274321021824 train.py:125] NOTE: Steps:25000/112603 [22.2%]
+Walltime:1d20h1m (0s eval)
+ETA:6d10h11m
+Total train time:8d6h11m
+I1129 17:52:10.080025 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:25000/112603 [22.2%]
+Walltime:1d20h1m (0s eval)
+ETA:6d10h11m
+Total train time:8d6h11m
+I1129 17:53:47.516324 137274321021824 utils.py:1231] [25000] val/acc@1 = 0.5568000637755102
+I1129 17:53:47.516568 137274321021824 utils.py:1231] [25000] val/loss = 1.8820168306024707
+I1129 17:53:47.516720 137274321021824 utils.py:1231] [25000] z/secs/eval/val = 97.43645310300053
+I1129 17:53:47.516789 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.43645310300053
+I1129 17:58:59.296841 137274321021824 utils.py:1231] [25050] l2_params = 331.2367223290889
+I1129 17:58:59.297110 137274321021824 utils.py:1231] [25050] train/loss = 5.317238390445709
+I1129 17:58:59.297209 137274321021824 utils.py:1231] [25050] l2_grads = 0.9426098465919495
+I1129 17:58:59.297281 137274321021824 utils.py:1231] [25050] lr = 0.000947852068064637
+I1129 17:58:59.297344 137274321021824 utils.py:1231] [25050] uptime = 158928.659705261
+I1129 17:58:59.297402 137274321021824 utils.py:1231] [25050] examples_seen = 25651200.0
+I1129 17:58:59.297460 137274321021824 utils.py:1231] [25050] progress = 0.22246298944077866
+I1129 17:58:59.297513 137274321021824 utils.py:1231] [25050] epoch = 20.021745798947364
+I1129 17:58:59.297570 137274321021824 utils.py:1231] [25050] img/sec/core = 125.01427143541095
+I1129 17:58:59.297632 137274321021824 utils.py:1231] [25050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.11253231577139
+I1129 17:58:59.297685 137274321021824 utils.py:1231] [25050] core_hours = 44.11253231577139
+I1129 17:58:59.297752 137274321021824 train.py:125] NOTE: Steps:25050/112603 [22.2%]
+Walltime:1d20h8m (0s eval)
+ETA:6d10h11m
+Total train time:8d6h18m
+I1129 18:04:10.962347 137274321021824 utils.py:1231] [25100] l2_params = 331.2340553666696
+I1129 18:04:10.962616 137274321021824 utils.py:1231] [25100] train/loss = 3.0683701038360596
+I1129 18:04:10.962713 137274321021824 utils.py:1231] [25100] l2_grads = 1.2754802703857422
+I1129 18:04:10.962786 137274321021824 utils.py:1231] [25100] lr = 0.000947511175422868
+I1129 18:04:10.962851 137274321021824 utils.py:1231] [25100] uptime = 159240.32521307998
+I1129 18:04:10.962911 137274321021824 utils.py:1231] [25100] examples_seen = 25702400.0
+I1129 18:04:10.962961 137274321021824 utils.py:1231] [25100] progress = 0.22290702734385406
+I1129 18:04:10.963009 137274321021824 utils.py:1231] [25100] epoch = 20.061709363416323
+I1129 18:04:10.963059 137274321021824 utils.py:1231] [25100] img/sec/core = 164.27868569189818
+I1129 18:04:10.963115 137274321021824 utils.py:1231] [25100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.199106067943326
+I1129 18:04:10.963167 137274321021824 utils.py:1231] [25100] core_hours = 44.199106067943326
+I1129 18:04:10.963234 137274321021824 train.py:125] NOTE: Steps:25100/112603 [22.3%]
+Walltime:1d20h14m (0s eval)
+ETA:6d10h5m
+Total train time:8d6h18m
+I1129 18:09:22.752986 137274321021824 utils.py:1231] [25150] l2_params = 331.28264079365596
+I1129 18:09:22.753262 137274321021824 utils.py:1231] [25150] train/loss = 3.2769213020801544
+I1129 18:09:22.753375 137274321021824 utils.py:1231] [25150] l2_grads = 1.2791671752929688
+I1129 18:09:22.753443 137274321021824 utils.py:1231] [25150] lr = 0.0009471692339068186
+I1129 18:09:22.753499 137274321021824 utils.py:1231] [25150] uptime = 159552.115854837
+I1129 18:09:22.753566 137274321021824 utils.py:1231] [25150] examples_seen = 25753600.0
+I1129 18:09:22.753616 137274321021824 utils.py:1231] [25150] progress = 0.22335106524692946
+I1129 18:09:22.753663 137274321021824 utils.py:1231] [25150] epoch = 20.10167292788528
+I1129 18:09:22.753712 137274321021824 utils.py:1231] [25150] img/sec/core = 164.21275414642022
+I1129 18:09:22.753767 137274321021824 utils.py:1231] [25150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.2857145795425
+I1129 18:09:22.753815 137274321021824 utils.py:1231] [25150] core_hours = 44.2857145795425
+I1129 18:09:22.753872 137274321021824 train.py:125] NOTE: Steps:25150/112603 [22.3%]
+Walltime:1d20h19m (0s eval)
+ETA:6d10h0m
+Total train time:8d6h17m
+I1129 18:14:34.544266 137274321021824 utils.py:1231] [25200] l2_params = 331.2874202949799
+I1129 18:14:34.544477 137274321021824 utils.py:1231] [25200] train/loss = 3.0583971738815308
+I1129 18:14:34.544584 137274321021824 utils.py:1231] [25200] l2_grads = 1.5490024089813232
+I1129 18:14:34.544655 137274321021824 utils.py:1231] [25200] lr = 0.0009468262443179307
+I1129 18:14:34.544721 137274321021824 utils.py:1231] [25200] uptime = 159863.907082745
+I1129 18:14:34.544782 137274321021824 utils.py:1231] [25200] examples_seen = 25804800.0
+I1129 18:14:34.544838 137274321021824 utils.py:1231] [25200] progress = 0.2237951031500049
+I1129 18:14:34.544902 137274321021824 utils.py:1231] [25200] epoch = 20.141636492354237
+I1129 18:14:34.544963 137274321021824 utils.py:1231] [25200] img/sec/core = 164.21244543515243
+I1129 18:14:34.545048 137274321021824 utils.py:1231] [25200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.37232325396139
+I1129 18:14:34.545109 137274321021824 utils.py:1231] [25200] core_hours = 44.37232325396139
+I1129 18:14:34.545171 137274321021824 train.py:125] NOTE: Steps:25200/112603 [22.4%]
+Walltime:1d20h24m (0s eval)
+ETA:6d9h54m
+Total train time:8d6h17m
+I1129 18:19:46.335532 137274321021824 utils.py:1231] [25250] l2_params = 331.3303259952966
+I1129 18:19:46.335777 137274321021824 utils.py:1231] [25250] train/loss = 4.560173273086548
+I1129 18:19:46.335916 137274321021824 utils.py:1231] [25250] l2_grads = 0.9794830679893494
+I1129 18:19:46.336006 137274321021824 utils.py:1231] [25250] lr = 0.0009464822074601006
+I1129 18:19:46.336070 137274321021824 utils.py:1231] [25250] uptime = 160175.69843182602
+I1129 18:19:46.336133 137274321021824 utils.py:1231] [25250] examples_seen = 25856000.0
+I1129 18:19:46.336197 137274321021824 utils.py:1231] [25250] progress = 0.2242391410530803
+I1129 18:19:46.336259 137274321021824 utils.py:1231] [25250] epoch = 20.181600056823193
+I1129 18:19:46.336328 137274321021824 utils.py:1231] [25250] img/sec/core = 164.21238161645852
+I1129 18:19:46.336395 137274321021824 utils.py:1231] [25250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.458931962039436
+I1129 18:19:46.336449 137274321021824 utils.py:1231] [25250] core_hours = 44.458931962039436
+I1129 18:19:46.336509 137274321021824 train.py:125] NOTE: Steps:25250/112603 [22.4%]
+Walltime:1d20h29m (0s eval)
+ETA:6d9h49m
+Total train time:8d6h16m
+I1129 18:24:58.129058 137274321021824 utils.py:1231] [25300] l2_params = 331.33931116633124
+I1129 18:24:58.129296 137274321021824 utils.py:1231] [25300] train/loss = 2.794214218854904
+I1129 18:24:58.129455 137274321021824 utils.py:1231] [25300] l2_grads = 1.2355332374572754
+I1129 18:24:58.129558 137274321021824 utils.py:1231] [25300] lr = 0.0009461371241396807
+I1129 18:24:58.129643 137274321021824 utils.py:1231] [25300] uptime = 160487.49200068502
+I1129 18:24:58.129740 137274321021824 utils.py:1231] [25300] examples_seen = 25907200.0
+I1129 18:24:58.129808 137274321021824 utils.py:1231] [25300] progress = 0.2246831789561557
+I1129 18:24:58.129878 137274321021824 utils.py:1231] [25300] epoch = 20.22156362129215
+I1129 18:24:58.129966 137274321021824 utils.py:1231] [25300] img/sec/core = 164.21121252553766
+I1129 18:24:58.130043 137274321021824 utils.py:1231] [25300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.54554128672249
+I1129 18:24:58.130114 137274321021824 utils.py:1231] [25300] core_hours = 44.54554128672249
+I1129 18:24:58.130197 137274321021824 train.py:125] NOTE: Steps:25300/112603 [22.5%]
+Walltime:1d20h34m (0s eval)
+ETA:6d9h43m
+Total train time:8d6h16m
+I1129 18:30:09.921584 137274321021824 utils.py:1231] [25350] l2_params = 331.35711966989265
+I1129 18:30:09.921964 137274321021824 utils.py:1231] [25350] train/loss = 2.9001066386699677
+I1129 18:30:09.922220 137274321021824 utils.py:1231] [25350] l2_grads = 1.203501582145691
+I1129 18:30:09.922331 137274321021824 utils.py:1231] [25350] lr = 0.000945790995165475
+I1129 18:30:09.922432 137274321021824 utils.py:1231] [25350] uptime = 160799.28479246498
+I1129 18:30:09.922497 137274321021824 utils.py:1231] [25350] examples_seen = 25958400.0
+I1129 18:30:09.922558 137274321021824 utils.py:1231] [25350] progress = 0.2251272168592311
+I1129 18:30:09.922619 137274321021824 utils.py:1231] [25350] epoch = 20.261527185761107
+I1129 18:30:09.922681 137274321021824 utils.py:1231] [25350] img/sec/core = 164.21162178799514
+I1129 18:30:09.922749 137274321021824 utils.py:1231] [25350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.63215039555027
+I1129 18:30:09.922812 137274321021824 utils.py:1231] [25350] core_hours = 44.63215039555027
+I1129 18:30:09.922901 137274321021824 train.py:125] NOTE: Steps:25350/112603 [22.5%]
+Walltime:1d20h39m (0s eval)
+ETA:6d9h37m
+Total train time:8d6h16m
+I1129 18:35:21.625727 137274321021824 utils.py:1231] [25400] l2_params = 331.3789813472713
+I1129 18:35:21.625940 137274321021824 utils.py:1231] [25400] train/loss = 3.026858329772949
+I1129 18:35:21.626035 137274321021824 utils.py:1231] [25400] l2_grads = 1.2872494459152222
+I1129 18:35:21.626094 137274321021824 utils.py:1231] [25400] lr = 0.0009454438213487387
+I1129 18:35:21.626144 137274321021824 utils.py:1231] [25400] uptime = 161110.988506208
+I1129 18:35:21.626196 137274321021824 utils.py:1231] [25400] examples_seen = 26009600.0
+I1129 18:35:21.626244 137274321021824 utils.py:1231] [25400] progress = 0.22557125476230652
+I1129 18:35:21.626291 137274321021824 utils.py:1231] [25400] epoch = 20.301490750230062
+I1129 18:35:21.626338 137274321021824 utils.py:1231] [25400] img/sec/core = 164.2585498426618
+I1129 18:35:21.626394 137274321021824 utils.py:1231] [25400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.71873476047889
+I1129 18:35:21.626443 137274321021824 utils.py:1231] [25400] core_hours = 44.71873476047889
+I1129 18:35:21.626502 137274321021824 train.py:125] NOTE: Steps:25400/112603 [22.6%]
+Walltime:1d20h45m (0s eval)
+ETA:6d9h32m
+Total train time:8d6h15m
+I1129 18:40:33.246124 137274321021824 utils.py:1231] [25450] l2_params = 331.44417408807806
+I1129 18:40:33.246332 137274321021824 utils.py:1231] [25450] train/loss = 3.073109954595566
+I1129 18:40:33.246446 137274321021824 utils.py:1231] [25450] l2_grads = 1.3938062191009521
+I1129 18:40:33.246516 137274321021824 utils.py:1231] [25450] lr = 0.0009450956035031757
+I1129 18:40:33.246575 137274321021824 utils.py:1231] [25450] uptime = 161422.608936764
+I1129 18:40:33.246636 137274321021824 utils.py:1231] [25450] examples_seen = 26060800.0
+I1129 18:40:33.246701 137274321021824 utils.py:1231] [25450] progress = 0.22601529266538192
+I1129 18:40:33.246757 137274321021824 utils.py:1231] [25450] epoch = 20.34145431469902
+I1129 18:40:33.246820 137274321021824 utils.py:1231] [25450] img/sec/core = 164.3024493248118
+I1129 18:40:33.246901 137274321021824 utils.py:1231] [25450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.805295991188885
+I1129 18:40:33.246961 137274321021824 utils.py:1231] [25450] core_hours = 44.805295991188885
+I1129 18:40:33.247023 137274321021824 train.py:125] NOTE: Steps:25450/112603 [22.6%]
+Walltime:1d20h50m (0s eval)
+ETA:6d9h26m
+Total train time:8d6h15m
+I1129 18:45:45.039534 137274321021824 utils.py:1231] [25500] l2_params = 331.4608649444986
+I1129 18:45:45.039777 137274321021824 utils.py:1231] [25500] train/loss = 2.9831913709640503
+I1129 18:45:45.039905 137274321021824 utils.py:1231] [25500] l2_grads = 1.4462063312530518
+I1129 18:45:45.040052 137274321021824 utils.py:1231] [25500] lr = 0.0009447463424449376
+I1129 18:45:45.040131 137274321021824 utils.py:1231] [25500] uptime = 161734.40249147202
+I1129 18:45:45.040206 137274321021824 utils.py:1231] [25500] examples_seen = 26112000.0
+I1129 18:45:45.040263 137274321021824 utils.py:1231] [25500] progress = 0.22645933056845732
+I1129 18:45:45.040333 137274321021824 utils.py:1231] [25500] epoch = 20.381417879167977
+I1129 18:45:45.040395 137274321021824 utils.py:1231] [25500] img/sec/core = 164.21121997838486
+I1129 18:45:45.040459 137274321021824 utils.py:1231] [25500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.8919053119411
+I1129 18:45:45.040519 137274321021824 utils.py:1231] [25500] core_hours = 44.8919053119411
+I1129 18:45:45.040593 137274321021824 train.py:125] NOTE: Steps:25500/112603 [22.6%]
+Walltime:1d20h55m (0s eval)
+ETA:6d9h21m
+Total train time:8d6h14m
+I1129 18:50:56.816787 137274321021824 utils.py:1231] [25550] l2_params = 331.4822040843699
+I1129 18:50:56.817073 137274321021824 utils.py:1231] [25550] train/loss = 2.7984738945961
+I1129 18:50:56.817260 137274321021824 utils.py:1231] [25550] l2_grads = 1.3539074659347534
+I1129 18:50:56.817357 137274321021824 utils.py:1231] [25550] lr = 0.0009443960389926206
+I1129 18:50:56.817439 137274321021824 utils.py:1231] [25550] uptime = 162046.179792349
+I1129 18:50:56.817502 137274321021824 utils.py:1231] [25550] examples_seen = 26163200.0
+I1129 18:50:56.817572 137274321021824 utils.py:1231] [25550] progress = 0.22690336847153272
+I1129 18:50:56.817635 137274321021824 utils.py:1231] [25550] epoch = 20.421381443636935
+I1129 18:50:56.817697 137274321021824 utils.py:1231] [25550] img/sec/core = 164.21978077293937
+I1129 18:50:56.817767 137274321021824 utils.py:1231] [25550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 44.97851011774027
+I1129 18:50:56.817835 137274321021824 utils.py:1231] [25550] core_hours = 44.97851011774027
+I1129 18:50:56.817919 137274321021824 train.py:125] NOTE: Steps:25550/112603 [22.7%]
+Walltime:1d21h0m (0s eval)
+ETA:6d9h15m
+Total train time:8d6h14m
+I1129 18:56:08.614760 137274321021824 utils.py:1231] [25600] l2_params = 331.50773510459936
+I1129 18:56:08.615001 137274321021824 utils.py:1231] [25600] train/loss = 2.934423804283142
+I1129 18:56:08.615158 137274321021824 utils.py:1231] [25600] l2_grads = 1.2922202348709106
+I1129 18:56:08.615238 137274321021824 utils.py:1231] [25600] lr = 0.0009440446939672631
+I1129 18:56:08.615304 137274321021824 utils.py:1231] [25600] uptime = 162357.97766497702
+I1129 18:56:08.615364 137274321021824 utils.py:1231] [25600] examples_seen = 26214400.0
+I1129 18:56:08.615420 137274321021824 utils.py:1231] [25600] progress = 0.22734740637460812
+I1129 18:56:08.615475 137274321021824 utils.py:1231] [25600] epoch = 20.46134500810589
+I1129 18:56:08.615538 137274321021824 utils.py:1231] [25600] img/sec/core = 164.2089459060744
+I1129 18:56:08.615601 137274321021824 utils.py:1231] [25600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.06512063791472
+I1129 18:56:08.615657 137274321021824 utils.py:1231] [25600] core_hours = 45.06512063791472
+I1129 18:56:08.615723 137274321021824 train.py:125] NOTE: Steps:25600/112603 [22.7%]
+Walltime:1d21h5m (0s eval)
+ETA:6d9h10m
+Total train time:8d6h14m
+I1129 19:01:20.403545 137274321021824 utils.py:1231] [25650] l2_params = 331.53678037973225
+I1129 19:01:20.403757 137274321021824 utils.py:1231] [25650] train/loss = 4.23326763510704
+I1129 19:01:20.403861 137274321021824 utils.py:1231] [25650] l2_grads = 1.1271984577178955
+I1129 19:01:20.403924 137274321021824 utils.py:1231] [25650] lr = 0.000943692308192347
+I1129 19:01:20.403976 137274321021824 utils.py:1231] [25650] uptime = 162669.76633869298
+I1129 19:01:20.404027 137274321021824 utils.py:1231] [25650] examples_seen = 26265600.0
+I1129 19:01:20.404073 137274321021824 utils.py:1231] [25650] progress = 0.22779144427768355
+I1129 19:01:20.404127 137274321021824 utils.py:1231] [25650] epoch = 20.501308572574846
+I1129 19:01:20.404176 137274321021824 utils.py:1231] [25650] img/sec/core = 164.21379067361184
+I1129 19:01:20.404232 137274321021824 utils.py:1231] [25650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.151728602835824
+I1129 19:01:20.404281 137274321021824 utils.py:1231] [25650] core_hours = 45.151728602835824
+I1129 19:01:20.404340 137274321021824 train.py:125] NOTE: Steps:25650/112603 [22.8%]
+Walltime:1d21h11m (0s eval)
+ETA:6d9h4m
+Total train time:8d6h13m
+I1129 19:06:32.199837 137274321021824 utils.py:1231] [25700] l2_params = 331.5270600646988
+I1129 19:06:32.200110 137274321021824 utils.py:1231] [25700] train/loss = 2.9966953694820404
+I1129 19:06:32.200232 137274321021824 utils.py:1231] [25700] l2_grads = 1.481960654258728
+I1129 19:06:32.200307 137274321021824 utils.py:1231] [25700] lr = 0.0009433388824937919
+I1129 19:06:32.200365 137274321021824 utils.py:1231] [25700] uptime = 162981.56272322603
+I1129 19:06:32.200424 137274321021824 utils.py:1231] [25700] examples_seen = 26316800.0
+I1129 19:06:32.200479 137274321021824 utils.py:1231] [25700] progress = 0.22823548218075895
+I1129 19:06:32.200530 137274321021824 utils.py:1231] [25700] epoch = 20.541272137043805
+I1129 19:06:32.200581 137274321021824 utils.py:1231] [25700] img/sec/core = 164.20972961787908
+I1129 19:06:32.200643 137274321021824 utils.py:1231] [25700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.23833870965055
+I1129 19:06:32.200707 137274321021824 utils.py:1231] [25700] core_hours = 45.23833870965055
+I1129 19:06:32.200768 137274321021824 train.py:125] NOTE: Steps:25700/112603 [22.8%]
+Walltime:1d21h16m (0s eval)
+ETA:6d8h58m
+Total train time:8d6h13m
+I1129 19:11:43.989579 137274321021824 utils.py:1231] [25750] l2_params = 331.5188152172139
+I1129 19:11:43.989806 137274321021824 utils.py:1231] [25750] train/loss = 2.780015140771866
+I1129 19:11:43.989920 137274321021824 utils.py:1231] [25750] l2_grads = 1.2728015184402466
+I1129 19:11:43.989995 137274321021824 utils.py:1231] [25750] lr = 0.0009429844176999541
+I1129 19:11:43.990058 137274321021824 utils.py:1231] [25750] uptime = 163293.35241676803
+I1129 19:11:43.990113 137274321021824 utils.py:1231] [25750] examples_seen = 26368000.0
+I1129 19:11:43.990164 137274321021824 utils.py:1231] [25750] progress = 0.22867952008383435
+I1129 19:11:43.990215 137274321021824 utils.py:1231] [25750] epoch = 20.58123570151276
+I1129 19:11:43.990268 137274321021824 utils.py:1231] [25750] img/sec/core = 164.21325355035552
+I1129 19:11:43.990326 137274321021824 utils.py:1231] [25750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.32494695785667
+I1129 19:11:43.990377 137274321021824 utils.py:1231] [25750] core_hours = 45.32494695785667
+I1129 19:11:43.990439 137274321021824 train.py:125] NOTE: Steps:25750/112603 [22.9%]
+Walltime:1d21h21m (0s eval)
+ETA:6d8h53m
+Total train time:8d6h13m
+I1129 19:16:55.789419 137274321021824 utils.py:1231] [25800] l2_params = 331.5099196333454
+I1129 19:16:55.789644 137274321021824 utils.py:1231] [25800] train/loss = 2.77773654460907
+I1129 19:16:55.789746 137274321021824 utils.py:1231] [25800] l2_grads = 1.4169960021972656
+I1129 19:16:55.789817 137274321021824 utils.py:1231] [25800] lr = 0.000942628914641627
+I1129 19:16:55.789890 137274321021824 utils.py:1231] [25800] uptime = 163605.152245829
+I1129 19:16:55.789962 137274321021824 utils.py:1231] [25800] examples_seen = 26419200.0
+I1129 19:16:55.790020 137274321021824 utils.py:1231] [25800] progress = 0.22912355798690975
+I1129 19:16:55.790077 137274321021824 utils.py:1231] [25800] epoch = 20.62119926598172
+I1129 19:16:55.790135 137274321021824 utils.py:1231] [25800] img/sec/core = 164.2079155533588
+I1129 19:16:55.790193 137274321021824 utils.py:1231] [25800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.411558021484716
+I1129 19:16:55.790248 137274321021824 utils.py:1231] [25800] core_hours = 45.411558021484716
+I1129 19:16:55.790324 137274321021824 train.py:125] NOTE: Steps:25800/112603 [22.9%]
+Walltime:1d21h26m (0s eval)
+ETA:6d8h47m
+Total train time:8d6h12m
+I1129 19:22:07.580928 137274321021824 utils.py:1231] [25850] l2_params = 331.5014498009703
+I1129 19:22:07.581202 137274321021824 utils.py:1231] [25850] train/loss = 4.472973167896271
+I1129 19:22:07.581377 137274321021824 utils.py:1231] [25850] l2_grads = 1.23674738407135
+I1129 19:22:07.581474 137274321021824 utils.py:1231] [25850] lr = 0.0009422723741520368
+I1129 19:22:07.581547 137274321021824 utils.py:1231] [25850] uptime = 163916.94390883
+I1129 19:22:07.581617 137274321021824 utils.py:1231] [25850] examples_seen = 26470400.0
+I1129 19:22:07.581691 137274321021824 utils.py:1231] [25850] progress = 0.22956759588998518
+I1129 19:22:07.581752 137274321021824 utils.py:1231] [25850] epoch = 20.661162830450674
+I1129 19:22:07.581810 137274321021824 utils.py:1231] [25850] img/sec/core = 164.2122162831394
+I1129 19:22:07.581873 137274321021824 utils.py:1231] [25850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.498166816762776
+I1129 19:22:07.581941 137274321021824 utils.py:1231] [25850] core_hours = 45.498166816762776
+I1129 19:22:07.582004 137274321021824 train.py:125] NOTE: Steps:25850/112603 [23.0%]
+Walltime:1d21h31m (0s eval)
+ETA:6d8h42m
+Total train time:8d6h12m
+I1129 19:27:19.457754 137274321021824 utils.py:1231] [25900] l2_params = 331.5196942375741
+I1129 19:27:19.457977 137274321021824 utils.py:1231] [25900] train/loss = 2.804853081703186
+I1129 19:27:19.458080 137274321021824 utils.py:1231] [25900] l2_grads = 1.3123424053192139
+I1129 19:27:19.458148 137274321021824 utils.py:1231] [25900] lr = 0.0009419147970668408
+I1129 19:27:19.458206 137274321021824 utils.py:1231] [25900] uptime = 164228.820567692
+I1129 19:27:19.458265 137274321021824 utils.py:1231] [25900] examples_seen = 26521600.0
+I1129 19:27:19.458324 137274321021824 utils.py:1231] [25900] progress = 0.23001163379306058
+I1129 19:27:19.458378 137274321021824 utils.py:1231] [25900] epoch = 20.701126394919633
+I1129 19:27:19.458433 137274321021824 utils.py:1231] [25900] img/sec/core = 164.16746346720544
+I1129 19:27:19.458494 137274321021824 utils.py:1231] [25900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.58479922200221
+I1129 19:27:19.458557 137274321021824 utils.py:1231] [25900] core_hours = 45.58479922200221
+I1129 19:27:19.458619 137274321021824 train.py:125] NOTE: Steps:25900/112603 [23.0%]
+Walltime:1d21h37m (0s eval)
+ETA:6d8h36m
+Total train time:8d6h12m
+I1129 19:32:31.240080 137274321021824 utils.py:1231] [25950] l2_params = 331.54174194792125
+I1129 19:32:31.240284 137274321021824 utils.py:1231] [25950] train/loss = 5.316332280635834
+I1129 19:32:31.240379 137274321021824 utils.py:1231] [25950] l2_grads = 1.032547116279602
+I1129 19:32:31.240437 137274321021824 utils.py:1231] [25950] lr = 0.0009415561842241264
+I1129 19:32:31.240486 137274321021824 utils.py:1231] [25950] uptime = 164540.60284824402
+I1129 19:32:31.240536 137274321021824 utils.py:1231] [25950] examples_seen = 26572800.0
+I1129 19:32:31.240583 137274321021824 utils.py:1231] [25950] progress = 0.23045567169613598
+I1129 19:32:31.240629 137274321021824 utils.py:1231] [25950] epoch = 20.74108995938859
+I1129 19:32:31.240679 137274321021824 utils.py:1231] [25950] img/sec/core = 164.21715791337655
+I1129 19:32:31.240732 137274321021824 utils.py:1231] [25950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.67140541104444
+I1129 19:32:31.240780 137274321021824 utils.py:1231] [25950] core_hours = 45.67140541104444
+I1129 19:32:31.240838 137274321021824 train.py:125] NOTE: Steps:25950/112603 [23.0%]
+Walltime:1d21h42m (0s eval)
+ETA:6d8h31m
+Total train time:8d6h11m
+I1129 19:37:42.946207 137274321021824 utils.py:1231] [26000] l2_params = 331.5349516404399
+I1129 19:37:42.946493 137274321021824 utils.py:1231] [26000] train/loss = 2.8798676133155823
+I1129 19:37:42.946662 137274321021824 utils.py:1231] [26000] l2_grads = 1.5394266843795776
+I1129 19:37:42.946731 137274321021824 utils.py:1231] [26000] lr = 0.0009411965364644084
+I1129 19:37:42.946792 137274321021824 utils.py:1231] [26000] uptime = 164852.309154447
+I1129 19:37:42.946849 137274321021824 utils.py:1231] [26000] examples_seen = 26624000.0
+I1129 19:37:42.946912 137274321021824 utils.py:1231] [26000] progress = 0.23089970959921138
+I1129 19:37:42.946967 137274321021824 utils.py:1231] [26000] epoch = 20.781053523857544
+I1129 19:37:42.947029 137274321021824 utils.py:1231] [26000] img/sec/core = 164.25718370501852
+I1129 19:37:42.947092 137274321021824 utils.py:1231] [26000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.757990496100824
+I1129 19:37:42.947146 137274321021824 utils.py:1231] [26000] core_hours = 45.757990496100824
+I1129 19:37:42.947222 137274321021824 train.py:125] NOTE: Steps:26000/112603 [23.1%]
+Walltime:1d21h47m (0s eval)
+ETA:6d8h25m
+Total train time:8d6h11m
+I1129 19:42:55.116558 137274321021824 utils.py:1231] [26050] l2_params = 331.56644521877445
+I1129 19:42:55.116765 137274321021824 utils.py:1231] [26050] train/loss = 2.889336884021759
+I1129 19:42:55.116885 137274321021824 utils.py:1231] [26050] l2_grads = 1.3535195589065552
+I1129 19:42:55.116961 137274321021824 utils.py:1231] [26050] lr = 0.000940835854630628
+I1129 19:42:55.117022 137274321021824 utils.py:1231] [26050] uptime = 165164.47938407602
+I1129 19:42:55.117082 137274321021824 utils.py:1231] [26050] examples_seen = 26675200.0
+I1129 19:42:55.117139 137274321021824 utils.py:1231] [26050] progress = 0.23134374750228678
+I1129 19:42:55.117195 137274321021824 utils.py:1231] [26050] epoch = 20.821017088326503
+I1129 19:42:55.117252 137274321021824 utils.py:1231] [26050] img/sec/core = 164.01307729070922
+I1129 19:42:55.117320 137274321021824 utils.py:1231] [26050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.84470444877555
+I1129 19:42:55.117375 137274321021824 utils.py:1231] [26050] core_hours = 45.84470444877555
+I1129 19:42:55.117440 137274321021824 train.py:125] NOTE: Steps:26050/112603 [23.1%]
+Walltime:1d21h52m (0s eval)
+ETA:6d8h20m
+Total train time:8d6h10m
+I1129 19:48:06.911210 137274321021824 utils.py:1231] [26100] l2_params = 331.57077741240914
+I1129 19:48:06.911441 137274321021824 utils.py:1231] [26100] train/loss = 3.4547001719474792
+I1129 19:48:06.911559 137274321021824 utils.py:1231] [26100] l2_grads = 1.1977217197418213
+I1129 19:48:06.911630 137274321021824 utils.py:1231] [26100] lr = 0.0009404741395681484
+I1129 19:48:06.911691 137274321021824 utils.py:1231] [26100] uptime = 165476.27405248198
+I1129 19:48:06.911761 137274321021824 utils.py:1231] [26100] examples_seen = 26726400.0
+I1129 19:48:06.911819 137274321021824 utils.py:1231] [26100] progress = 0.2317877854053622
+I1129 19:48:06.911888 137274321021824 utils.py:1231] [26100] epoch = 20.860980652795458
+I1129 19:48:06.911947 137274321021824 utils.py:1231] [26100] img/sec/core = 164.2106334330668
+I1129 19:48:06.912010 137274321021824 utils.py:1231] [26100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 45.93131407888833
+I1129 19:48:06.912067 137274321021824 utils.py:1231] [26100] core_hours = 45.93131407888833
+I1129 19:48:06.912135 137274321021824 train.py:125] NOTE: Steps:26100/112603 [23.2%]
+Walltime:1d21h57m (0s eval)
+ETA:6d8h14m
+Total train time:8d6h10m
+I1129 19:53:18.702284 137274321021824 utils.py:1231] [26150] l2_params = 331.6170262132267
+I1129 19:53:18.702521 137274321021824 utils.py:1231] [26150] train/loss = 3.276287764310837
+I1129 19:53:18.702632 137274321021824 utils.py:1231] [26150] l2_grads = 1.177394986152649
+I1129 19:53:18.702707 137274321021824 utils.py:1231] [26150] lr = 0.0009401113921247558
+I1129 19:53:18.702768 137274321021824 utils.py:1231] [26150] uptime = 165788.065129488
+I1129 19:53:18.702830 137274321021824 utils.py:1231] [26150] examples_seen = 26777600.0
+I1129 19:53:18.702895 137274321021824 utils.py:1231] [26150] progress = 0.2322318233084376
+I1129 19:53:18.702955 137274321021824 utils.py:1231] [26150] epoch = 20.900944217264417
+I1129 19:53:18.703015 137274321021824 utils.py:1231] [26150] img/sec/core = 164.21252491140152
+I1129 19:53:18.703075 137274321021824 utils.py:1231] [26150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.01792271138999
+I1129 19:53:18.703132 137274321021824 utils.py:1231] [26150] core_hours = 46.01792271138999
+I1129 19:53:18.703206 137274321021824 train.py:125] NOTE: Steps:26150/112603 [23.2%]
+Walltime:1d22h3m (0s eval)
+ETA:6d8h8m
+Total train time:8d6h10m
+I1129 19:58:30.493851 137274321021824 utils.py:1231] [26200] l2_params = 331.62418517685916
+I1129 19:58:30.494099 137274321021824 utils.py:1231] [26200] train/loss = 5.073564648628235
+I1129 19:58:30.494217 137274321021824 utils.py:1231] [26200] l2_grads = 0.9902811646461487
+I1129 19:58:30.494283 137274321021824 utils.py:1231] [26200] lr = 0.0009397476131506551
+I1129 19:58:30.494342 137274321021824 utils.py:1231] [26200] uptime = 166099.85670423502
+I1129 19:58:30.494401 137274321021824 utils.py:1231] [26200] examples_seen = 26828800.0
+I1129 19:58:30.494448 137274321021824 utils.py:1231] [26200] progress = 0.232675861211513
+I1129 19:58:30.494493 137274321021824 utils.py:1231] [26200] epoch = 20.940907781733372
+I1129 19:58:30.494541 137274321021824 utils.py:1231] [26200] img/sec/core = 164.21226276413645
+I1129 19:58:30.494595 137274321021824 utils.py:1231] [26200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.10453148215306
+I1129 19:58:30.494645 137274321021824 utils.py:1231] [26200] core_hours = 46.10453148215306
+I1129 19:58:30.494703 137274321021824 train.py:125] NOTE: Steps:26200/112603 [23.3%]
+Walltime:1d22h8m (0s eval)
+ETA:6d8h3m
+Total train time:8d6h9m
+I1129 20:03:42.305783 137274321021824 utils.py:1231] [26250] l2_params = 331.6286070431052
+I1129 20:03:42.306002 137274321021824 utils.py:1231] [26250] train/loss = 3.0127586126327515
+I1129 20:03:42.306110 137274321021824 utils.py:1231] [26250] l2_grads = 1.2730847597122192
+I1129 20:03:42.306173 137274321021824 utils.py:1231] [26250] lr = 0.0009393828034984707
+I1129 20:03:42.306240 137274321021824 utils.py:1231] [26250] uptime = 166411.66860226903
+I1129 20:03:42.306300 137274321021824 utils.py:1231] [26250] examples_seen = 26880000.0
+I1129 20:03:42.306348 137274321021824 utils.py:1231] [26250] progress = 0.2331198991145884
+I1129 20:03:42.306397 137274321021824 utils.py:1231] [26250] epoch = 20.98087134620233
+I1129 20:03:42.306448 137274321021824 utils.py:1231] [26250] img/sec/core = 164.2015597314246
+I1129 20:03:42.306504 137274321021824 utils.py:1231] [26250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.191145898273604
+I1129 20:03:42.306555 137274321021824 utils.py:1231] [26250] core_hours = 46.191145898273604
+I1129 20:03:42.306617 137274321021824 train.py:125] NOTE: Steps:26250/112603 [23.3%]
+Walltime:1d22h13m (0s eval)
+ETA:6d7h57m
+Total train time:8d6h9m
+I1129 20:09:06.856548 137274321021824 utils.py:1231] [26300] l2_params = 331.67179333748766
+I1129 20:09:06.856848 137274321021824 utils.py:1231] [26300] train/loss = 5.369876146316528
+I1129 20:09:06.857021 137274321021824 utils.py:1231] [26300] l2_grads = 0.9942061901092529
+I1129 20:09:06.857100 137274321021824 utils.py:1231] [26300] lr = 0.0009390169640232402
+I1129 20:09:06.857161 137274321021824 utils.py:1231] [26300] uptime = 166736.21952210699
+I1129 20:09:06.857227 137274321021824 utils.py:1231] [26300] examples_seen = 26931200.0
+I1129 20:09:06.857288 137274321021824 utils.py:1231] [26300] progress = 0.23356393701766381
+I1129 20:09:06.857362 137274321021824 utils.py:1231] [26300] epoch = 21.020834910671287
+I1129 20:09:06.857419 137274321021824 utils.py:1231] [26300] img/sec/core = 157.75644704861253
+I1129 20:09:06.857492 137274321021824 utils.py:1231] [26300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.28129893156194
+I1129 20:09:06.857568 137274321021824 utils.py:1231] [26300] core_hours = 46.28129893156194
+I1129 20:09:06.857657 137274321021824 train.py:125] NOTE: Steps:26300/112603 [23.4%]
+Walltime:1d22h18m (0s eval)
+ETA:6d7h52m
+Total train time:8d6h10m
+I1129 20:14:18.771672 137274321021824 utils.py:1231] [26350] l2_params = 331.6735658068346
+I1129 20:14:18.771932 137274321021824 utils.py:1231] [26350] train/loss = 3.497067928314209
+I1129 20:14:18.772068 137274321021824 utils.py:1231] [26350] l2_grads = 1.1276178359985352
+I1129 20:14:18.772166 137274321021824 utils.py:1231] [26350] lr = 0.0009386500955824175
+I1129 20:14:18.772248 137274321021824 utils.py:1231] [26350] uptime = 167048.134603902
+I1129 20:14:18.772341 137274321021824 utils.py:1231] [26350] examples_seen = 26982400.0
+I1129 20:14:18.772425 137274321021824 utils.py:1231] [26350] progress = 0.23400797492073924
+I1129 20:14:18.772497 137274321021824 utils.py:1231] [26350] epoch = 21.060798475140242
+I1129 20:14:18.772582 137274321021824 utils.py:1231] [26350] img/sec/core = 164.1472406699715
+I1129 20:14:18.772696 137274321021824 utils.py:1231] [26350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.36794200983833
+I1129 20:14:18.772775 137274321021824 utils.py:1231] [26350] core_hours = 46.36794200983833
+I1129 20:14:18.772857 137274321021824 train.py:125] NOTE: Steps:26350/112603 [23.4%]
+Walltime:1d22h24m (0s eval)
+ETA:6d7h47m
+Total train time:8d6h9m
+I1129 20:19:30.649059 137274321021824 utils.py:1231] [26400] l2_params = 331.68099138315614
+I1129 20:19:30.649321 137274321021824 utils.py:1231] [26400] train/loss = 4.470608711242676
+I1129 20:19:30.649488 137274321021824 utils.py:1231] [26400] l2_grads = 0.9802024960517883
+I1129 20:19:30.649564 137274321021824 utils.py:1231] [26400] lr = 0.000938282199035866
+I1129 20:19:30.649623 137274321021824 utils.py:1231] [26400] uptime = 167360.01198365598
+I1129 20:19:30.649676 137274321021824 utils.py:1231] [26400] examples_seen = 27033600.0
+I1129 20:19:30.649729 137274321021824 utils.py:1231] [26400] progress = 0.23445201282381464
+I1129 20:19:30.649781 137274321021824 utils.py:1231] [26400] epoch = 21.1007620396092
+I1129 20:19:30.649850 137274321021824 utils.py:1231] [26400] img/sec/core = 164.16708400073458
+I1129 20:19:30.649926 137274321021824 utils.py:1231] [26400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.45457461532555
+I1129 20:19:30.649978 137274321021824 utils.py:1231] [26400] core_hours = 46.45457461532555
+I1129 20:19:30.650059 137274321021824 train.py:125] NOTE: Steps:26400/112603 [23.4%]
+Walltime:1d22h29m (0s eval)
+ETA:6d7h41m
+Total train time:8d6h9m
+I1129 20:24:46.228269 137274321021824 utils.py:1231] [26450] l2_params = 331.7137254594975
+I1129 20:24:46.228499 137274321021824 utils.py:1231] [26450] train/loss = 3.302089273929596
+I1129 20:24:46.228628 137274321021824 utils.py:1231] [26450] l2_grads = 1.2140856981277466
+I1129 20:24:46.228724 137274321021824 utils.py:1231] [26450] lr = 0.0009379132752458597
+I1129 20:24:46.228793 137274321021824 utils.py:1231] [26450] uptime = 167675.591154018
+I1129 20:24:46.228861 137274321021824 utils.py:1231] [26450] examples_seen = 27084800.0
+I1129 20:24:46.228932 137274321021824 utils.py:1231] [26450] progress = 0.23489605072689004
+I1129 20:24:46.228992 137274321021824 utils.py:1231] [26450] epoch = 21.140725604078156
+I1129 20:24:46.229050 137274321021824 utils.py:1231] [26450] img/sec/core = 162.24137968696658
+I1129 20:24:46.229114 137274321021824 utils.py:1231] [26450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.54223549598166
+I1129 20:24:46.229180 137274321021824 utils.py:1231] [26450] core_hours = 46.54223549598166
+I1129 20:24:46.229264 137274321021824 train.py:125] NOTE: Steps:26450/112603 [23.5%]
+Walltime:1d22h34m (0s eval)
+ETA:6d7h36m
+Total train time:8d6h9m
+I1129 20:29:58.107800 137274321021824 utils.py:1231] [26500] l2_params = 331.7276955820111
+I1129 20:29:58.108035 137274321021824 utils.py:1231] [26500] train/loss = 2.9530281722545624
+I1129 20:29:58.108150 137274321021824 utils.py:1231] [26500] l2_grads = 1.2937902212142944
+I1129 20:29:58.108227 137274321021824 utils.py:1231] [26500] lr = 0.000937543325077081
+I1129 20:29:58.108289 137274321021824 utils.py:1231] [26500] uptime = 167987.47064973298
+I1129 20:29:58.108361 137274321021824 utils.py:1231] [26500] examples_seen = 27136000.0
+I1129 20:29:58.108419 137274321021824 utils.py:1231] [26500] progress = 0.23534008862996544
+I1129 20:29:58.108478 137274321021824 utils.py:1231] [26500] epoch = 21.180689168547115
+I1129 20:29:58.108537 137274321021824 utils.py:1231] [26500] img/sec/core = 164.16597020147856
+I1129 20:29:58.108603 137274321021824 utils.py:1231] [26500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.62886868923582
+I1129 20:29:58.108661 137274321021824 utils.py:1231] [26500] core_hours = 46.62886868923582
+I1129 20:29:58.108729 137274321021824 train.py:125] NOTE: Steps:26500/112603 [23.5%]
+Walltime:1d22h39m (0s eval)
+ETA:6d7h30m
+Total train time:8d6h8m
+I1129 20:35:21.084656 137274321021824 utils.py:1231] [26550] l2_params = 331.7587618357107
+I1129 20:35:21.084956 137274321021824 utils.py:1231] [26550] train/loss = 5.236930310726166
+I1129 20:35:21.085138 137274321021824 utils.py:1231] [26550] l2_grads = 0.9663101434707642
+I1129 20:35:21.085216 137274321021824 utils.py:1231] [26550] lr = 0.0009371723493966155
+I1129 20:35:21.085267 137274321021824 utils.py:1231] [26550] uptime = 168310.44762856
+I1129 20:35:21.085319 137274321021824 utils.py:1231] [26550] examples_seen = 27187200.0
+I1129 20:35:21.085367 137274321021824 utils.py:1231] [26550] progress = 0.23578412653304087
+I1129 20:35:21.085414 137274321021824 utils.py:1231] [26550] epoch = 21.22065273301607
+I1129 20:35:21.085466 137274321021824 utils.py:1231] [26550] img/sec/core = 158.5252304543474
+I1129 20:35:21.085521 137274321021824 utils.py:1231] [26550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.71858451668777
+I1129 20:35:21.085570 137274321021824 utils.py:1231] [26550] core_hours = 46.71858451668777
+I1129 20:35:21.085632 137274321021824 train.py:125] NOTE: Steps:26550/112603 [23.6%]
+Walltime:1d22h45m (0s eval)
+ETA:6d7h26m
+Total train time:8d6h9m
+I1129 20:40:48.121048 137274321021824 utils.py:1231] [26600] l2_params = 331.74569584908215
+I1129 20:40:48.121348 137274321021824 utils.py:1231] [26600] train/loss = 5.302427709102631
+I1129 20:40:48.121542 137274321021824 utils.py:1231] [26600] l2_grads = 0.9995370507240295
+I1129 20:40:48.121630 137274321021824 utils.py:1231] [26600] lr = 0.000936800349073955
+I1129 20:40:48.121700 137274321021824 utils.py:1231] [26600] uptime = 168637.48406124202
+I1129 20:40:48.121762 137274321021824 utils.py:1231] [26600] examples_seen = 27238400.0
+I1129 20:40:48.121820 137274321021824 utils.py:1231] [26600] progress = 0.23622816443611627
+I1129 20:40:48.121877 137274321021824 utils.py:1231] [26600] epoch = 21.260616297485026
+I1129 20:40:48.121948 137274321021824 utils.py:1231] [26600] img/sec/core = 156.5574807066933
+I1129 20:40:48.122014 137274321021824 utils.py:1231] [26600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.80942797021055
+I1129 20:40:48.122071 137274321021824 utils.py:1231] [26600] core_hours = 46.80942797021055
+I1129 20:40:48.122141 137274321021824 train.py:125] NOTE: Steps:26600/112603 [23.6%]
+Walltime:1d22h50m (0s eval)
+ETA:6d7h21m
+Total train time:8d6h10m
+I1129 20:46:06.728112 137274321021824 utils.py:1231] [26650] l2_params = 331.74257045449116
+I1129 20:46:06.728420 137274321021824 utils.py:1231] [26650] train/loss = 4.448175549507141
+I1129 20:46:06.728602 137274321021824 utils.py:1231] [26650] l2_grads = 1.128232717514038
+I1129 20:46:06.728695 137274321021824 utils.py:1231] [26650] lr = 0.0009364273249809921
+I1129 20:46:06.728775 137274321021824 utils.py:1231] [26650] uptime = 168956.09113225003
+I1129 20:46:06.728842 137274321021824 utils.py:1231] [26650] examples_seen = 27289600.0
+I1129 20:46:06.728911 137274321021824 utils.py:1231] [26650] progress = 0.23667220233919167
+I1129 20:46:06.728969 137274321021824 utils.py:1231] [26650] epoch = 21.300579861953985
+I1129 20:46:06.729026 137274321021824 utils.py:1231] [26650] img/sec/core = 160.69950939259883
+I1129 20:46:06.729089 137274321021824 utils.py:1231] [26650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.89792993437945
+I1129 20:46:06.729143 137274321021824 utils.py:1231] [26650] core_hours = 46.89792993437945
+I1129 20:46:06.729224 137274321021824 train.py:125] NOTE: Steps:26650/112603 [23.7%]
+Walltime:1d22h55m (0s eval)
+ETA:6d7h16m
+Total train time:8d6h10m
+I1129 20:51:18.757074 137274321021824 utils.py:1231] [26700] l2_params = 331.7576870726583
+I1129 20:51:18.757297 137274321021824 utils.py:1231] [26700] train/loss = 2.7893969118595123
+I1129 20:51:18.757409 137274321021824 utils.py:1231] [26700] l2_grads = 1.3896313905715942
+I1129 20:51:18.757499 137274321021824 utils.py:1231] [26700] lr = 0.0009360532779920185
+I1129 20:51:18.757570 137274321021824 utils.py:1231] [26700] uptime = 169268.11993164202
+I1129 20:51:18.757638 137274321021824 utils.py:1231] [26700] examples_seen = 27340800.0
+I1129 20:51:18.757702 137274321021824 utils.py:1231] [26700] progress = 0.23711624024226707
+I1129 20:51:18.757764 137274321021824 utils.py:1231] [26700] epoch = 21.34054342642294
+I1129 20:51:18.757836 137274321021824 utils.py:1231] [26700] img/sec/core = 164.08741789144617
+I1129 20:51:18.757915 137274321021824 utils.py:1231] [26700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 46.98460460087722
+I1129 20:51:18.757975 137274321021824 utils.py:1231] [26700] core_hours = 46.98460460087722
+I1129 20:51:18.758044 137274321021824 train.py:125] NOTE: Steps:26700/112603 [23.7%]
+Walltime:1d23h1m (0s eval)
+ETA:6d7h10m
+Total train time:8d6h9m
+I1129 20:56:30.614623 137274321021824 utils.py:1231] [26750] l2_params = 331.7979691811869
+I1129 20:56:30.614852 137274321021824 utils.py:1231] [26750] train/loss = 5.263117015361786
+I1129 20:56:30.614954 137274321021824 utils.py:1231] [26750] l2_grads = 1.114357352256775
+I1129 20:56:30.615015 137274321021824 utils.py:1231] [26750] lr = 0.0009356782089837228
+I1129 20:56:30.615066 137274321021824 utils.py:1231] [26750] uptime = 169579.977428335
+I1129 20:56:30.615118 137274321021824 utils.py:1231] [26750] examples_seen = 27392000.0
+I1129 20:56:30.615167 137274321021824 utils.py:1231] [26750] progress = 0.23756027814534247
+I1129 20:56:30.615214 137274321021824 utils.py:1231] [26750] epoch = 21.3805069908919
+I1129 20:56:30.615265 137274321021824 utils.py:1231] [26750] img/sec/core = 164.17755078179843
+I1129 20:56:30.615323 137274321021824 utils.py:1231] [26750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.07123168329194
+I1129 20:56:30.615372 137274321021824 utils.py:1231] [26750] core_hours = 47.07123168329194
+I1129 20:56:30.615432 137274321021824 train.py:125] NOTE: Steps:26750/112603 [23.8%]
+Walltime:1d23h6m (0s eval)
+ETA:6d7h5m
+Total train time:8d6h9m
+I1129 21:01:42.474397 137274321021824 utils.py:1231] [26800] l2_params = 331.82994475182073
+I1129 21:01:42.474678 137274321021824 utils.py:1231] [26800] train/loss = 5.263943314552307
+I1129 21:01:42.474857 137274321021824 utils.py:1231] [26800] l2_grads = 1.0805171728134155
+I1129 21:01:42.474955 137274321021824 utils.py:1231] [26800] lr = 0.000935302118835191
+I1129 21:01:42.475017 137274321021824 utils.py:1231] [26800] uptime = 169891.83737801499
+I1129 21:01:42.475077 137274321021824 utils.py:1231] [26800] examples_seen = 27443200.0
+I1129 21:01:42.475133 137274321021824 utils.py:1231] [26800] progress = 0.2380043160484179
+I1129 21:01:42.475188 137274321021824 utils.py:1231] [26800] epoch = 21.420470555360854
+I1129 21:01:42.475247 137274321021824 utils.py:1231] [26800] img/sec/core = 164.17625941560797
+I1129 21:01:42.475310 137274321021824 utils.py:1231] [26800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.15785944709194
+I1129 21:01:42.475373 137274321021824 utils.py:1231] [26800] core_hours = 47.15785944709194
+I1129 21:01:42.475439 137274321021824 train.py:125] NOTE: Steps:26800/112603 [23.8%]
+Walltime:1d23h11m (0s eval)
+ETA:6d6h59m
+Total train time:8d6h9m
+I1129 21:06:54.316407 137274321021824 utils.py:1231] [26850] l2_params = 331.83364160112825
+I1129 21:06:54.316710 137274321021824 utils.py:1231] [26850] train/loss = 3.049359142780304
+I1129 21:06:54.316867 137274321021824 utils.py:1231] [26850] l2_grads = 1.264172911643982
+I1129 21:06:54.316950 137274321021824 utils.py:1231] [26850] lr = 0.0009349250084279001
+I1129 21:06:54.317002 137274321021824 utils.py:1231] [26850] uptime = 170203.679364222
+I1129 21:06:54.317055 137274321021824 utils.py:1231] [26850] examples_seen = 27494400.0
+I1129 21:06:54.317106 137274321021824 utils.py:1231] [26850] progress = 0.2384483539514933
+I1129 21:06:54.317154 137274321021824 utils.py:1231] [26850] epoch = 21.460434119829813
+I1129 21:06:54.317205 137274321021824 utils.py:1231] [26850] img/sec/core = 164.18571669182958
+I1129 21:06:54.317262 137274321021824 utils.py:1231] [26850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.244482221038325
+I1129 21:06:54.317314 137274321021824 utils.py:1231] [26850] core_hours = 47.244482221038325
+I1129 21:06:54.317384 137274321021824 train.py:125] NOTE: Steps:26850/112603 [23.8%]
+Walltime:1d23h16m (0s eval)
+ETA:6d6h53m
+Total train time:8d6h8m
+I1129 21:12:06.142085 137274321021824 utils.py:1231] [26900] l2_params = 331.84599513231916
+I1129 21:12:06.142451 137274321021824 utils.py:1231] [26900] train/loss = 4.417676508426666
+I1129 21:12:06.142630 137274321021824 utils.py:1231] [26900] l2_grads = 1.1248445510864258
+I1129 21:12:06.142728 137274321021824 utils.py:1231] [26900] lr = 0.0009345468786457198
+I1129 21:12:06.142819 137274321021824 utils.py:1231] [26900] uptime = 170515.50516796298
+I1129 21:12:06.142893 137274321021824 utils.py:1231] [26900] examples_seen = 27545600.0
+I1129 21:12:06.142967 137274321021824 utils.py:1231] [26900] progress = 0.2388923918545687
+I1129 21:12:06.143027 137274321021824 utils.py:1231] [26900] epoch = 21.50039768429877
+I1129 21:12:06.143086 137274321021824 utils.py:1231] [26900] img/sec/core = 164.1942372496109
+I1129 21:12:06.143148 137274321021824 utils.py:1231] [26900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.331100499855275
+I1129 21:12:06.143205 137274321021824 utils.py:1231] [26900] core_hours = 47.331100499855275
+I1129 21:12:06.143275 137274321021824 train.py:125] NOTE: Steps:26900/112603 [23.9%]
+Walltime:1d23h21m (0s eval)
+ETA:6d6h48m
+Total train time:8d6h8m
+I1129 21:17:17.970534 137274321021824 utils.py:1231] [26950] l2_params = 331.851727047968
+I1129 21:17:17.970817 137274321021824 utils.py:1231] [26950] train/loss = 4.775468051433563
+I1129 21:17:17.970985 137274321021824 utils.py:1231] [26950] l2_grads = 1.1690709590911865
+I1129 21:17:17.971062 137274321021824 utils.py:1231] [26950] lr = 0.0009341677303749086
+I1129 21:17:17.971125 137274321021824 utils.py:1231] [26950] uptime = 170827.333486758
+I1129 21:17:17.971188 137274321021824 utils.py:1231] [26950] examples_seen = 27596800.0
+I1129 21:17:17.971247 137274321021824 utils.py:1231] [26950] progress = 0.2393364297576441
+I1129 21:17:17.971305 137274321021824 utils.py:1231] [26950] epoch = 21.540361248767724
+I1129 21:17:17.971364 137274321021824 utils.py:1231] [26950] img/sec/core = 164.19291293956258
+I1129 21:17:17.971426 137274321021824 utils.py:1231] [26950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.41771947729833
+I1129 21:17:17.971484 137274321021824 utils.py:1231] [26950] core_hours = 47.41771947729833
+I1129 21:17:17.971552 137274321021824 train.py:125] NOTE: Steps:26950/112603 [23.9%]
+Walltime:1d23h27m (0s eval)
+ETA:6d6h42m
+Total train time:8d6h8m
+I1129 21:22:29.733758 137274321021824 utils.py:1231] [27000] l2_params = 331.90198916205594
+I1129 21:22:29.733959 137274321021824 utils.py:1231] [27000] train/loss = 3.1052234768867493
+I1129 21:22:29.734051 137274321021824 utils.py:1231] [27000] l2_grads = 1.3650296926498413
+I1129 21:22:29.734119 137274321021824 utils.py:1231] [27000] lr = 0.0009337875645041122
+I1129 21:22:29.734187 137274321021824 utils.py:1231] [27000] uptime = 171139.09654834203
+I1129 21:22:29.734234 137274321021824 utils.py:1231] [27000] examples_seen = 27648000.0
+I1129 21:22:29.734279 137274321021824 utils.py:1231] [27000] progress = 0.23978046766071953
+I1129 21:22:29.734324 137274321021824 utils.py:1231] [27000] epoch = 21.580324813236683
+I1129 21:22:29.734371 137274321021824 utils.py:1231] [27000] img/sec/core = 164.22728125603965
+I1129 21:22:29.734422 137274321021824 utils.py:1231] [27000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.504320327738334
+I1129 21:22:29.734467 137274321021824 utils.py:1231] [27000] core_hours = 47.504320327738334
+I1129 21:22:29.734523 137274321021824 train.py:125] NOTE: Steps:27000/112603 [24.0%]
+Walltime:1d23h32m (0s eval)
+ETA:6d6h37m
+Total train time:8d6h7m
+I1129 21:27:41.818002 137274321021824 utils.py:1231] [27050] l2_params = 331.8991970973678
+I1129 21:27:41.818334 137274321021824 utils.py:1231] [27050] train/loss = 2.915053367614746
+I1129 21:27:41.818519 137274321021824 utils.py:1231] [27050] l2_grads = 1.2515937089920044
+I1129 21:27:41.818591 137274321021824 utils.py:1231] [27050] lr = 0.0009334063819243609
+I1129 21:27:41.818648 137274321021824 utils.py:1231] [27050] uptime = 171451.18100898498
+I1129 21:27:41.818707 137274321021824 utils.py:1231] [27050] examples_seen = 27699200.0
+I1129 21:27:41.818757 137274321021824 utils.py:1231] [27050] progress = 0.24022450556379493
+I1129 21:27:41.818808 137274321021824 utils.py:1231] [27050] epoch = 21.620288377705638
+I1129 21:27:41.818861 137274321021824 utils.py:1231] [27050] img/sec/core = 164.05815238129696
+I1129 21:27:41.818931 137274321021824 utils.py:1231] [27050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.59101045569471
+I1129 21:27:41.818984 137274321021824 utils.py:1231] [27050] core_hours = 47.59101045569471
+I1129 21:27:41.819049 137274321021824 train.py:125] NOTE: Steps:27050/112603 [24.0%]
+Walltime:1d23h37m (0s eval)
+ETA:6d6h31m
+Total train time:8d6h7m
+I1129 21:32:53.877245 137274321021824 utils.py:1231] [27100] l2_params = 331.9321597773527
+I1129 21:32:53.877532 137274321021824 utils.py:1231] [27100] train/loss = 3.0379551351070404
+I1129 21:32:53.877684 137274321021824 utils.py:1231] [27100] l2_grads = 1.234205961227417
+I1129 21:32:53.877754 137274321021824 utils.py:1231] [27100] lr = 0.0009330241835290697
+I1129 21:32:53.877811 137274321021824 utils.py:1231] [27100] uptime = 171763.240173224
+I1129 21:32:53.877870 137274321021824 utils.py:1231] [27100] examples_seen = 27750400.0
+I1129 21:32:53.877929 137274321021824 utils.py:1231] [27100] progress = 0.24066854346687033
+I1129 21:32:53.877978 137274321021824 utils.py:1231] [27100] epoch = 21.660251942174597
+I1129 21:32:53.878027 137274321021824 utils.py:1231] [27100] img/sec/core = 164.07145140203568
+I1129 21:32:53.878084 137274321021824 utils.py:1231] [27100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.677693556872214
+I1129 21:32:53.878134 137274321021824 utils.py:1231] [27100] core_hours = 47.677693556872214
+I1129 21:32:53.878195 137274321021824 train.py:125] NOTE: Steps:27100/112603 [24.1%]
+Walltime:1d23h42m (0s eval)
+ETA:6d6h26m
+Total train time:8d6h7m
+I1129 21:38:05.716152 137274321021824 utils.py:1231] [27150] l2_params = 331.9567164042982
+I1129 21:38:05.716390 137274321021824 utils.py:1231] [27150] train/loss = 5.2710652351379395
+I1129 21:38:05.716500 137274321021824 utils.py:1231] [27150] l2_grads = 1.001693844795227
+I1129 21:38:05.716607 137274321021824 utils.py:1231] [27150] lr = 0.0009326409702140319
+I1129 21:38:05.716692 137274321021824 utils.py:1231] [27150] uptime = 172075.07905106398
+I1129 21:38:05.716756 137274321021824 utils.py:1231] [27150] examples_seen = 27801600.0
+I1129 21:38:05.716814 137274321021824 utils.py:1231] [27150] progress = 0.24111258136994573
+I1129 21:38:05.716875 137274321021824 utils.py:1231] [27150] epoch = 21.700215506643552
+I1129 21:38:05.716944 137274321021824 utils.py:1231] [27150] img/sec/core = 164.18735327245025
+I1129 21:38:05.717005 137274321021824 utils.py:1231] [27150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.76431546738333
+I1129 21:38:05.717060 137274321021824 utils.py:1231] [27150] core_hours = 47.76431546738333
+I1129 21:38:05.717131 137274321021824 train.py:125] NOTE: Steps:27150/112603 [24.1%]
+Walltime:1d23h47m (0s eval)
+ETA:6d6h20m
+Total train time:8d6h6m
+I1129 21:43:17.523497 137274321021824 utils.py:1231] [27200] l2_params = 331.96657143542046
+I1129 21:43:17.523713 137274321021824 utils.py:1231] [27200] train/loss = 5.210066258907318
+I1129 21:43:17.523824 137274321021824 utils.py:1231] [27200] l2_grads = 0.9989801049232483
+I1129 21:43:17.523926 137274321021824 utils.py:1231] [27200] lr = 0.000932256742877421
+I1129 21:43:17.524021 137274321021824 utils.py:1231] [27200] uptime = 172386.886374094
+I1129 21:43:17.524120 137274321021824 utils.py:1231] [27200] examples_seen = 27852800.0
+I1129 21:43:17.524235 137274321021824 utils.py:1231] [27200] progress = 0.24155661927302113
+I1129 21:43:17.524342 137274321021824 utils.py:1231] [27200] epoch = 21.74017907111251
+I1129 21:43:17.524462 137274321021824 utils.py:1231] [27200] img/sec/core = 164.2039689846231
+I1129 21:43:17.524574 137274321021824 utils.py:1231] [27200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.85092861266944
+I1129 21:43:17.524653 137274321021824 utils.py:1231] [27200] core_hours = 47.85092861266944
+I1129 21:43:17.524765 137274321021824 train.py:125] NOTE: Steps:27200/112603 [24.2%]
+Walltime:1d23h53m (0s eval)
+ETA:6d6h15m
+Total train time:8d6h6m
+I1129 21:48:29.292402 137274321021824 utils.py:1231] [27250] l2_params = 331.97483947775737
+I1129 21:48:29.292636 137274321021824 utils.py:1231] [27250] train/loss = 5.010462701320648
+I1129 21:48:29.292746 137274321021824 utils.py:1231] [27250] l2_grads = 1.0310701131820679
+I1129 21:48:29.292819 137274321021824 utils.py:1231] [27250] lr = 0.000931871502419787
+I1129 21:48:29.292885 137274321021824 utils.py:1231] [27250] uptime = 172698.65524243598
+I1129 21:48:29.292946 137274321021824 utils.py:1231] [27250] examples_seen = 27904000.0
+I1129 21:48:29.293004 137274321021824 utils.py:1231] [27250] progress = 0.24200065717609656
+I1129 21:48:29.293059 137274321021824 utils.py:1231] [27250] epoch = 21.780142635581466
+I1129 21:48:29.293115 137274321021824 utils.py:1231] [27250] img/sec/core = 164.22422248984137
+I1129 21:48:29.293181 137274321021824 utils.py:1231] [27250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 47.93753107609777
+I1129 21:48:29.293232 137274321021824 utils.py:1231] [27250] core_hours = 47.93753107609777
+I1129 21:48:29.293294 137274321021824 train.py:125] NOTE: Steps:27250/112603 [24.2%]
+Walltime:1d23h58m (0s eval)
+ETA:6d6h9m
+Total train time:8d6h6m
+I1129 21:53:41.124368 137274321021824 utils.py:1231] [27300] l2_params = 331.9773742975211
+I1129 21:53:41.124627 137274321021824 utils.py:1231] [27300] train/loss = 3.8753133714199066
+I1129 21:53:41.124757 137274321021824 utils.py:1231] [27300] l2_grads = 1.1529580354690552
+I1129 21:53:41.124823 137274321021824 utils.py:1231] [27300] lr = 0.0009314852497440542
+I1129 21:53:41.124896 137274321021824 utils.py:1231] [27300] uptime = 173010.48725777102
+I1129 21:53:41.124947 137274321021824 utils.py:1231] [27300] examples_seen = 27955200.0
+I1129 21:53:41.124994 137274321021824 utils.py:1231] [27300] progress = 0.24244469507917196
+I1129 21:53:41.125041 137274321021824 utils.py:1231] [27300] epoch = 21.82010620005042
+I1129 21:53:41.125091 137274321021824 utils.py:1231] [27300] img/sec/core = 164.1909665529208
+I1129 21:53:41.125147 137274321021824 utils.py:1231] [27300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.02415108035749
+I1129 21:53:41.125194 137274321021824 utils.py:1231] [27300] core_hours = 48.02415108035749
+I1129 21:53:41.125252 137274321021824 train.py:125] NOTE: Steps:27300/112603 [24.2%]
+Walltime:2d0h3m (0s eval)
+ETA:6d6h4m
+Total train time:8d6h5m
+I1129 21:58:52.990557 137274321021824 utils.py:1231] [27350] l2_params = 331.9890346572637
+I1129 21:58:52.990863 137274321021824 utils.py:1231] [27350] train/loss = 3.3380364775657654
+I1129 21:58:52.991048 137274321021824 utils.py:1231] [27350] l2_grads = 1.166892647743225
+I1129 21:58:52.991120 137274321021824 utils.py:1231] [27350] lr = 0.0009310979857555206
+I1129 21:58:52.991179 137274321021824 utils.py:1231] [27350] uptime = 173322.353540199
+I1129 21:58:52.991257 137274321021824 utils.py:1231] [27350] examples_seen = 28006400.0
+I1129 21:58:52.991314 137274321021824 utils.py:1231] [27350] progress = 0.24288873298224736
+I1129 21:58:52.991369 137274321021824 utils.py:1231] [27350] epoch = 21.86006976451938
+I1129 21:58:52.991425 137274321021824 utils.py:1231] [27350] img/sec/core = 164.17292565707183
+I1129 21:58:52.991502 137274321021824 utils.py:1231] [27350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.11078060325416
+I1129 21:58:52.991565 137274321021824 utils.py:1231] [27350] core_hours = 48.11078060325416
+I1129 21:58:52.991631 137274321021824 train.py:125] NOTE: Steps:27350/112603 [24.3%]
+Walltime:2d0h8m (0s eval)
+ETA:6d5h58m
+Total train time:8d6h5m
+I1129 22:04:04.827097 137274321021824 utils.py:1231] [27400] l2_params = 331.97531731567955
+I1129 22:04:04.827371 137274321021824 utils.py:1231] [27400] train/loss = 3.715266466140747
+I1129 22:04:04.827498 137274321021824 utils.py:1231] [27400] l2_grads = 1.2326489686965942
+I1129 22:04:04.827602 137274321021824 utils.py:1231] [27400] lr = 0.000930709711361852
+I1129 22:04:04.827733 137274321021824 utils.py:1231] [27400] uptime = 173634.19008742
+I1129 22:04:04.827814 137274321021824 utils.py:1231] [27400] examples_seen = 28057600.0
+I1129 22:04:04.827877 137274321021824 utils.py:1231] [27400] progress = 0.24333277088532276
+I1129 22:04:04.827953 137274321021824 utils.py:1231] [27400] epoch = 21.900033328988336
+I1129 22:04:04.828011 137274321021824 utils.py:1231] [27400] img/sec/core = 164.1885803837948
+I1129 22:04:04.828076 137274321021824 utils.py:1231] [27400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.19740186637111
+I1129 22:04:04.828132 137274321021824 utils.py:1231] [27400] core_hours = 48.19740186637111
+I1129 22:04:04.828194 137274321021824 train.py:125] NOTE: Steps:27400/112603 [24.3%]
+Walltime:2d0h13m (0s eval)
+ETA:6d5h53m
+Total train time:8d6h5m
+I1129 22:09:16.662886 137274321021824 utils.py:1231] [27450] l2_params = 331.93683169261203
+I1129 22:09:16.663108 137274321021824 utils.py:1231] [27450] train/loss = 2.937664270401001
+I1129 22:09:16.663216 137274321021824 utils.py:1231] [27450] l2_grads = 1.3064409494400024
+I1129 22:09:16.663296 137274321021824 utils.py:1231] [27450] lr = 0.000930320427473085
+I1129 22:09:16.663358 137274321021824 utils.py:1231] [27450] uptime = 173946.02571877
+I1129 22:09:16.663422 137274321021824 utils.py:1231] [27450] examples_seen = 28108800.0
+I1129 22:09:16.663481 137274321021824 utils.py:1231] [27450] progress = 0.24377680878839816
+I1129 22:09:16.663541 137274321021824 utils.py:1231] [27450] epoch = 21.939996893457295
+I1129 22:09:16.925459 137274321021824 utils.py:1231] [27450] img/sec/core = 164.1890626107903
+I1129 22:09:16.925746 137274321021824 utils.py:1231] [27450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.284022875079444
+I1129 22:09:16.925936 137274321021824 utils.py:1231] [27450] core_hours = 48.284022875079444
+I1129 22:09:16.926061 137274321021824 train.py:125] NOTE: Steps:27450/112603 [24.4%]
+Walltime:2d0h19m (0s eval)
+ETA:6d5h47m
+Total train time:8d6h4m
+I1129 22:14:30.167012 137274321021824 utils.py:1231] [27500] l2_params = 331.9424945800843
+I1129 22:14:30.167260 137274321021824 utils.py:1231] [27500] train/loss = 4.073179095983505
+I1129 22:14:30.167416 137274321021824 utils.py:1231] [27500] l2_grads = 1.0500880479812622
+I1129 22:14:30.167501 137274321021824 utils.py:1231] [27500] lr = 0.0009299301350016202
+I1129 22:14:30.167583 137274321021824 utils.py:1231] [27500] uptime = 174259.52994115
+I1129 22:14:30.167655 137274321021824 utils.py:1231] [27500] examples_seen = 28160000.0
+I1129 22:14:30.167726 137274321021824 utils.py:1231] [27500] progress = 0.2442208466914736
+I1129 22:14:30.167805 137274321021824 utils.py:1231] [27500] epoch = 21.97996045792625
+I1129 22:14:30.167942 137274321021824 utils.py:1231] [27500] img/sec/core = 163.3151847567165
+I1129 22:14:30.168030 137274321021824 utils.py:1231] [27500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.371107381296106
+I1129 22:14:30.168100 137274321021824 utils.py:1231] [27500] core_hours = 48.371107381296106
+I1129 22:14:30.168184 137274321021824 train.py:125] NOTE: Steps:27500/112603 [24.4%]
+Walltime:2d0h24m (0s eval)
+ETA:6d5h42m
+Total train time:8d6h4m
+I1129 22:14:30.168338 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:27500/112603 [24.4%]
+Walltime:2d0h24m (0s eval)
+ETA:6d5h42m
+Total train time:8d6h4m
+I1129 22:16:11.517525 137274321021824 utils.py:1231] [27500] val/acc@1 = 0.5676219706632653
+I1129 22:16:11.517948 137274321021824 utils.py:1231] [27500] val/loss = 1.8715924666244157
+I1129 22:16:11.518328 137274321021824 utils.py:1231] [27500] z/secs/eval/val = 101.34988454999984
+I1129 22:16:11.518463 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 101.34988454999984
+I1129 22:21:23.385573 137274321021824 utils.py:1231] [27550] l2_params = 331.97445409810547
+I1129 22:21:23.385958 137274321021824 utils.py:1231] [27550] train/loss = 3.146942585706711
+I1129 22:21:23.386143 137274321021824 utils.py:1231] [27550] l2_grads = 1.212480902671814
+I1129 22:21:23.386225 137274321021824 utils.py:1231] [27550] lr = 0.0009295388348622241
+I1129 22:21:23.386302 137274321021824 utils.py:1231] [27550] uptime = 174672.748658453
+I1129 22:21:23.386367 137274321021824 utils.py:1231] [27550] examples_seen = 28211200.0
+I1129 22:21:23.386420 137274321021824 utils.py:1231] [27550] progress = 0.244664884594549
+I1129 22:21:23.386491 137274321021824 utils.py:1231] [27550] epoch = 22.019924022395205
+I1129 22:21:23.386547 137274321021824 utils.py:1231] [27550] img/sec/core = 123.90532629831327
+I1129 22:21:23.386619 137274321021824 utils.py:1231] [27550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.48589035832472
+I1129 22:21:23.386673 137274321021824 utils.py:1231] [27550] core_hours = 48.48589035832472
+I1129 22:21:23.386738 137274321021824 train.py:125] NOTE: Steps:27550/112603 [24.5%]
+Walltime:2d0h31m (0s eval)
+ETA:6d5h41m
+Total train time:8d6h11m
+I1129 22:26:35.217157 137274321021824 utils.py:1231] [27600] l2_params = 331.9728739682596
+I1129 22:26:35.217442 137274321021824 utils.py:1231] [27600] train/loss = 2.7079440653324127
+I1129 22:26:35.217563 137274321021824 utils.py:1231] [27600] l2_grads = 1.3734630346298218
+I1129 22:26:35.217653 137274321021824 utils.py:1231] [27600] lr = 0.0009291465279720226
+I1129 22:26:35.217725 137274321021824 utils.py:1231] [27600] uptime = 174984.580086743
+I1129 22:26:35.217781 137274321021824 utils.py:1231] [27600] examples_seen = 28262400.0
+I1129 22:26:35.217833 137274321021824 utils.py:1231] [27600] progress = 0.2451089224976244
+I1129 22:26:35.217903 137274321021824 utils.py:1231] [27600] epoch = 22.059887586864164
+I1129 22:26:35.217957 137274321021824 utils.py:1231] [27600] img/sec/core = 164.1912756541822
+I1129 22:26:35.218029 137274321021824 utils.py:1231] [27600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.57251019951639
+I1129 22:26:35.218095 137274321021824 utils.py:1231] [27600] core_hours = 48.57251019951639
+I1129 22:26:35.218168 137274321021824 train.py:125] NOTE: Steps:27600/112603 [24.5%]
+Walltime:2d0h36m (0s eval)
+ETA:6d5h36m
+Total train time:8d6h10m
+I1129 22:31:47.055565 137274321021824 utils.py:1231] [27650] l2_params = 331.99216903244815
+I1129 22:31:47.055799 137274321021824 utils.py:1231] [27650] train/loss = 2.6446287035942078
+I1129 22:31:47.055910 137274321021824 utils.py:1231] [27650] l2_grads = 1.3681474924087524
+I1129 22:31:47.055990 137274321021824 utils.py:1231] [27650] lr = 0.0009287532152505033
+I1129 22:31:47.056050 137274321021824 utils.py:1231] [27650] uptime = 175296.41841077898
+I1129 22:31:47.056110 137274321021824 utils.py:1231] [27650] examples_seen = 28313600.0
+I1129 22:31:47.056165 137274321021824 utils.py:1231] [27650] progress = 0.2455529604006998
+I1129 22:31:47.056223 137274321021824 utils.py:1231] [27650] epoch = 22.09985115133312
+I1129 22:31:47.056280 137274321021824 utils.py:1231] [27650] img/sec/core = 164.18764485821168
+I1129 22:31:47.056344 137274321021824 utils.py:1231] [27650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.65913195619304
+I1129 22:31:47.056400 137274321021824 utils.py:1231] [27650] core_hours = 48.65913195619304
+I1129 22:31:47.056466 137274321021824 train.py:125] NOTE: Steps:27650/112603 [24.6%]
+Walltime:2d0h41m (0s eval)
+ETA:6d5h30m
+Total train time:8d6h10m
+I1129 22:36:58.886264 137274321021824 utils.py:1231] [27700] l2_params = 331.95822924359186
+I1129 22:36:58.886497 137274321021824 utils.py:1231] [27700] train/loss = 3.5863168835639954
+I1129 22:36:58.886599 137274321021824 utils.py:1231] [27700] l2_grads = 1.1446880102157593
+I1129 22:36:58.886669 137274321021824 utils.py:1231] [27700] lr = 0.0009283588976195102
+I1129 22:36:58.886732 137274321021824 utils.py:1231] [27700] uptime = 175608.24909254303
+I1129 22:36:58.886795 137274321021824 utils.py:1231] [27700] examples_seen = 28364800.0
+I1129 22:36:58.886852 137274321021824 utils.py:1231] [27700] progress = 0.24599699830377522
+I1129 22:36:58.886913 137274321021824 utils.py:1231] [27700] epoch = 22.13981471580208
+I1129 22:36:58.886972 137274321021824 utils.py:1231] [27700] img/sec/core = 164.19166872984286
+I1129 22:36:58.887032 137274321021824 utils.py:1231] [27700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.74575159001639
+I1129 22:36:58.887083 137274321021824 utils.py:1231] [27700] core_hours = 48.74575159001639
+I1129 22:36:58.887151 137274321021824 train.py:125] NOTE: Steps:27700/112603 [24.6%]
+Walltime:2d0h46m (0s eval)
+ETA:6d5h25m
+Total train time:8d6h10m
+I1129 22:42:10.710694 137274321021824 utils.py:1231] [27750] l2_params = 331.94016950943814
+I1129 22:42:10.710958 137274321021824 utils.py:1231] [27750] train/loss = 3.2639785408973694
+I1129 22:42:10.711063 137274321021824 utils.py:1231] [27750] l2_grads = 1.1919920444488525
+I1129 22:42:10.711123 137274321021824 utils.py:1231] [27750] lr = 0.000927963576003242
+I1129 22:42:10.711185 137274321021824 utils.py:1231] [27750] uptime = 175920.073547923
+I1129 22:42:10.711242 137274321021824 utils.py:1231] [27750] examples_seen = 28416000.0
+I1129 22:42:10.711298 137274321021824 utils.py:1231] [27750] progress = 0.24644103620685062
+I1129 22:42:10.711351 137274321021824 utils.py:1231] [27750] epoch = 22.179778280271034
+I1129 22:42:10.711400 137274321021824 utils.py:1231] [27750] img/sec/core = 164.1949472423841
+I1129 22:42:10.711455 137274321021824 utils.py:1231] [27750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.832369494288606
+I1129 22:42:10.711507 137274321021824 utils.py:1231] [27750] core_hours = 48.832369494288606
+I1129 22:42:10.711571 137274321021824 train.py:125] NOTE: Steps:27750/112603 [24.6%]
+Walltime:2d0h52m (0s eval)
+ETA:6d5h19m
+Total train time:8d6h9m
+I1129 22:47:22.532112 137274321021824 utils.py:1231] [27800] l2_params = 331.9224117085083
+I1129 22:47:22.532376 137274321021824 utils.py:1231] [27800] train/loss = 5.2096996903419495
+I1129 22:47:22.532508 137274321021824 utils.py:1231] [27800] l2_grads = 0.994568407535553
+I1129 22:47:22.532615 137274321021824 utils.py:1231] [27800] lr = 0.0009275672513282529
+I1129 22:47:22.532686 137274321021824 utils.py:1231] [27800] uptime = 176231.89504494902
+I1129 22:47:22.532758 137274321021824 utils.py:1231] [27800] examples_seen = 28467200.0
+I1129 22:47:22.532822 137274321021824 utils.py:1231] [27800] progress = 0.24688507410992602
+I1129 22:47:22.532897 137274321021824 utils.py:1231] [27800] epoch = 22.219741844739993
+I1129 22:47:22.532984 137274321021824 utils.py:1231] [27800] img/sec/core = 164.19650501430394
+I1129 22:47:22.533061 137274321021824 utils.py:1231] [27800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 48.91898657679583
+I1129 22:47:22.533129 137274321021824 utils.py:1231] [27800] core_hours = 48.91898657679583
+I1129 22:47:22.533212 137274321021824 train.py:125] NOTE: Steps:27800/112603 [24.7%]
+Walltime:2d0h57m (0s eval)
+ETA:6d5h14m
+Total train time:8d6h9m
+I1129 22:52:34.372359 137274321021824 utils.py:1231] [27850] l2_params = 331.9191547123045
+I1129 22:52:34.372565 137274321021824 utils.py:1231] [27850] train/loss = 2.756732761859894
+I1129 22:52:34.372667 137274321021824 utils.py:1231] [27850] l2_grads = 1.2370933294296265
+I1129 22:52:34.372738 137274321021824 utils.py:1231] [27850] lr = 0.0009271699245234455
+I1129 22:52:34.372798 137274321021824 utils.py:1231] [27850] uptime = 176543.73515983898
+I1129 22:52:34.372857 137274321021824 utils.py:1231] [27850] examples_seen = 28518400.0
+I1129 22:52:34.372919 137274321021824 utils.py:1231] [27850] progress = 0.24732911201300142
+I1129 22:52:34.372976 137274321021824 utils.py:1231] [27850] epoch = 22.259705409208948
+I1129 22:52:34.373034 137274321021824 utils.py:1231] [27850] img/sec/core = 164.18670195161627
+I1129 22:52:34.373093 137274321021824 utils.py:1231] [27850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.00560883093193
+I1129 22:52:34.373148 137274321021824 utils.py:1231] [27850] core_hours = 49.00560883093193
+I1129 22:52:34.373218 137274321021824 train.py:125] NOTE: Steps:27850/112603 [24.7%]
+Walltime:2d1h2m (0s eval)
+ETA:6d5h8m
+Total train time:8d6h9m
+I1129 22:57:46.112860 137274321021824 utils.py:1231] [27900] l2_params = 331.95091371892795
+I1129 22:57:46.113084 137274321021824 utils.py:1231] [27900] train/loss = 2.7934634387493134
+I1129 22:57:46.113188 137274321021824 utils.py:1231] [27900] l2_grads = 1.3339416980743408
+I1129 22:57:46.113256 137274321021824 utils.py:1231] [27900] lr = 0.0009267715965200722
+I1129 22:57:46.113323 137274321021824 utils.py:1231] [27900] uptime = 176855.475676761
+I1129 22:57:46.113390 137274321021824 utils.py:1231] [27900] examples_seen = 28569600.0
+I1129 22:57:46.113445 137274321021824 utils.py:1231] [27900] progress = 0.24777314991607682
+I1129 22:57:46.113500 137274321021824 utils.py:1231] [27900] epoch = 22.299668973677903
+I1129 22:57:46.113556 137274321021824 utils.py:1231] [27900] img/sec/core = 164.23915795587982
+I1129 22:57:46.113618 137274321021824 utils.py:1231] [27900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.09220341896582
+I1129 22:57:46.113672 137274321021824 utils.py:1231] [27900] core_hours = 49.09220341896582
+I1129 22:57:46.113732 137274321021824 train.py:125] NOTE: Steps:27900/112603 [24.8%]
+Walltime:2d1h7m (0s eval)
+ETA:6d5h3m
+Total train time:8d6h8m
+I1129 23:02:57.902762 137274321021824 utils.py:1231] [27950] l2_params = 331.93956639491165
+I1129 23:02:57.903024 137274321021824 utils.py:1231] [27950] train/loss = 4.825970113277435
+I1129 23:02:57.903137 137274321021824 utils.py:1231] [27950] l2_grads = 1.013520359992981
+I1129 23:02:57.903210 137274321021824 utils.py:1231] [27950] lr = 0.000926372268251733
+I1129 23:02:57.903277 137274321021824 utils.py:1231] [27950] uptime = 177167.26563177002
+I1129 23:02:57.903336 137274321021824 utils.py:1231] [27950] examples_seen = 28620800.0
+I1129 23:02:57.903395 137274321021824 utils.py:1231] [27950] progress = 0.24821718781915225
+I1129 23:02:57.903453 137274321021824 utils.py:1231] [27950] epoch = 22.339632538146862
+I1129 23:02:57.903510 137274321021824 utils.py:1231] [27950] img/sec/core = 164.2131158411463
+I1129 23:02:57.903575 137274321021824 utils.py:1231] [27950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.17881173980166
+I1129 23:02:57.903639 137274321021824 utils.py:1231] [27950] core_hours = 49.17881173980166
+I1129 23:02:57.903709 137274321021824 train.py:125] NOTE: Steps:27950/112603 [24.8%]
+Walltime:2d1h12m (0s eval)
+ETA:6d4h57m
+Total train time:8d6h8m
+I1129 23:08:09.703378 137274321021824 utils.py:1231] [28000] l2_params = 331.95105444391896
+I1129 23:08:09.703580 137274321021824 utils.py:1231] [28000] train/loss = 5.322449207305908
+I1129 23:08:09.703687 137274321021824 utils.py:1231] [28000] l2_grads = 1.3331650495529175
+I1129 23:08:09.703759 137274321021824 utils.py:1231] [28000] lr = 0.0009259719406543708
+I1129 23:08:09.703818 137274321021824 utils.py:1231] [28000] uptime = 177479.066179929
+I1129 23:08:09.703890 137274321021824 utils.py:1231] [28000] examples_seen = 28672000.0
+I1129 23:08:09.703956 137274321021824 utils.py:1231] [28000] progress = 0.24866122572222765
+I1129 23:08:09.704033 137274321021824 utils.py:1231] [28000] epoch = 22.379596102615817
+I1129 23:08:09.704095 137274321021824 utils.py:1231] [28000] img/sec/core = 164.20753684465004
+I1129 23:08:09.704163 137274321021824 utils.py:1231] [28000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.26542300317916
+I1129 23:08:09.704224 137274321021824 utils.py:1231] [28000] core_hours = 49.26542300317916
+I1129 23:08:09.704293 137274321021824 train.py:125] NOTE: Steps:28000/112603 [24.9%]
+Walltime:2d1h17m (0s eval)
+ETA:6d4h52m
+Total train time:8d6h8m
+I1129 23:13:21.860252 137274321021824 utils.py:1231] [28050] l2_params = 331.95829700023916
+I1129 23:13:21.860467 137274321021824 utils.py:1231] [28050] train/loss = 2.8331134617328644
+I1129 23:13:21.860570 137274321021824 utils.py:1231] [28050] l2_grads = 1.378697395324707
+I1129 23:13:21.860633 137274321021824 utils.py:1231] [28050] lr = 0.0009255706146662708
+I1129 23:13:21.860684 137274321021824 utils.py:1231] [28050] uptime = 177791.22304585198
+I1129 23:13:21.860735 137274321021824 utils.py:1231] [28050] examples_seen = 28723200.0
+I1129 23:13:21.860785 137274321021824 utils.py:1231] [28050] progress = 0.24910526362530305
+I1129 23:13:21.860832 137274321021824 utils.py:1231] [28050] epoch = 22.419559667084776
+I1129 23:13:21.860887 137274321021824 utils.py:1231] [28050] img/sec/core = 164.02009883271378
+I1129 23:13:21.860943 137274321021824 utils.py:1231] [28050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.352133243713325
+I1129 23:13:21.860991 137274321021824 utils.py:1231] [28050] core_hours = 49.352133243713325
+I1129 23:13:21.861049 137274321021824 train.py:125] NOTE: Steps:28050/112603 [24.9%]
+Walltime:2d1h23m (0s eval)
+ETA:6d4h46m
+Total train time:8d6h7m
+I1129 23:18:33.651073 137274321021824 utils.py:1231] [28100] l2_params = 331.9274826850187
+I1129 23:18:33.651340 137274321021824 utils.py:1231] [28100] train/loss = 2.8712803423404694
+I1129 23:18:33.651568 137274321021824 utils.py:1231] [28100] l2_grads = 1.262227177619934
+I1129 23:18:33.651716 137274321021824 utils.py:1231] [28100] lr = 0.0009251682912280589
+I1129 23:18:33.651819 137274321021824 utils.py:1231] [28100] uptime = 178103.01416680298
+I1129 23:18:33.651952 137274321021824 utils.py:1231] [28100] examples_seen = 28774400.0
+I1129 23:18:33.652059 137274321021824 utils.py:1231] [28100] progress = 0.24954930152837845
+I1129 23:18:33.652177 137274321021824 utils.py:1231] [28100] epoch = 22.45952323155373
+I1129 23:18:33.652276 137274321021824 utils.py:1231] [28100] img/sec/core = 164.21250176667306
+I1129 23:18:33.652358 137274321021824 utils.py:1231] [28100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.43874188842194
+I1129 23:18:33.652435 137274321021824 utils.py:1231] [28100] core_hours = 49.43874188842194
+I1129 23:18:33.652547 137274321021824 train.py:125] NOTE: Steps:28100/112603 [25.0%]
+Walltime:2d1h28m (0s eval)
+ETA:6d4h41m
+Total train time:8d6h7m
+I1129 23:23:45.438067 137274321021824 utils.py:1231] [28150] l2_params = 331.97331568309136
+I1129 23:23:45.438275 137274321021824 utils.py:1231] [28150] train/loss = 2.6816800236701965
+I1129 23:23:45.438392 137274321021824 utils.py:1231] [28150] l2_grads = 1.3775222301483154
+I1129 23:23:45.438461 137274321021824 utils.py:1231] [28150] lr = 0.0009247649712826991
+I1129 23:23:45.438514 137274321021824 utils.py:1231] [28150] uptime = 178414.80087627698
+I1129 23:23:45.438567 137274321021824 utils.py:1231] [28150] examples_seen = 28825600.0
+I1129 23:23:45.438617 137274321021824 utils.py:1231] [28150] progress = 0.24999333943145388
+I1129 23:23:45.438667 137274321021824 utils.py:1231] [28150] epoch = 22.499486796022687
+I1129 23:23:45.438718 137274321021824 utils.py:1231] [28150] img/sec/core = 164.2148252129742
+I1129 23:23:45.438775 137274321021824 utils.py:1231] [28150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.52534930772027
+I1129 23:23:45.438827 137274321021824 utils.py:1231] [28150] core_hours = 49.52534930772027
+I1129 23:23:45.438894 137274321021824 train.py:125] NOTE: Steps:28150/112603 [25.0%]
+Walltime:2d1h33m (0s eval)
+ETA:6d4h35m
+Total train time:8d6h7m
+I1129 23:28:57.230708 137274321021824 utils.py:1231] [28200] l2_params = 331.9452229557403
+I1129 23:28:57.230952 137274321021824 utils.py:1231] [28200] train/loss = 3.2660330832004547
+I1129 23:28:57.231078 137274321021824 utils.py:1231] [28200] l2_grads = 1.174411416053772
+I1129 23:28:57.231166 137274321021824 utils.py:1231] [28200] lr = 0.0009243606557754892
+I1129 23:28:57.231244 137274321021824 utils.py:1231] [28200] uptime = 178726.593605223
+I1129 23:28:57.231314 137274321021824 utils.py:1231] [28200] examples_seen = 28876800.0
+I1129 23:28:57.231385 137274321021824 utils.py:1231] [28200] progress = 0.25043737733452925
+I1129 23:28:57.231441 137274321021824 utils.py:1231] [28200] epoch = 22.539450360491646
+I1129 23:28:57.231499 137274321021824 utils.py:1231] [28200] img/sec/core = 164.21165488071432
+I1129 23:28:57.231560 137274321021824 utils.py:1231] [28200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.61195839909416
+I1129 23:28:57.231623 137274321021824 utils.py:1231] [28200] core_hours = 49.61195839909416
+I1129 23:28:57.231690 137274321021824 train.py:125] NOTE: Steps:28200/112603 [25.0%]
+Walltime:2d1h38m (0s eval)
+ETA:6d4h29m
+Total train time:8d6h6m
+I1129 23:34:09.133706 137274321021824 utils.py:1231] [28250] l2_params = 331.94874239457243
+I1129 23:34:09.133915 137274321021824 utils.py:1231] [28250] train/loss = 4.531778335571289
+I1129 23:34:09.134024 137274321021824 utils.py:1231] [28250] l2_grads = 1.037584900856018
+I1129 23:34:09.134106 137274321021824 utils.py:1231] [28250] lr = 0.0009239553456540635
+I1129 23:34:09.134193 137274321021824 utils.py:1231] [28250] uptime = 179038.49654833
+I1129 23:34:09.134302 137274321021824 utils.py:1231] [28250] examples_seen = 28928000.0
+I1129 23:34:09.134392 137274321021824 utils.py:1231] [28250] progress = 0.25088141523760465
+I1129 23:34:09.134483 137274321021824 utils.py:1231] [28250] epoch = 22.5794139249606
+I1129 23:34:09.134577 137274321021824 utils.py:1231] [28250] img/sec/core = 164.15362897822658
+I1129 23:34:09.134676 137274321021824 utils.py:1231] [28250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.69859810551277
+I1129 23:34:09.134761 137274321021824 utils.py:1231] [28250] core_hours = 49.69859810551277
+I1129 23:34:09.134834 137274321021824 train.py:125] NOTE: Steps:28250/112603 [25.1%]
+Walltime:2d1h43m (0s eval)
+ETA:6d4h24m
+Total train time:8d6h6m
+I1129 23:39:20.924651 137274321021824 utils.py:1231] [28300] l2_params = 331.95963266098323
+I1129 23:39:20.924899 137274321021824 utils.py:1231] [28300] train/loss = 3.3222171664237976
+I1129 23:39:20.925002 137274321021824 utils.py:1231] [28300] l2_grads = 1.3115736246109009
+I1129 23:39:20.925063 137274321021824 utils.py:1231] [28300] lr = 0.0009235490418683851
+I1129 23:39:20.925115 137274321021824 utils.py:1231] [28300] uptime = 179350.287477176
+I1129 23:39:20.925168 137274321021824 utils.py:1231] [28300] examples_seen = 28979200.0
+I1129 23:39:20.925217 137274321021824 utils.py:1231] [28300] progress = 0.2513254531406801
+I1129 23:39:20.925264 137274321021824 utils.py:1231] [28300] epoch = 22.61937748942956
+I1129 23:39:20.925313 137274321021824 utils.py:1231] [28300] img/sec/core = 164.2126029435745
+I1129 23:39:20.925368 137274321021824 utils.py:1231] [28300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.78520669685888
+I1129 23:39:20.925416 137274321021824 utils.py:1231] [28300] core_hours = 49.78520669685888
+I1129 23:39:20.925480 137274321021824 train.py:125] NOTE: Steps:28300/112603 [25.1%]
+Walltime:2d1h49m (0s eval)
+ETA:6d4h18m
+Total train time:8d6h6m
+I1129 23:44:32.715950 137274321021824 utils.py:1231] [28350] l2_params = 331.92734947120096
+I1129 23:44:32.716201 137274321021824 utils.py:1231] [28350] train/loss = 2.883903294801712
+I1129 23:44:32.716324 137274321021824 utils.py:1231] [28350] l2_grads = 1.3512505292892456
+I1129 23:44:32.716397 137274321021824 utils.py:1231] [28350] lr = 0.0009231417453707457
+I1129 23:44:32.716460 137274321021824 utils.py:1231] [28350] uptime = 179662.078815214
+I1129 23:44:32.716516 137274321021824 utils.py:1231] [28350] examples_seen = 29030400.0
+I1129 23:44:32.716568 137274321021824 utils.py:1231] [28350] progress = 0.2517694910437555
+I1129 23:44:32.716624 137274321021824 utils.py:1231] [28350] epoch = 22.659341053898515
+I1129 23:44:32.716674 137274321021824 utils.py:1231] [28350] img/sec/core = 164.21238743252545
+I1129 23:44:32.716735 137274321021824 utils.py:1231] [28350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.87181540186943
+I1129 23:44:32.716784 137274321021824 utils.py:1231] [28350] core_hours = 49.87181540186943
+I1129 23:44:32.716845 137274321021824 train.py:125] NOTE: Steps:28350/112603 [25.2%]
+Walltime:2d1h54m (0s eval)
+ETA:6d4h13m
+Total train time:8d6h5m
+I1129 23:49:44.444261 137274321021824 utils.py:1231] [28400] l2_params = 331.9336152306393
+I1129 23:49:44.444558 137274321021824 utils.py:1231] [28400] train/loss = 4.2980717420578
+I1129 23:49:44.444738 137274321021824 utils.py:1231] [28400] l2_grads = 1.1025290489196777
+I1129 23:49:44.444827 137274321021824 utils.py:1231] [28400] lr = 0.0009227334571157655
+I1129 23:49:44.444907 137274321021824 utils.py:1231] [28400] uptime = 179973.80726515298
+I1129 23:49:44.444985 137274321021824 utils.py:1231] [28400] examples_seen = 29081600.0
+I1129 23:49:44.445061 137274321021824 utils.py:1231] [28400] progress = 0.2522135289468309
+I1129 23:49:44.445137 137274321021824 utils.py:1231] [28400] epoch = 22.699304618367474
+I1129 23:49:44.445208 137274321021824 utils.py:1231] [28400] img/sec/core = 164.2455156403596
+I1129 23:49:44.445281 137274321021824 utils.py:1231] [28400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 49.9584066379636
+I1129 23:49:44.445352 137274321021824 utils.py:1231] [28400] core_hours = 49.9584066379636
+I1129 23:49:44.445444 137274321021824 train.py:125] NOTE: Steps:28400/112603 [25.2%]
+Walltime:2d1h59m (0s eval)
+ETA:6d4h7m
+Total train time:8d6h5m
+I1129 23:54:56.214496 137274321021824 utils.py:1231] [28450] l2_params = 331.909417540633
+I1129 23:54:56.214700 137274321021824 utils.py:1231] [28450] train/loss = 3.1018678843975067
+I1129 23:54:56.214793 137274321021824 utils.py:1231] [28450] l2_grads = 1.2552064657211304
+I1129 23:54:56.214866 137274321021824 utils.py:1231] [28450] lr = 0.0009223241780603879
+I1129 23:54:56.214934 137274321021824 utils.py:1231] [28450] uptime = 180285.57729185402
+I1129 23:54:56.214991 137274321021824 utils.py:1231] [28450] examples_seen = 29132800.0
+I1129 23:54:56.215046 137274321021824 utils.py:1231] [28450] progress = 0.2526575668499063
+I1129 23:54:56.215097 137274321021824 utils.py:1231] [28450] epoch = 22.73926818283643
+I1129 23:54:56.215149 137274321021824 utils.py:1231] [28450] img/sec/core = 164.22361232659992
+I1129 23:54:56.215208 137274321021824 utils.py:1231] [28450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.045009423158334
+I1129 23:54:56.215274 137274321021824 utils.py:1231] [28450] core_hours = 50.045009423158334
+I1129 23:54:56.215334 137274321021824 train.py:125] NOTE: Steps:28450/112603 [25.3%]
+Walltime:2d2h4m (0s eval)
+ETA:6d4h2m
+Total train time:8d6h5m
+I1130 00:00:08.001071 137274321021824 utils.py:1231] [28500] l2_params = 331.8648523771048
+I1130 00:00:08.001320 137274321021824 utils.py:1231] [28500] train/loss = 4.762643814086914
+I1130 00:00:08.001479 137274321021824 utils.py:1231] [28500] l2_grads = 1.0561015605926514
+I1130 00:00:08.001584 137274321021824 utils.py:1231] [28500] lr = 0.0009219139091638795
+I1130 00:00:08.001664 137274321021824 utils.py:1231] [28500] uptime = 180597.36402231402
+I1130 00:00:08.001769 137274321021824 utils.py:1231] [28500] examples_seen = 29184000.0
+I1130 00:00:08.001857 137274321021824 utils.py:1231] [28500] progress = 0.2531016047529817
+I1130 00:00:08.001950 137274321021824 utils.py:1231] [28500] epoch = 22.779231747305385
+I1130 00:00:08.002029 137274321021824 utils.py:1231] [28500] img/sec/core = 164.21481415985832
+I1130 00:00:08.002110 137274321021824 utils.py:1231] [28500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.13161684828611
+I1130 00:00:08.002188 137274321021824 utils.py:1231] [28500] core_hours = 50.13161684828611
+I1130 00:00:08.002282 137274321021824 train.py:125] NOTE: Steps:28500/112603 [25.3%]
+Walltime:2d2h9m (0s eval)
+ETA:6d3h56m
+Total train time:8d6h4m
+I1130 00:05:19.777432 137274321021824 utils.py:1231] [28550] l2_params = 331.90272355960616
+I1130 00:05:19.777688 137274321021824 utils.py:1231] [28550] train/loss = 3.3123176395893097
+I1130 00:05:19.777822 137274321021824 utils.py:1231] [28550] l2_grads = 1.2202562093734741
+I1130 00:05:19.777904 137274321021824 utils.py:1231] [28550] lr = 0.0009215026513878257
+I1130 00:05:19.777968 137274321021824 utils.py:1231] [28550] uptime = 180909.140328918
+I1130 00:05:19.778030 137274321021824 utils.py:1231] [28550] examples_seen = 29235200.0
+I1130 00:05:19.778089 137274321021824 utils.py:1231] [28550] progress = 0.2535456426560571
+I1130 00:05:19.778153 137274321021824 utils.py:1231] [28550] epoch = 22.819195311774344
+I1130 00:05:19.778221 137274321021824 utils.py:1231] [28550] img/sec/core = 164.22030447948774
+I1130 00:05:19.778328 137274321021824 utils.py:1231] [28550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.21822137789833
+I1130 00:05:19.778401 137274321021824 utils.py:1231] [28550] core_hours = 50.21822137789833
+I1130 00:05:19.778477 137274321021824 train.py:125] NOTE: Steps:28550/112603 [25.4%]
+Walltime:2d2h15m (0s eval)
+ETA:6d3h51m
+Total train time:8d6h4m
+I1130 00:10:31.561345 137274321021824 utils.py:1231] [28600] l2_params = 331.9145445955862
+I1130 00:10:31.561633 137274321021824 utils.py:1231] [28600] train/loss = 2.871606171131134
+I1130 00:10:31.561874 137274321021824 utils.py:1231] [28600] l2_grads = 1.3388524055480957
+I1130 00:10:31.561980 137274321021824 utils.py:1231] [28600] lr = 0.0009210904056961307
+I1130 00:10:31.562057 137274321021824 utils.py:1231] [28600] uptime = 181220.92441556003
+I1130 00:10:31.562128 137274321021824 utils.py:1231] [28600] examples_seen = 29286400.0
+I1130 00:10:31.562210 137274321021824 utils.py:1231] [28600] progress = 0.2539896805591325
+I1130 00:10:31.562288 137274321021824 utils.py:1231] [28600] epoch = 22.8591588762433
+I1130 00:10:31.562366 137274321021824 utils.py:1231] [28600] img/sec/core = 164.21620664299402
+I1130 00:10:31.562450 137274321021824 utils.py:1231] [28600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.30482806863222
+I1130 00:10:31.562524 137274321021824 utils.py:1231] [28600] core_hours = 50.30482806863222
+I1130 00:10:31.562607 137274321021824 train.py:125] NOTE: Steps:28600/112603 [25.4%]
+Walltime:2d2h20m (0s eval)
+ETA:6d3h45m
+Total train time:8d6h4m
+I1130 00:15:43.449843 137274321021824 utils.py:1231] [28650] l2_params = 331.88214644121007
+I1130 00:15:43.450086 137274321021824 utils.py:1231] [28650] train/loss = 2.872728705406189
+I1130 00:15:43.450222 137274321021824 utils.py:1231] [28650] l2_grads = 1.3280428647994995
+I1130 00:15:43.450321 137274321021824 utils.py:1231] [28650] lr = 0.0009206771730550133
+I1130 00:15:43.450397 137274321021824 utils.py:1231] [28650] uptime = 181532.812754179
+I1130 00:15:43.450475 137274321021824 utils.py:1231] [28650] examples_seen = 29337600.0
+I1130 00:15:43.450551 137274321021824 utils.py:1231] [28650] progress = 0.2544337184622079
+I1130 00:15:43.450620 137274321021824 utils.py:1231] [28650] epoch = 22.899122440712258
+I1130 00:15:43.450699 137274321021824 utils.py:1231] [28650] img/sec/core = 164.16131563849717
+I1130 00:15:43.450791 137274321021824 utils.py:1231] [28650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.39146371824861
+I1130 00:15:43.450871 137274321021824 utils.py:1231] [28650] core_hours = 50.39146371824861
+I1130 00:15:43.450992 137274321021824 train.py:125] NOTE: Steps:28650/112603 [25.4%]
+Walltime:2d2h25m (0s eval)
+ETA:6d3h40m
+Total train time:8d6h4m
+I1130 00:20:55.234526 137274321021824 utils.py:1231] [28700] l2_params = 331.8505065326049
+I1130 00:20:55.234819 137274321021824 utils.py:1231] [28700] train/loss = 4.627461135387421
+I1130 00:20:55.235046 137274321021824 utils.py:1231] [28700] l2_grads = 1.0441447496414185
+I1130 00:20:55.235189 137274321021824 utils.py:1231] [28700] lr = 0.0009202629544330066
+I1130 00:20:55.235281 137274321021824 utils.py:1231] [28700] uptime = 181844.59763630503
+I1130 00:20:55.235391 137274321021824 utils.py:1231] [28700] examples_seen = 29388800.0
+I1130 00:20:55.235475 137274321021824 utils.py:1231] [28700] progress = 0.2548777563652833
+I1130 00:20:55.235553 137274321021824 utils.py:1231] [28700] epoch = 22.939086005181213
+I1130 00:20:55.235639 137274321021824 utils.py:1231] [28700] img/sec/core = 164.2157876638457
+I1130 00:20:55.235721 137274321021824 utils.py:1231] [28700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.478070629950274
+I1130 00:20:55.235790 137274321021824 utils.py:1231] [28700] core_hours = 50.478070629950274
+I1130 00:20:55.235877 137274321021824 train.py:125] NOTE: Steps:28700/112603 [25.5%]
+Walltime:2d2h30m (0s eval)
+ETA:6d3h34m
+Total train time:8d6h3m
+I1130 00:26:07.020191 137274321021824 utils.py:1231] [28750] l2_params = 331.8537788502102
+I1130 00:26:07.020434 137274321021824 utils.py:1231] [28750] train/loss = 2.792011708021164
+I1130 00:26:07.020543 137274321021824 utils.py:1231] [28750] l2_grads = 1.3370131254196167
+I1130 00:26:07.020617 137274321021824 utils.py:1231] [28750] lr = 0.0009198477508009541
+I1130 00:26:07.020684 137274321021824 utils.py:1231] [28750] uptime = 182156.38304537302
+I1130 00:26:07.020745 137274321021824 utils.py:1231] [28750] examples_seen = 29440000.0
+I1130 00:26:07.020802 137274321021824 utils.py:1231] [28750] progress = 0.25532179426835877
+I1130 00:26:07.020859 137274321021824 utils.py:1231] [28750] epoch = 22.979049569650172
+I1130 00:26:07.020925 137274321021824 utils.py:1231] [28750] img/sec/core = 164.21551012618087
+I1130 00:26:07.020989 137274321021824 utils.py:1231] [28750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.564677688024716
+I1130 00:26:07.021046 137274321021824 utils.py:1231] [28750] core_hours = 50.564677688024716
+I1130 00:26:07.021113 137274321021824 train.py:125] NOTE: Steps:28750/112603 [25.5%]
+Walltime:2d2h35m (0s eval)
+ETA:6d3h29m
+Total train time:8d6h3m
+I1130 00:31:18.792274 137274321021824 utils.py:1231] [28800] l2_params = 331.89580885886437
+I1130 00:31:18.792522 137274321021824 utils.py:1231] [28800] train/loss = 2.879154086112976
+I1130 00:31:18.792673 137274321021824 utils.py:1231] [28800] l2_grads = 1.2924660444259644
+I1130 00:31:18.792784 137274321021824 utils.py:1231] [28800] lr = 0.0009194315631320064
+I1130 00:31:18.792873 137274321021824 utils.py:1231] [28800] uptime = 182468.155230314
+I1130 00:31:18.792959 137274321021824 utils.py:1231] [28800] examples_seen = 29491200.0
+I1130 00:31:18.793046 137274321021824 utils.py:1231] [28800] progress = 0.25576583217143417
+I1130 00:31:18.793133 137274321021824 utils.py:1231] [28800] epoch = 23.019013134119128
+I1130 00:31:18.793206 137274321021824 utils.py:1231] [28800] img/sec/core = 164.2224754902089
+I1130 00:31:18.793283 137274321021824 utils.py:1231] [28800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.65128107273054
+I1130 00:31:18.793357 137274321021824 utils.py:1231] [28800] core_hours = 50.65128107273054
+I1130 00:31:18.793447 137274321021824 train.py:125] NOTE: Steps:28800/112603 [25.6%]
+Walltime:2d2h41m (0s eval)
+ETA:6d3h23m
+Total train time:8d6h3m
+I1130 00:36:30.572001 137274321021824 utils.py:1231] [28850] l2_params = 331.8948492479325
+I1130 00:36:30.572191 137274321021824 utils.py:1231] [28850] train/loss = 5.134329617023468
+I1130 00:36:30.572296 137274321021824 utils.py:1231] [28850] l2_grads = 1.122025728225708
+I1130 00:36:30.572377 137274321021824 utils.py:1231] [28850] lr = 0.0009190143924016231
+I1130 00:36:30.572426 137274321021824 utils.py:1231] [28850] uptime = 182779.934788367
+I1130 00:36:30.572475 137274321021824 utils.py:1231] [28850] examples_seen = 29542400.0
+I1130 00:36:30.572521 137274321021824 utils.py:1231] [28850] progress = 0.25620987007450957
+I1130 00:36:30.572566 137274321021824 utils.py:1231] [28850] epoch = 23.058976698588083
+I1130 00:36:30.572611 137274321021824 utils.py:1231] [28850] img/sec/core = 164.21859187860935
+I1130 00:36:30.572663 137274321021824 utils.py:1231] [28850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.73788650552305
+I1130 00:36:30.572709 137274321021824 utils.py:1231] [28850] core_hours = 50.73788650552305
+I1130 00:36:30.572764 137274321021824 train.py:125] NOTE: Steps:28850/112603 [25.6%]
+Walltime:2d2h46m (0s eval)
+ETA:6d3h18m
+Total train time:8d6h2m
+I1130 00:41:42.354713 137274321021824 utils.py:1231] [28900] l2_params = 331.9076211665925
+I1130 00:41:42.354922 137274321021824 utils.py:1231] [28900] train/loss = 2.764690637588501
+I1130 00:41:42.355029 137274321021824 utils.py:1231] [28900] l2_grads = 1.4068748950958252
+I1130 00:41:42.355104 137274321021824 utils.py:1231] [28900] lr = 0.0009185962395875661
+I1130 00:41:42.355155 137274321021824 utils.py:1231] [28900] uptime = 183091.717516859
+I1130 00:41:42.355205 137274321021824 utils.py:1231] [28900] examples_seen = 29593600.0
+I1130 00:41:42.355257 137274321021824 utils.py:1231] [28900] progress = 0.25665390797758497
+I1130 00:41:42.355305 137274321021824 utils.py:1231] [28900] epoch = 23.09894026305704
+I1130 00:41:42.355354 137274321021824 utils.py:1231] [28900] img/sec/core = 164.21692198166508
+I1130 00:41:42.355408 137274321021824 utils.py:1231] [28900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.82449281899304
+I1130 00:41:42.355456 137274321021824 utils.py:1231] [28900] core_hours = 50.82449281899304
+I1130 00:41:42.355518 137274321021824 train.py:125] NOTE: Steps:28900/112603 [25.7%]
+Walltime:2d2h51m (0s eval)
+ETA:6d3h12m
+Total train time:8d6h2m
+I1130 00:46:54.132405 137274321021824 utils.py:1231] [28950] l2_params = 331.89873619641725
+I1130 00:46:54.132665 137274321021824 utils.py:1231] [28950] train/loss = 5.011765420436859
+I1130 00:46:54.132781 137274321021824 utils.py:1231] [28950] l2_grads = 1.0581880807876587
+I1130 00:46:54.132849 137274321021824 utils.py:1231] [28950] lr = 0.0009181771056699002
+I1130 00:46:54.132913 137274321021824 utils.py:1231] [28950] uptime = 183403.49527408503
+I1130 00:46:54.132965 137274321021824 utils.py:1231] [28950] examples_seen = 29644800.0
+I1130 00:46:54.133015 137274321021824 utils.py:1231] [28950] progress = 0.2570979458806604
+I1130 00:46:54.133064 137274321021824 utils.py:1231] [28950] epoch = 23.138903827525997
+I1130 00:46:54.133115 137274321021824 utils.py:1231] [28950] img/sec/core = 164.21954040448907
+I1130 00:46:54.133172 137274321021824 utils.py:1231] [28950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.911097751555836
+I1130 00:46:54.133231 137274321021824 utils.py:1231] [28950] core_hours = 50.911097751555836
+I1130 00:46:54.133301 137274321021824 train.py:125] NOTE: Steps:28950/112603 [25.7%]
+Walltime:2d2h56m (0s eval)
+ETA:6d3h7m
+Total train time:8d6h2m
+I1130 00:52:05.901631 137274321021824 utils.py:1231] [29000] l2_params = 331.89938100754296
+I1130 00:52:05.901829 137274321021824 utils.py:1231] [29000] train/loss = 2.809863567352295
+I1130 00:52:05.901934 137274321021824 utils.py:1231] [29000] l2_grads = 1.3595685958862305
+I1130 00:52:05.902020 137274321021824 utils.py:1231] [29000] lr = 0.0009177569916309891
+I1130 00:52:05.902081 137274321021824 utils.py:1231] [29000] uptime = 183715.264442674
+I1130 00:52:05.902141 137274321021824 utils.py:1231] [29000] examples_seen = 29696000.0
+I1130 00:52:05.902197 137274321021824 utils.py:1231] [29000] progress = 0.2575419837837358
+I1130 00:52:05.902248 137274321021824 utils.py:1231] [29000] epoch = 23.178867391994956
+I1130 00:52:05.902303 137274321021824 utils.py:1231] [29000] img/sec/core = 164.22406433490454
+I1130 00:52:05.902373 137274321021824 utils.py:1231] [29000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 50.99770029838611
+I1130 00:52:05.902428 137274321021824 utils.py:1231] [29000] core_hours = 50.99770029838611
+I1130 00:52:05.902504 137274321021824 train.py:125] NOTE: Steps:29000/112603 [25.8%]
+Walltime:2d3h1m (0s eval)
+ETA:6d3h1m
+Total train time:8d6h1m
+I1130 00:57:18.095462 137274321021824 utils.py:1231] [29050] l2_params = 331.91221309172215
+I1130 00:57:18.095671 137274321021824 utils.py:1231] [29050] train/loss = 3.5712770819664
+I1130 00:57:18.095781 137274321021824 utils.py:1231] [29050] l2_grads = 1.0778497457504272
+I1130 00:57:18.095853 137274321021824 utils.py:1231] [29050] lr = 0.0009173358984554937
+I1130 00:57:18.095917 137274321021824 utils.py:1231] [29050] uptime = 184027.458278574
+I1130 00:57:18.095978 137274321021824 utils.py:1231] [29050] examples_seen = 29747200.0
+I1130 00:57:18.096036 137274321021824 utils.py:1231] [29050] progress = 0.2579860216868112
+I1130 00:57:18.096094 137274321021824 utils.py:1231] [29050] epoch = 23.21883095646391
+I1130 00:57:18.096152 137274321021824 utils.py:1231] [29050] img/sec/core = 164.0006755815692
+I1130 00:57:18.096215 137274321021824 utils.py:1231] [29050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.08442080835833
+I1130 00:57:18.096271 137274321021824 utils.py:1231] [29050] core_hours = 51.08442080835833
+I1130 00:57:18.096336 137274321021824 train.py:125] NOTE: Steps:29050/112603 [25.8%]
+Walltime:2d3h7m (0s eval)
+ETA:6d2h56m
+Total train time:8d6h1m
+I1130 01:02:29.866776 137274321021824 utils.py:1231] [29100] l2_params = 331.95392682936375
+I1130 01:02:29.866994 137274321021824 utils.py:1231] [29100] train/loss = 2.8138042092323303
+I1130 01:02:29.867093 137274321021824 utils.py:1231] [29100] l2_grads = 1.403037190437317
+I1130 01:02:29.867170 137274321021824 utils.py:1231] [29100] lr = 0.0009169138271303703
+I1130 01:02:29.867228 137274321021824 utils.py:1231] [29100] uptime = 184339.22958971298
+I1130 01:02:29.867286 137274321021824 utils.py:1231] [29100] examples_seen = 29798400.0
+I1130 01:02:29.867340 137274321021824 utils.py:1231] [29100] progress = 0.2584300595898866
+I1130 01:02:29.867398 137274321021824 utils.py:1231] [29100] epoch = 23.258794520932867
+I1130 01:02:29.867452 137274321021824 utils.py:1231] [29100] img/sec/core = 164.22293575682707
+I1130 01:02:29.867510 137274321021824 utils.py:1231] [29100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.17102395034138
+I1130 01:02:29.867587 137274321021824 utils.py:1231] [29100] core_hours = 51.17102395034138
+I1130 01:02:29.867653 137274321021824 train.py:125] NOTE: Steps:29100/112603 [25.8%]
+Walltime:2d3h12m (0s eval)
+ETA:6d2h50m
+Total train time:8d6h1m
+I1130 01:07:41.641978 137274321021824 utils.py:1231] [29150] l2_params = 331.9797221566145
+I1130 01:07:41.642175 137274321021824 utils.py:1231] [29150] train/loss = 2.9180034399032593
+I1130 01:07:41.642272 137274321021824 utils.py:1231] [29150] l2_grads = 1.3982269763946533
+I1130 01:07:41.642333 137274321021824 utils.py:1231] [29150] lr = 0.0009164907786448665
+I1130 01:07:41.642392 137274321021824 utils.py:1231] [29150] uptime = 184651.004753666
+I1130 01:07:41.642445 137274321021824 utils.py:1231] [29150] examples_seen = 29849600.0
+I1130 01:07:41.642495 137274321021824 utils.py:1231] [29150] progress = 0.258874097492962
+I1130 01:07:41.642544 137274321021824 utils.py:1231] [29150] epoch = 23.298758085401825
+I1130 01:07:41.642594 137274321021824 utils.py:1231] [29150] img/sec/core = 164.22090634427647
+I1130 01:07:41.642654 137274321021824 utils.py:1231] [29150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.25762816255055
+I1130 01:07:41.642704 137274321021824 utils.py:1231] [29150] core_hours = 51.25762816255055
+I1130 01:07:41.642765 137274321021824 train.py:125] NOTE: Steps:29150/112603 [25.9%]
+Walltime:2d3h17m (0s eval)
+ETA:6d2h45m
+Total train time:8d6h0m
+I1130 01:12:53.413481 137274321021824 utils.py:1231] [29200] l2_params = 331.9599575552466
+I1130 01:12:53.413752 137274321021824 utils.py:1231] [29200] train/loss = 2.6771544218063354
+I1130 01:12:53.413891 137274321021824 utils.py:1231] [29200] l2_grads = 1.3059247732162476
+I1130 01:12:53.413992 137274321021824 utils.py:1231] [29200] lr = 0.0009160667539905218
+I1130 01:12:53.414058 137274321021824 utils.py:1231] [29200] uptime = 184962.776418838
+I1130 01:12:53.414114 137274321021824 utils.py:1231] [29200] examples_seen = 29900800.0
+I1130 01:12:53.414168 137274321021824 utils.py:1231] [29200] progress = 0.25931813539603743
+I1130 01:12:53.414237 137274321021824 utils.py:1231] [29200] epoch = 23.33872164987078
+I1130 01:12:53.414297 137274321021824 utils.py:1231] [29200] img/sec/core = 164.2227492731151
+I1130 01:12:53.414355 137274321021824 utils.py:1231] [29200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.3442314028761
+I1130 01:12:53.414407 137274321021824 utils.py:1231] [29200] core_hours = 51.3442314028761
+I1130 01:12:53.414469 137274321021824 train.py:125] NOTE: Steps:29200/112603 [25.9%]
+Walltime:2d3h22m (0s eval)
+ETA:6d2h39m
+Total train time:8d6h0m
+I1130 01:18:05.200642 137274321021824 utils.py:1231] [29250] l2_params = 331.9162301463139
+I1130 01:18:05.200876 137274321021824 utils.py:1231] [29250] train/loss = 2.7293890714645386
+I1130 01:18:05.201016 137274321021824 utils.py:1231] [29250] l2_grads = 1.2650291919708252
+I1130 01:18:05.201129 137274321021824 utils.py:1231] [29250] lr = 0.0009156417541611634
+I1130 01:18:05.201195 137274321021824 utils.py:1231] [29250] uptime = 185274.56355649
+I1130 01:18:05.201257 137274321021824 utils.py:1231] [29250] examples_seen = 29952000.0
+I1130 01:18:05.201313 137274321021824 utils.py:1231] [29250] progress = 0.25976217329911283
+I1130 01:18:05.201369 137274321021824 utils.py:1231] [29250] epoch = 23.37868521433974
+I1130 01:18:05.201425 137274321021824 utils.py:1231] [29250] img/sec/core = 164.214599696359
+I1130 01:18:05.201486 137274321021824 utils.py:1231] [29250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.430838941112775
+I1130 01:18:05.201542 137274321021824 utils.py:1231] [29250] core_hours = 51.430838941112775
+I1130 01:18:05.201612 137274321021824 train.py:125] NOTE: Steps:29250/112603 [26.0%]
+Walltime:2d3h27m (0s eval)
+ETA:6d2h34m
+Total train time:8d6h0m
+I1130 01:23:16.998042 137274321021824 utils.py:1231] [29300] l2_params = 331.93185039119686
+I1130 01:23:16.998274 137274321021824 utils.py:1231] [29300] train/loss = 2.8485957384109497
+I1130 01:23:16.998405 137274321021824 utils.py:1231] [29300] l2_grads = 1.2617563009262085
+I1130 01:23:16.998486 137274321021824 utils.py:1231] [29300] lr = 0.0009152157801529028
+I1130 01:23:16.998554 137274321021824 utils.py:1231] [29300] uptime = 185586.360912563
+I1130 01:23:16.998624 137274321021824 utils.py:1231] [29300] examples_seen = 30003200.0
+I1130 01:23:16.998697 137274321021824 utils.py:1231] [29300] progress = 0.26020621120218823
+I1130 01:23:16.998804 137274321021824 utils.py:1231] [29300] epoch = 23.418648778808695
+I1130 01:23:16.998876 137274321021824 utils.py:1231] [29300] img/sec/core = 164.20921795120904
+I1130 01:23:16.998969 137274321021824 utils.py:1231] [29300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.517449317799716
+I1130 01:23:16.999034 137274321021824 utils.py:1231] [29300] core_hours = 51.517449317799716
+I1130 01:23:16.999107 137274321021824 train.py:125] NOTE: Steps:29300/112603 [26.0%]
+Walltime:2d3h33m (0s eval)
+ETA:6d2h28m
+Total train time:8d6h0m
+I1130 01:28:27.907207 137274321021824 utils.py:1231] [29350] l2_params = 331.9471142293571
+I1130 01:28:27.907478 137274321021824 utils.py:1231] [29350] train/loss = 2.89619043469429
+I1130 01:28:27.907601 137274321021824 utils.py:1231] [29350] l2_grads = 1.4156969785690308
+I1130 01:28:27.907700 137274321021824 utils.py:1231] [29350] lr = 0.0009147888329641353
+I1130 01:28:27.907778 137274321021824 utils.py:1231] [29350] uptime = 185897.270138836
+I1130 01:28:27.907878 137274321021824 utils.py:1231] [29350] examples_seen = 30054400.0
+I1130 01:28:27.907948 137274321021824 utils.py:1231] [29350] progress = 0.26065024910526363
+I1130 01:28:27.908005 137274321021824 utils.py:1231] [29350] epoch = 23.458612343277654
+I1130 01:28:27.908061 137274321021824 utils.py:1231] [29350] img/sec/core = 164.67829087529512
+I1130 01:28:27.908121 137274321021824 utils.py:1231] [29350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.60381299176444
+I1130 01:28:27.908177 137274321021824 utils.py:1231] [29350] core_hours = 51.60381299176444
+I1130 01:28:27.908242 137274321021824 train.py:125] NOTE: Steps:29350/112603 [26.1%]
+Walltime:2d3h38m (0s eval)
+ETA:6d2h23m
+Total train time:8d5h59m
+I1130 01:33:39.688183 137274321021824 utils.py:1231] [29400] l2_params = 331.949866933314
+I1130 01:33:39.688435 137274321021824 utils.py:1231] [29400] train/loss = 4.170697450637817
+I1130 01:33:39.688565 137274321021824 utils.py:1231] [29400] l2_grads = 1.115350365638733
+I1130 01:33:39.688651 137274321021824 utils.py:1231] [29400] lr = 0.0009143609135955377
+I1130 01:33:39.688720 137274321021824 utils.py:1231] [29400] uptime = 186209.05107731
+I1130 01:33:39.688786 137274321021824 utils.py:1231] [29400] examples_seen = 30105600.0
+I1130 01:33:39.688841 137274321021824 utils.py:1231] [29400] progress = 0.26109428700833903
+I1130 01:33:39.688908 137274321021824 utils.py:1231] [29400] epoch = 23.49857590774661
+I1130 01:33:39.688966 137274321021824 utils.py:1231] [29400] img/sec/core = 164.21786479506142
+I1130 01:33:39.689038 137274321021824 utils.py:1231] [29400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.690418808007216
+I1130 01:33:39.689092 137274321021824 utils.py:1231] [29400] core_hours = 51.690418808007216
+I1130 01:33:39.689156 137274321021824 train.py:125] NOTE: Steps:29400/112603 [26.1%]
+Walltime:2d3h43m (0s eval)
+ETA:6d2h17m
+Total train time:8d5h59m
+I1130 01:38:51.470991 137274321021824 utils.py:1231] [29450] l2_params = 331.93962308735854
+I1130 01:38:51.471197 137274321021824 utils.py:1231] [29450] train/loss = 2.7721298038959503
+I1130 01:38:51.471289 137274321021824 utils.py:1231] [29450] l2_grads = 1.3221651315689087
+I1130 01:38:51.471354 137274321021824 utils.py:1231] [29450] lr = 0.0009139320230500652
+I1130 01:38:51.471404 137274321021824 utils.py:1231] [29450] uptime = 186520.83376652998
+I1130 01:38:51.471455 137274321021824 utils.py:1231] [29450] examples_seen = 30156800.0
+I1130 01:38:51.471503 137274321021824 utils.py:1231] [29450] progress = 0.26153832491141443
+I1130 01:38:51.471550 137274321021824 utils.py:1231] [29450] epoch = 23.538539472215565
+I1130 01:38:51.471600 137274321021824 utils.py:1231] [29450] img/sec/core = 164.21694266635632
+I1130 01:38:51.471655 137274321021824 utils.py:1231] [29450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.777025110568324
+I1130 01:38:51.471704 137274321021824 utils.py:1231] [29450] core_hours = 51.777025110568324
+I1130 01:38:51.471763 137274321021824 train.py:125] NOTE: Steps:29450/112603 [26.2%]
+Walltime:2d3h48m (0s eval)
+ETA:6d2h12m
+Total train time:8d5h59m
+I1130 01:44:03.259282 137274321021824 utils.py:1231] [29500] l2_params = 331.90972926554997
+I1130 01:44:03.259548 137274321021824 utils.py:1231] [29500] train/loss = 2.754889190196991
+I1130 01:44:03.259673 137274321021824 utils.py:1231] [29500] l2_grads = 1.3704277276992798
+I1130 01:44:03.259756 137274321021824 utils.py:1231] [29500] lr = 0.0009135021623329497
+I1130 01:44:03.259818 137274321021824 utils.py:1231] [29500] uptime = 186832.62217993703
+I1130 01:44:03.259876 137274321021824 utils.py:1231] [29500] examples_seen = 30208000.0
+I1130 01:44:03.259947 137274321021824 utils.py:1231] [29500] progress = 0.26198236281448983
+I1130 01:44:03.259999 137274321021824 utils.py:1231] [29500] epoch = 23.578503036684523
+I1130 01:44:03.260054 137274321021824 utils.py:1231] [29500] img/sec/core = 164.21392777403113
+I1130 01:44:03.260118 137274321021824 utils.py:1231] [29500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.863633003181384
+I1130 01:44:03.260186 137274321021824 utils.py:1231] [29500] core_hours = 51.863633003181384
+I1130 01:44:03.260253 137274321021824 train.py:125] NOTE: Steps:29500/112603 [26.2%]
+Walltime:2d3h53m (0s eval)
+ETA:6d2h6m
+Total train time:8d5h58m
+I1130 01:49:13.505251 137274321021824 utils.py:1231] [29550] l2_params = 331.90326408804475
+I1130 01:49:13.505522 137274321021824 utils.py:1231] [29550] train/loss = 3.357947438955307
+I1130 01:49:13.505630 137274321021824 utils.py:1231] [29550] l2_grads = 1.1048035621643066
+I1130 01:49:13.505706 137274321021824 utils.py:1231] [29550] lr = 0.0009130713324516961
+I1130 01:49:13.505762 137274321021824 utils.py:1231] [29550] uptime = 187142.86812391103
+I1130 01:49:13.505817 137274321021824 utils.py:1231] [29550] examples_seen = 30259200.0
+I1130 01:49:13.505866 137274321021824 utils.py:1231] [29550] progress = 0.26242640071756523
+I1130 01:49:13.505925 137274321021824 utils.py:1231] [29550] epoch = 23.61846660115348
+I1130 01:49:13.505976 137274321021824 utils.py:1231] [29550] img/sec/core = 165.03036057190147
+I1130 01:49:13.506032 137274321021824 utils.py:1231] [29550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 51.94981243206305
+I1130 01:49:13.506081 137274321021824 utils.py:1231] [29550] core_hours = 51.94981243206305
+I1130 01:49:13.506142 137274321021824 train.py:125] NOTE: Steps:29550/112603 [26.2%]
+Walltime:2d3h59m (0s eval)
+ETA:6d2h1m
+Total train time:8d5h58m
+I1130 01:54:25.296252 137274321021824 utils.py:1231] [29600] l2_params = 331.89831777049267
+I1130 01:54:25.296473 137274321021824 utils.py:1231] [29600] train/loss = 2.89040943980217
+I1130 01:54:25.296577 137274321021824 utils.py:1231] [29600] l2_grads = 1.4807915687561035
+I1130 01:54:25.296650 137274321021824 utils.py:1231] [29600] lr = 0.0009126395344160805
+I1130 01:54:25.296725 137274321021824 utils.py:1231] [29600] uptime = 187454.659085227
+I1130 01:54:25.296786 137274321021824 utils.py:1231] [29600] examples_seen = 30310400.0
+I1130 01:54:25.296836 137274321021824 utils.py:1231] [29600] progress = 0.26287043862064063
+I1130 01:54:25.296898 137274321021824 utils.py:1231] [29600] epoch = 23.658430165622438
+I1130 01:54:25.296952 137274321021824 utils.py:1231] [29600] img/sec/core = 164.2125858424378
+I1130 01:54:25.297008 137274321021824 utils.py:1231] [29600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.036421032428606
+I1130 01:54:25.297058 137274321021824 utils.py:1231] [29600] core_hours = 52.036421032428606
+I1130 01:54:25.297119 137274321021824 train.py:125] NOTE: Steps:29600/112603 [26.3%]
+Walltime:2d4h4m (0s eval)
+ETA:6d1h55m
+Total train time:8d5h58m
+I1130 01:59:34.742147 137274321021824 utils.py:1231] [29650] l2_params = 331.8634619699821
+I1130 01:59:34.742335 137274321021824 utils.py:1231] [29650] train/loss = 3.3989475667476654
+I1130 01:59:34.742447 137274321021824 utils.py:1231] [29650] l2_grads = 1.2057181596755981
+I1130 01:59:34.742528 137274321021824 utils.py:1231] [29650] lr = 0.0009122067692381491
+I1130 01:59:34.742582 137274321021824 utils.py:1231] [29650] uptime = 187764.104944858
+I1130 01:59:34.742649 137274321021824 utils.py:1231] [29650] examples_seen = 30361600.0
+I1130 01:59:34.742729 137274321021824 utils.py:1231] [29650] progress = 0.2633144765237161
+I1130 01:59:34.742779 137274321021824 utils.py:1231] [29650] epoch = 23.698393730091393
+I1130 01:59:34.742841 137274321021824 utils.py:1231] [29650] img/sec/core = 165.45705300776592
+I1130 01:59:34.742901 137274321021824 utils.py:1231] [29650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.12237821565944
+I1130 01:59:34.742969 137274321021824 utils.py:1231] [29650] core_hours = 52.12237821565944
+I1130 01:59:34.743071 137274321021824 train.py:125] NOTE: Steps:29650/112603 [26.3%]
+Walltime:2d4h9m (0s eval)
+ETA:6d1h50m
+Total train time:8d5h57m
+I1130 02:04:45.075422 137274321021824 utils.py:1231] [29700] l2_params = 331.8653266312342
+I1130 02:04:45.075683 137274321021824 utils.py:1231] [29700] train/loss = 2.6909259557724
+I1130 02:04:45.075796 137274321021824 utils.py:1231] [29700] l2_grads = 1.3424897193908691
+I1130 02:04:45.075864 137274321021824 utils.py:1231] [29700] lr = 0.0009117730379322153
+I1130 02:04:45.075926 137274321021824 utils.py:1231] [29700] uptime = 188074.438288728
+I1130 02:04:45.075976 137274321021824 utils.py:1231] [29700] examples_seen = 30412800.0
+I1130 02:04:45.076023 137274321021824 utils.py:1231] [29700] progress = 0.2637585144267915
+I1130 02:04:45.076068 137274321021824 utils.py:1231] [29700] epoch = 23.738357294560352
+I1130 02:04:45.076117 137274321021824 utils.py:1231] [29700] img/sec/core = 164.98388269051628
+I1130 02:04:45.076170 137274321021824 utils.py:1231] [29700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.208581922289994
+I1130 02:04:45.076217 137274321021824 utils.py:1231] [29700] core_hours = 52.208581922289994
+I1130 02:04:45.076274 137274321021824 train.py:125] NOTE: Steps:29700/112603 [26.4%]
+Walltime:2d4h14m (0s eval)
+ETA:6d1h44m
+Total train time:8d5h57m
+I1130 02:09:56.846673 137274321021824 utils.py:1231] [29750] l2_params = 331.8423308458813
+I1130 02:09:56.846872 137274321021824 utils.py:1231] [29750] train/loss = 4.931036293506622
+I1130 02:09:56.846975 137274321021824 utils.py:1231] [29750] l2_grads = 1.136217713356018
+I1130 02:09:56.847043 137274321021824 utils.py:1231] [29750] lr = 0.0009113383415148558
+I1130 02:09:56.847133 137274321021824 utils.py:1231] [29750] uptime = 188386.20948509
+I1130 02:09:56.847208 137274321021824 utils.py:1231] [29750] examples_seen = 30464000.0
+I1130 02:09:56.847264 137274321021824 utils.py:1231] [29750] progress = 0.2642025523298669
+I1130 02:09:56.847320 137274321021824 utils.py:1231] [29750] epoch = 23.778320859029307
+I1130 02:09:56.847386 137274321021824 utils.py:1231] [29750] img/sec/core = 164.22299621466544
+I1130 02:09:56.847463 137274321021824 utils.py:1231] [29750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.29518503239055
+I1130 02:09:56.847522 137274321021824 utils.py:1231] [29750] core_hours = 52.29518503239055
+I1130 02:09:56.847584 137274321021824 train.py:125] NOTE: Steps:29750/112603 [26.4%]
+Walltime:2d4h19m (0s eval)
+ETA:6d1h39m
+Total train time:8d5h56m
+I1130 02:15:04.571729 137274321021824 utils.py:1231] [29800] l2_params = 331.84504821691166
+I1130 02:15:04.572028 137274321021824 utils.py:1231] [29800] train/loss = 3.3069345355033875
+I1130 02:15:04.572263 137274321021824 utils.py:1231] [29800] l2_grads = 1.1633715629577637
+I1130 02:15:04.572390 137274321021824 utils.py:1231] [29800] lr = 0.0009109026810049097
+I1130 02:15:04.572451 137274321021824 utils.py:1231] [29800] uptime = 188693.93481257698
+I1130 02:15:04.572518 137274321021824 utils.py:1231] [29800] examples_seen = 30515200.0
+I1130 02:15:04.572573 137274321021824 utils.py:1231] [29800] progress = 0.2646465902329423
+I1130 02:15:04.572629 137274321021824 utils.py:1231] [29800] epoch = 23.818284423498262
+I1130 02:15:04.572688 137274321021824 utils.py:1231] [29800] img/sec/core = 166.38214481115497
+I1130 02:15:04.572750 137274321021824 utils.py:1231] [29800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.38066429002583
+I1130 02:15:04.572805 137274321021824 utils.py:1231] [29800] core_hours = 52.38066429002583
+I1130 02:15:04.572869 137274321021824 train.py:125] NOTE: Steps:29800/112603 [26.5%]
+Walltime:2d4h24m (0s eval)
+ETA:6d1h33m
+Total train time:8d5h56m
+I1130 02:20:16.350904 137274321021824 utils.py:1231] [29850] l2_params = 331.8045915749589
+I1130 02:20:16.351162 137274321021824 utils.py:1231] [29850] train/loss = 2.9520691335201263
+I1130 02:20:16.351286 137274321021824 utils.py:1231] [29850] l2_grads = 1.263121247291565
+I1130 02:20:16.351374 137274321021824 utils.py:1231] [29850] lr = 0.000910466057423475
+I1130 02:20:16.351450 137274321021824 utils.py:1231] [29850] uptime = 189005.713811255
+I1130 02:20:16.351522 137274321021824 utils.py:1231] [29850] examples_seen = 30566400.0
+I1130 02:20:16.351596 137274321021824 utils.py:1231] [29850] progress = 0.2650906281360177
+I1130 02:20:16.351657 137274321021824 utils.py:1231] [29850] epoch = 23.85824798796722
+I1130 02:20:16.351719 137274321021824 utils.py:1231] [29850] img/sec/core = 164.21888650965286
+I1130 02:20:16.351792 137274321021824 utils.py:1231] [29850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.46726956743638
+I1130 02:20:16.351852 137274321021824 utils.py:1231] [29850] core_hours = 52.46726956743638
+I1130 02:20:16.351934 137274321021824 train.py:125] NOTE: Steps:29850/112603 [26.5%]
+Walltime:2d4h30m (0s eval)
+ETA:6d1h27m
+Total train time:8d5h56m
+I1130 02:25:28.131077 137274321021824 utils.py:1231] [29900] l2_params = 331.75656026920313
+I1130 02:25:28.131296 137274321021824 utils.py:1231] [29900] train/loss = 3.055136024951935
+I1130 02:25:28.131398 137274321021824 utils.py:1231] [29900] l2_grads = 1.2778325080871582
+I1130 02:25:28.131474 137274321021824 utils.py:1231] [29900] lr = 0.0009100284717939087
+I1130 02:25:28.131541 137274321021824 utils.py:1231] [29900] uptime = 189317.493902385
+I1130 02:25:28.131600 137274321021824 utils.py:1231] [29900] examples_seen = 30617600.0
+I1130 02:25:28.131655 137274321021824 utils.py:1231] [29900] progress = 0.2655346660390931
+I1130 02:25:28.131709 137274321021824 utils.py:1231] [29900] epoch = 23.898211552436177
+I1130 02:25:28.131770 137274321021824 utils.py:1231] [29900] img/sec/core = 164.21831110008756
+I1130 02:25:28.131829 137274321021824 utils.py:1231] [29900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.55387514830583
+I1130 02:25:28.131888 137274321021824 utils.py:1231] [29900] core_hours = 52.55387514830583
+I1130 02:25:28.131959 137274321021824 train.py:125] NOTE: Steps:29900/112603 [26.6%]
+Walltime:2d4h35m (0s eval)
+ETA:6d1h22m
+Total train time:8d5h55m
+I1130 02:30:35.904339 137274321021824 utils.py:1231] [29950] l2_params = 331.69996309954405
+I1130 02:30:35.904612 137274321021824 utils.py:1231] [29950] train/loss = 3.0644319653511047
+I1130 02:30:35.904788 137274321021824 utils.py:1231] [29950] l2_grads = 1.3023663759231567
+I1130 02:30:35.904892 137274321021824 utils.py:1231] [29950] lr = 0.0009095899251418212
+I1130 02:30:35.904953 137274321021824 utils.py:1231] [29950] uptime = 189625.26731373603
+I1130 02:30:35.905018 137274321021824 utils.py:1231] [29950] examples_seen = 30668800.0
+I1130 02:30:35.905077 137274321021824 utils.py:1231] [29950] progress = 0.2659787039421685
+I1130 02:30:35.905137 137274321021824 utils.py:1231] [29950] epoch = 23.938175116905136
+I1130 02:30:35.905194 137274321021824 utils.py:1231] [29950] img/sec/core = 166.35615069948315
+I1130 02:30:35.905254 137274321021824 utils.py:1231] [29950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.63936776257
+I1130 02:30:35.905312 137274321021824 utils.py:1231] [29950] core_hours = 52.63936776257
+I1130 02:30:35.905375 137274321021824 train.py:125] NOTE: Steps:29950/112603 [26.6%]
+Walltime:2d4h40m (0s eval)
+ETA:6d1h16m
+Total train time:8d5h55m
+I1130 02:35:47.673835 137274321021824 utils.py:1231] [30000] l2_params = 331.6803625485186
+I1130 02:35:47.674107 137274321021824 utils.py:1231] [30000] train/loss = 4.066617727279663
+I1130 02:35:47.674231 137274321021824 utils.py:1231] [30000] l2_grads = 1.0621674060821533
+I1130 02:35:47.674317 137274321021824 utils.py:1231] [30000] lr = 0.0009091504184950754
+I1130 02:35:47.674395 137274321021824 utils.py:1231] [30000] uptime = 189937.03675641498
+I1130 02:35:47.674458 137274321021824 utils.py:1231] [30000] examples_seen = 30720000.0
+I1130 02:35:47.674527 137274321021824 utils.py:1231] [30000] progress = 0.2664227418452439
+I1130 02:35:47.674585 137274321021824 utils.py:1231] [30000] epoch = 23.97813868137409
+I1130 02:35:47.674658 137274321021824 utils.py:1231] [30000] img/sec/core = 164.22391995843802
+I1130 02:35:47.674734 137274321021824 utils.py:1231] [30000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.72597038553638
+I1130 02:35:47.674787 137274321021824 utils.py:1231] [30000] core_hours = 52.72597038553638
+I1130 02:35:47.674864 137274321021824 train.py:125] NOTE: Steps:30000/112603 [26.6%]
+Walltime:2d4h45m (0s eval)
+ETA:6d1h11m
+Total train time:8d5h54m
+I1130 02:35:48.012427 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:30000/112603 [26.6%]
+Walltime:2d4h45m (0s eval)
+ETA:6d1h11m
+Total train time:8d5h54m
+I1130 02:37:20.440173 137274321021824 utils.py:1231] [30000] val/acc@1 = 0.5828085140306123
+I1130 02:37:20.440375 137274321021824 utils.py:1231] [30000] val/loss = 1.767182093189687
+I1130 02:37:20.440524 137274321021824 utils.py:1231] [30000] z/secs/eval/val = 92.42785263300175
+I1130 02:37:20.440582 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 92.42785263300175
+I1130 02:42:27.168564 137274321021824 utils.py:1231] [30050] l2_params = 331.6710238187696
+I1130 02:42:27.168771 137274321021824 utils.py:1231] [30050] train/loss = 4.650609791278839
+I1130 02:42:27.168868 137274321021824 utils.py:1231] [30050] l2_grads = 1.1607706546783447
+I1130 02:42:27.168957 137274321021824 utils.py:1231] [30050] lr = 0.0009087099528837858
+I1130 02:42:27.169027 137274321021824 utils.py:1231] [30050] uptime = 190336.53138881997
+I1130 02:42:27.169097 137274321021824 utils.py:1231] [30050] examples_seen = 30771200.0
+I1130 02:42:27.169176 137274321021824 utils.py:1231] [30050] progress = 0.2668667797483193
+I1130 02:42:27.169253 137274321021824 utils.py:1231] [30050] epoch = 24.018102245843046
+I1130 02:42:27.169326 137274321021824 utils.py:1231] [30050] img/sec/core = 128.1619222059908
+I1130 02:42:27.169391 137274321021824 utils.py:1231] [30050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.836941116759995
+I1130 02:42:27.169446 137274321021824 utils.py:1231] [30050] core_hours = 52.836941116759995
+I1130 02:42:27.169504 137274321021824 train.py:125] NOTE: Steps:30050/112603 [26.7%]
+Walltime:2d4h52m (0s eval)
+ETA:6d1h9m
+Total train time:8d6h0m
+I1130 02:47:38.947055 137274321021824 utils.py:1231] [30100] l2_params = 331.6532381758112
+I1130 02:47:38.947280 137274321021824 utils.py:1231] [30100] train/loss = 2.6947683691978455
+I1130 02:47:38.947409 137274321021824 utils.py:1231] [30100] l2_grads = 1.3840500116348267
+I1130 02:47:38.947487 137274321021824 utils.py:1231] [30100] lr = 0.0009082685293403128
+I1130 02:47:38.947548 137274321021824 utils.py:1231] [30100] uptime = 190648.309910073
+I1130 02:47:38.947623 137274321021824 utils.py:1231] [30100] examples_seen = 30822400.0
+I1130 02:47:38.947679 137274321021824 utils.py:1231] [30100] progress = 0.26731081765139475
+I1130 02:47:38.947734 137274321021824 utils.py:1231] [30100] epoch = 24.058065810312005
+I1130 02:47:38.947789 137274321021824 utils.py:1231] [30100] img/sec/core = 164.21913797726884
+I1130 02:47:38.947852 137274321021824 utils.py:1231] [30100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 52.923546261552495
+I1130 02:47:38.947914 137274321021824 utils.py:1231] [30100] core_hours = 52.923546261552495
+I1130 02:47:38.947977 137274321021824 train.py:125] NOTE: Steps:30100/112603 [26.7%]
+Walltime:2d4h57m (0s eval)
+ETA:6d1h4m
+Total train time:8d5h59m
+I1130 02:52:46.541968 137274321021824 utils.py:1231] [30150] l2_params = 331.61609076045534
+I1130 02:52:46.542197 137274321021824 utils.py:1231] [30150] train/loss = 4.888002634048462
+I1130 02:52:46.542301 137274321021824 utils.py:1231] [30150] l2_grads = 0.9720342755317688
+I1130 02:52:46.542384 137274321021824 utils.py:1231] [30150] lr = 0.000907826148899262
+I1130 02:52:46.542459 137274321021824 utils.py:1231] [30150] uptime = 190955.904812855
+I1130 02:52:46.542527 137274321021824 utils.py:1231] [30150] examples_seen = 30873600.0
+I1130 02:52:46.542591 137274321021824 utils.py:1231] [30150] progress = 0.26775485555447015
+I1130 02:52:46.542646 137274321021824 utils.py:1231] [30150] epoch = 24.09802937478096
+I1130 02:52:46.542706 137274321021824 utils.py:1231] [30150] img/sec/core = 166.45269325640004
+I1130 02:52:46.542762 137274321021824 utils.py:1231] [30150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.008989290103045
+I1130 02:52:46.542815 137274321021824 utils.py:1231] [30150] core_hours = 53.008989290103045
+I1130 02:52:46.542880 137274321021824 train.py:125] NOTE: Steps:30150/112603 [26.8%]
+Walltime:2d5h2m (0s eval)
+ETA:6d0h58m
+Total train time:8d5h59m
+I1130 02:57:56.903935 137274321021824 utils.py:1231] [30200] l2_params = 331.5678882198809
+I1130 02:57:56.904173 137274321021824 utils.py:1231] [30200] train/loss = 5.336273670196533
+I1130 02:57:56.904282 137274321021824 utils.py:1231] [30200] l2_grads = 1.2519183158874512
+I1130 02:57:56.904369 137274321021824 utils.py:1231] [30200] lr = 0.0009073828125974818
+I1130 02:57:56.904434 137274321021824 utils.py:1231] [30200] uptime = 191266.26679534698
+I1130 02:57:56.904499 137274321021824 utils.py:1231] [30200] examples_seen = 30924800.0
+I1130 02:57:56.904559 137274321021824 utils.py:1231] [30200] progress = 0.26819889345754555
+I1130 02:57:56.904624 137274321021824 utils.py:1231] [30200] epoch = 24.13799293924992
+I1130 02:57:56.904695 137274321021824 utils.py:1231] [30200] img/sec/core = 164.96865881864042
+I1130 02:57:56.904758 137274321021824 utils.py:1231] [30200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.09520095190638
+I1130 02:57:56.904815 137274321021824 utils.py:1231] [30200] core_hours = 53.09520095190638
+I1130 02:57:56.904904 137274321021824 train.py:125] NOTE: Steps:30200/112603 [26.8%]
+Walltime:2d5h7m (0s eval)
+ETA:6d0h53m
+Total train time:8d5h58m
+I1130 03:03:08.671910 137274321021824 utils.py:1231] [30250] l2_params = 331.5506710850195
+I1130 03:03:08.672145 137274321021824 utils.py:1231] [30250] train/loss = 2.8055796027183533
+I1130 03:03:08.672252 137274321021824 utils.py:1231] [30250] l2_grads = 1.3895068168640137
+I1130 03:03:08.672374 137274321021824 utils.py:1231] [30250] lr = 0.0009069385214740631
+I1130 03:03:08.672445 137274321021824 utils.py:1231] [30250] uptime = 191578.034805774
+I1130 03:03:08.672504 137274321021824 utils.py:1231] [30250] examples_seen = 30976000.0
+I1130 03:03:08.672561 137274321021824 utils.py:1231] [30250] progress = 0.26864293136062095
+I1130 03:03:08.672615 137274321021824 utils.py:1231] [30250] epoch = 24.177956503718875
+I1130 03:03:08.672673 137274321021824 utils.py:1231] [30250] img/sec/core = 164.22467439771717
+I1130 03:03:08.672735 137274321021824 utils.py:1231] [30250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.181803177024996
+I1130 03:03:08.672791 137274321021824 utils.py:1231] [30250] core_hours = 53.181803177024996
+I1130 03:03:08.672856 137274321021824 train.py:125] NOTE: Steps:30250/112603 [26.9%]
+Walltime:2d5h12m (0s eval)
+ETA:6d0h47m
+Total train time:8d5h58m
+I1130 03:08:16.903821 137274321021824 utils.py:1231] [30300] l2_params = 331.535392456233
+I1130 03:08:16.904074 137274321021824 utils.py:1231] [30300] train/loss = 2.9056167900562286
+I1130 03:08:16.904175 137274321021824 utils.py:1231] [30300] l2_grads = 1.4630610942840576
+I1130 03:08:16.904256 137274321021824 utils.py:1231] [30300] lr = 0.0009064932765703317
+I1130 03:08:16.904332 137274321021824 utils.py:1231] [30300] uptime = 191886.26668725902
+I1130 03:08:16.904415 137274321021824 utils.py:1231] [30300] examples_seen = 31027200.0
+I1130 03:08:16.904478 137274321021824 utils.py:1231] [30300] progress = 0.26908696926369635
+I1130 03:08:16.904543 137274321021824 utils.py:1231] [30300] epoch = 24.217920068187833
+I1130 03:08:16.904617 137274321021824 utils.py:1231] [30300] img/sec/core = 166.10870930459396
+I1130 03:08:16.904705 137274321021824 utils.py:1231] [30300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.267423144104164
+I1130 03:08:16.904763 137274321021824 utils.py:1231] [30300] core_hours = 53.267423144104164
+I1130 03:08:16.904842 137274321021824 train.py:125] NOTE: Steps:30300/112603 [26.9%]
+Walltime:2d5h18m (0s eval)
+ETA:6d0h41m
+Total train time:8d5h58m
+I1130 03:13:25.795100 137274321021824 utils.py:1231] [30350] l2_params = 331.50662252106355
+I1130 03:13:25.795364 137274321021824 utils.py:1231] [30350] train/loss = 2.798760950565338
+I1130 03:13:25.795497 137274321021824 utils.py:1231] [30350] l2_grads = 1.4612596035003662
+I1130 03:13:25.795608 137274321021824 utils.py:1231] [30350] lr = 0.0009060470789298505
+I1130 03:13:25.795696 137274321021824 utils.py:1231] [30350] uptime = 192195.158051453
+I1130 03:13:25.795759 137274321021824 utils.py:1231] [30350] examples_seen = 31078400.0
+I1130 03:13:25.795822 137274321021824 utils.py:1231] [30350] progress = 0.26953100716677175
+I1130 03:13:25.795911 137274321021824 utils.py:1231] [30350] epoch = 24.25788363265679
+I1130 03:13:25.796011 137274321021824 utils.py:1231] [30350] img/sec/core = 165.75406740036232
+I1130 03:13:25.796107 137274321021824 utils.py:1231] [30350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.353226300824716
+I1130 03:13:25.796195 137274321021824 utils.py:1231] [30350] core_hours = 53.353226300824716
+I1130 03:13:25.796304 137274321021824 train.py:125] NOTE: Steps:30350/112603 [27.0%]
+Walltime:2d5h23m (0s eval)
+ETA:6d0h36m
+Total train time:8d5h57m
+I1130 03:18:33.627500 137274321021824 utils.py:1231] [30400] l2_params = 331.5088606713041
+I1130 03:18:33.627693 137274321021824 utils.py:1231] [30400] train/loss = 3.330028623342514
+I1130 03:18:33.627783 137274321021824 utils.py:1231] [30400] l2_grads = 1.1791187524795532
+I1130 03:18:33.627845 137274321021824 utils.py:1231] [30400] lr = 0.0009055999295984157
+I1130 03:18:33.627903 137274321021824 utils.py:1231] [30400] uptime = 192502.99026461103
+I1130 03:18:33.627955 137274321021824 utils.py:1231] [30400] examples_seen = 31129600.0
+I1130 03:18:33.628004 137274321021824 utils.py:1231] [30400] progress = 0.26997504506984715
+I1130 03:18:33.628053 137274321021824 utils.py:1231] [30400] epoch = 24.297847197125744
+I1130 03:18:33.628104 137274321021824 utils.py:1231] [30400] img/sec/core = 166.32437351095967
+I1130 03:18:33.628171 137274321021824 utils.py:1231] [30400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.43873524892417
+I1130 03:18:33.628223 137274321021824 utils.py:1231] [30400] core_hours = 53.43873524892417
+I1130 03:18:33.628285 137274321021824 train.py:125] NOTE: Steps:30400/112603 [27.0%]
+Walltime:2d5h28m (0s eval)
+ETA:6d0h30m
+Total train time:8d5h57m
+I1130 03:23:45.412328 137274321021824 utils.py:1231] [30450] l2_params = 331.48857745658586
+I1130 03:23:45.412653 137274321021824 utils.py:1231] [30450] train/loss = 3.3749110400676727
+I1130 03:23:45.412888 137274321021824 utils.py:1231] [30450] l2_grads = 1.1607258319854736
+I1130 03:23:45.413025 137274321021824 utils.py:1231] [30450] lr = 0.0009051518296240534
+I1130 03:23:45.413120 137274321021824 utils.py:1231] [30450] uptime = 192814.77546997397
+I1130 03:23:45.413224 137274321021824 utils.py:1231] [30450] examples_seen = 31180800.0
+I1130 03:23:45.413314 137274321021824 utils.py:1231] [30450] progress = 0.27041908297292255
+I1130 03:23:45.413392 137274321021824 utils.py:1231] [30450] epoch = 24.337810761594703
+I1130 03:23:45.413467 137274321021824 utils.py:1231] [30450] img/sec/core = 164.2156174164637
+I1130 03:23:45.413557 137274321021824 utils.py:1231] [30450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.525342250413885
+I1130 03:23:45.413621 137274321021824 utils.py:1231] [30450] core_hours = 53.525342250413885
+I1130 03:23:45.413711 137274321021824 train.py:125] NOTE: Steps:30450/112603 [27.0%]
+Walltime:2d5h33m (0s eval)
+ETA:6d0h25m
+Total train time:8d5h56m
+I1130 03:28:54.705855 137274321021824 utils.py:1231] [30500] l2_params = 331.47606708716427
+I1130 03:28:54.706077 137274321021824 utils.py:1231] [30500] train/loss = 2.7336867451667786
+I1130 03:28:54.706174 137274321021824 utils.py:1231] [30500] l2_grads = 1.3787248134613037
+I1130 03:28:54.706244 137274321021824 utils.py:1231] [30500] lr = 0.000904702780057017
+I1130 03:28:54.706322 137274321021824 utils.py:1231] [30500] uptime = 193124.06867987203
+I1130 03:28:54.706376 137274321021824 utils.py:1231] [30500] examples_seen = 31232000.0
+I1130 03:28:54.706434 137274321021824 utils.py:1231] [30500] progress = 0.27086312087599795
+I1130 03:28:54.706484 137274321021824 utils.py:1231] [30500] epoch = 24.37777432606366
+I1130 03:28:54.706534 137274321021824 utils.py:1231] [30500] img/sec/core = 165.53871330340888
+I1130 03:28:54.706587 137274321021824 utils.py:1231] [30500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.61125703094111
+I1130 03:28:54.706634 137274321021824 utils.py:1231] [30500] core_hours = 53.61125703094111
+I1130 03:28:54.706692 137274321021824 train.py:125] NOTE: Steps:30500/112603 [27.1%]
+Walltime:2d5h38m (0s eval)
+ETA:6d0h19m
+Total train time:8d5h56m
+I1130 03:34:03.201805 137274321021824 utils.py:1231] [30550] l2_params = 331.47555583187915
+I1130 03:34:03.202044 137274321021824 utils.py:1231] [30550] train/loss = 3.60325688123703
+I1130 03:34:03.202141 137274321021824 utils.py:1231] [30550] l2_grads = 1.1496623754501343
+I1130 03:34:03.202200 137274321021824 utils.py:1231] [30550] lr = 0.0009042527819497874
+I1130 03:34:03.202252 137274321021824 utils.py:1231] [30550] uptime = 193432.56461456
+I1130 03:34:03.202311 137274321021824 utils.py:1231] [30550] examples_seen = 31283200.0
+I1130 03:34:03.202364 137274321021824 utils.py:1231] [30550] progress = 0.27130715877907335
+I1130 03:34:03.202412 137274321021824 utils.py:1231] [30550] epoch = 24.417737890532617
+I1130 03:34:03.202462 137274321021824 utils.py:1231] [30550] img/sec/core = 165.96653065066477
+I1130 03:34:03.202517 137274321021824 utils.py:1231] [30550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.69695034613222
+I1130 03:34:03.202566 137274321021824 utils.py:1231] [30550] core_hours = 53.69695034613222
+I1130 03:34:03.202625 137274321021824 train.py:125] NOTE: Steps:30550/112603 [27.1%]
+Walltime:2d5h43m (0s eval)
+ETA:6d0h13m
+Total train time:8d5h55m
+I1130 03:39:09.044481 137274321021824 utils.py:1231] [30600] l2_params = 331.45314378962684
+I1130 03:39:09.044755 137274321021824 utils.py:1231] [30600] train/loss = 2.62238410115242
+I1130 03:39:09.044872 137274321021824 utils.py:1231] [30600] l2_grads = 1.324079155921936
+I1130 03:39:09.044965 137274321021824 utils.py:1231] [30600] lr = 0.0009038018363570673
+I1130 03:39:09.045023 137274321021824 utils.py:1231] [30600] uptime = 193738.407384501
+I1130 03:39:09.045078 137274321021824 utils.py:1231] [30600] examples_seen = 31334400.0
+I1130 03:39:09.045133 137274321021824 utils.py:1231] [30600] progress = 0.2717511966821488
+I1130 03:39:09.045190 137274321021824 utils.py:1231] [30600] epoch = 24.457701455001573
+I1130 03:39:09.045246 137274321021824 utils.py:1231] [30600] img/sec/core = 167.40627875518464
+I1130 03:39:09.045316 137274321021824 utils.py:1231] [30600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.78190667111583
+I1130 03:39:09.045369 137274321021824 utils.py:1231] [30600] core_hours = 53.78190667111583
+I1130 03:39:09.045431 137274321021824 train.py:125] NOTE: Steps:30600/112603 [27.2%]
+Walltime:2d5h48m (0s eval)
+ETA:6d0h8m
+Total train time:8d5h55m
+I1130 03:44:20.858837 137274321021824 utils.py:1231] [30650] l2_params = 331.430410625195
+I1130 03:44:20.859077 137274321021824 utils.py:1231] [30650] train/loss = 2.7639191150665283
+I1130 03:44:20.859191 137274321021824 utils.py:1231] [30650] l2_grads = 1.4036376476287842
+I1130 03:44:20.859276 137274321021824 utils.py:1231] [30650] lr = 0.0009033499443357809
+I1130 03:44:20.859363 137274321021824 utils.py:1231] [30650] uptime = 194050.22171563603
+I1130 03:44:20.859447 137274321021824 utils.py:1231] [30650] examples_seen = 31385600.0
+I1130 03:44:20.859506 137274321021824 utils.py:1231] [30650] progress = 0.2721952345852242
+I1130 03:44:20.859565 137274321021824 utils.py:1231] [30650] epoch = 24.497665019470528
+I1130 03:44:20.859615 137274321021824 utils.py:1231] [30650] img/sec/core = 164.2002784593972
+I1130 03:44:20.859669 137274321021824 utils.py:1231] [30650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.868521763097775
+I1130 03:44:20.859718 137274321021824 utils.py:1231] [30650] core_hours = 53.868521763097775
+I1130 03:44:20.859778 137274321021824 train.py:125] NOTE: Steps:30650/112603 [27.2%]
+Walltime:2d5h54m (0s eval)
+ETA:6d0h2m
+Total train time:8d5h55m
+I1130 03:49:28.815543 137274321021824 utils.py:1231] [30700] l2_params = 331.41090825713013
+I1130 03:49:28.815825 137274321021824 utils.py:1231] [30700] train/loss = 2.8562611639499664
+I1130 03:49:28.816014 137274321021824 utils.py:1231] [30700] l2_grads = 1.4189358949661255
+I1130 03:49:28.816106 137274321021824 utils.py:1231] [30700] lr = 0.0009028971069450698
+I1130 03:49:28.816180 137274321021824 utils.py:1231] [30700] uptime = 194358.178540988
+I1130 03:49:28.816252 137274321021824 utils.py:1231] [30700] examples_seen = 31436800.0
+I1130 03:49:28.816315 137274321021824 utils.py:1231] [30700] progress = 0.2726392724882996
+I1130 03:49:28.816382 137274321021824 utils.py:1231] [30700] epoch = 24.537628583939487
+I1130 03:49:28.816444 137274321021824 utils.py:1231] [30700] img/sec/core = 166.25707172257137
+I1130 03:49:28.816517 137274321021824 utils.py:1231] [30700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 53.954065325695545
+I1130 03:49:28.816574 137274321021824 utils.py:1231] [30700] core_hours = 53.954065325695545
+I1130 03:49:28.816662 137274321021824 train.py:125] NOTE: Steps:30700/112603 [27.3%]
+Walltime:2d5h59m (0s eval)
+ETA:5d23h57m
+Total train time:8d5h54m
+I1130 03:54:34.974072 137274321021824 utils.py:1231] [30750] l2_params = 331.4051308508235
+I1130 03:54:34.974316 137274321021824 utils.py:1231] [30750] train/loss = 2.728028506040573
+I1130 03:54:34.974448 137274321021824 utils.py:1231] [30750] l2_grads = 1.3262228965759277
+I1130 03:54:34.974536 137274321021824 utils.py:1231] [30750] lr = 0.0009024433252462924
+I1130 03:54:34.974601 137274321021824 utils.py:1231] [30750] uptime = 194664.336962256
+I1130 03:54:34.974686 137274321021824 utils.py:1231] [30750] examples_seen = 31488000.0
+I1130 03:54:34.974778 137274321021824 utils.py:1231] [30750] progress = 0.273083310391375
+I1130 03:54:34.974859 137274321021824 utils.py:1231] [30750] epoch = 24.577592148408442
+I1130 03:54:34.974944 137274321021824 utils.py:1231] [30750] img/sec/core = 167.23368179109627
+I1130 03:54:34.975008 137274321021824 utils.py:1231] [30750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.03910933160333
+I1130 03:54:34.975074 137274321021824 utils.py:1231] [30750] core_hours = 54.03910933160333
+I1130 03:54:34.975154 137274321021824 train.py:125] NOTE: Steps:30750/112603 [27.3%]
+Walltime:2d6h4m (0s eval)
+ETA:5d23h51m
+Total train time:8d5h53m
+I1130 03:59:46.761771 137274321021824 utils.py:1231] [30800] l2_params = 331.3864259716869
+I1130 03:59:46.761976 137274321021824 utils.py:1231] [30800] train/loss = 2.851132720708847
+I1130 03:59:46.762073 137274321021824 utils.py:1231] [30800] l2_grads = 1.3892076015472412
+I1130 03:59:46.762142 137274321021824 utils.py:1231] [30800] lr = 0.0009019886003030189
+I1130 03:59:46.762194 137274321021824 utils.py:1231] [30800] uptime = 194976.12455608498
+I1130 03:59:46.762253 137274321021824 utils.py:1231] [30800] examples_seen = 31539200.0
+I1130 03:59:46.762306 137274321021824 utils.py:1231] [30800] progress = 0.2735273482944504
+I1130 03:59:46.762359 137274321021824 utils.py:1231] [30800] epoch = 24.6175557128774
+I1130 03:59:46.762427 137274321021824 utils.py:1231] [30800] img/sec/core = 164.21435943369818
+I1130 03:59:46.762485 137274321021824 utils.py:1231] [30800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.12571699655583
+I1130 03:59:46.762537 137274321021824 utils.py:1231] [30800] core_hours = 54.12571699655583
+I1130 03:59:46.762613 137274321021824 train.py:125] NOTE: Steps:30800/112603 [27.4%]
+Walltime:2d6h9m (0s eval)
+ETA:5d23h45m
+Total train time:8d5h53m
+I1130 04:04:54.216179 137274321021824 utils.py:1231] [30850] l2_params = 331.3558124596801
+I1130 04:04:54.216398 137274321021824 utils.py:1231] [30850] train/loss = 2.643772840499878
+I1130 04:04:54.216498 137274321021824 utils.py:1231] [30850] l2_grads = 1.3105690479278564
+I1130 04:04:54.216561 137274321021824 utils.py:1231] [30850] lr = 0.0009015329331810322
+I1130 04:04:54.216613 137274321021824 utils.py:1231] [30850] uptime = 195283.578975323
+I1130 04:04:54.216667 137274321021824 utils.py:1231] [30850] examples_seen = 31590400.0
+I1130 04:04:54.216716 137274321021824 utils.py:1231] [30850] progress = 0.2739713861975258
+I1130 04:04:54.216767 137274321021824 utils.py:1231] [30850] epoch = 24.657519277346356
+I1130 04:04:54.216817 137274321021824 utils.py:1231] [30850] img/sec/core = 166.5287496172448
+I1130 04:04:54.216873 137274321021824 utils.py:1231] [30850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.21112100189971
+I1130 04:04:54.216928 137274321021824 utils.py:1231] [30850] core_hours = 54.21112100189971
+I1130 04:04:54.216989 137274321021824 train.py:125] NOTE: Steps:30850/112603 [27.4%]
+Walltime:2d6h14m (0s eval)
+ETA:5d23h40m
+Total train time:8d5h53m
+I1130 04:10:03.601325 137274321021824 utils.py:1231] [30900] l2_params = 331.354734470397
+I1130 04:10:03.601601 137274321021824 utils.py:1231] [30900] train/loss = 2.9434549510478973
+I1130 04:10:03.601772 137274321021824 utils.py:1231] [30900] l2_grads = 1.308390498161316
+I1130 04:10:03.601860 137274321021824 utils.py:1231] [30900] lr = 0.0009010763249483218
+I1130 04:10:03.601942 137274321021824 utils.py:1231] [30900] uptime = 195592.96429724002
+I1130 04:10:03.602010 137274321021824 utils.py:1231] [30900] examples_seen = 31641600.0
+I1130 04:10:03.602062 137274321021824 utils.py:1231] [30900] progress = 0.2744154241006012
+I1130 04:10:03.602114 137274321021824 utils.py:1231] [30900] epoch = 24.697482841815315
+I1130 04:10:03.602175 137274321021824 utils.py:1231] [30900] img/sec/core = 165.48942814337386
+I1130 04:10:03.602250 137274321021824 utils.py:1231] [30900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.297061369098884
+I1130 04:10:03.602301 137274321021824 utils.py:1231] [30900] core_hours = 54.297061369098884
+I1130 04:10:03.602363 137274321021824 train.py:125] NOTE: Steps:30900/112603 [27.4%]
+Walltime:2d6h19m (0s eval)
+ETA:5d23h34m
+Total train time:8d5h52m
+I1130 04:15:10.219392 137274321021824 utils.py:1231] [30950] l2_params = 331.32138503275087
+I1130 04:15:10.219624 137274321021824 utils.py:1231] [30950] train/loss = 2.828224778175354
+I1130 04:15:10.219772 137274321021824 utils.py:1231] [30950] l2_grads = 1.3981541395187378
+I1130 04:15:10.219887 137274321021824 utils.py:1231] [30950] lr = 0.0009006187766750842
+I1130 04:15:10.219948 137274321021824 utils.py:1231] [30950] uptime = 195899.582310241
+I1130 04:15:10.219999 137274321021824 utils.py:1231] [30950] examples_seen = 31692800.0
+I1130 04:15:10.220049 137274321021824 utils.py:1231] [30950] progress = 0.2748594620036766
+I1130 04:15:10.220095 137274321021824 utils.py:1231] [30950] epoch = 24.73744640628427
+I1130 04:15:10.220144 137274321021824 utils.py:1231] [30950] img/sec/core = 166.98301413829486
+I1130 04:15:10.220198 137274321021824 utils.py:1231] [30950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.382233039376935
+I1130 04:15:10.220245 137274321021824 utils.py:1231] [30950] core_hours = 54.382233039376935
+I1130 04:15:10.220303 137274321021824 train.py:125] NOTE: Steps:30950/112603 [27.5%]
+Walltime:2d6h24m (0s eval)
+ETA:5d23h28m
+Total train time:8d5h52m
+I1130 04:20:17.206681 137274321021824 utils.py:1231] [31000] l2_params = 331.28823943421014
+I1130 04:20:17.206908 137274321021824 utils.py:1231] [31000] train/loss = 3.0469507575035095
+I1130 04:20:17.207012 137274321021824 utils.py:1231] [31000] l2_grads = 1.2904772758483887
+I1130 04:20:17.207082 137274321021824 utils.py:1231] [31000] lr = 0.0009001602894337176
+I1130 04:20:17.207142 137274321021824 utils.py:1231] [31000] uptime = 196206.569502779
+I1130 04:20:17.207205 137274321021824 utils.py:1231] [31000] examples_seen = 31744000.0
+I1130 04:20:17.207262 137274321021824 utils.py:1231] [31000] progress = 0.275303499906752
+I1130 04:20:17.207318 137274321021824 utils.py:1231] [31000] epoch = 24.777409970753226
+I1130 04:20:17.207375 137274321021824 utils.py:1231] [31000] img/sec/core = 166.78220213913306
+I1130 04:20:17.207463 137274321021824 utils.py:1231] [31000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.467507259526386
+I1130 04:20:17.207532 137274321021824 utils.py:1231] [31000] core_hours = 54.467507259526386
+I1130 04:20:17.207620 137274321021824 train.py:125] NOTE: Steps:31000/112603 [27.5%]
+Walltime:2d6h30m (0s eval)
+ETA:5d23h23m
+Total train time:8d5h51m
+I1130 04:25:27.368006 137274321021824 utils.py:1231] [31050] l2_params = 331.3026753483527
+I1130 04:25:27.368273 137274321021824 utils.py:1231] [31050] train/loss = 4.359769105911255
+I1130 04:25:27.368405 137274321021824 utils.py:1231] [31050] l2_grads = 1.1179107427597046
+I1130 04:25:27.368518 137274321021824 utils.py:1231] [31050] lr = 0.000899700864298823
+I1130 04:25:27.368582 137274321021824 utils.py:1231] [31050] uptime = 196516.73094401503
+I1130 04:25:27.368641 137274321021824 utils.py:1231] [31050] examples_seen = 31795200.0
+I1130 04:25:27.368701 137274321021824 utils.py:1231] [31050] progress = 0.27574753780982747
+I1130 04:25:27.368758 137274321021824 utils.py:1231] [31050] epoch = 24.817373535222185
+I1130 04:25:27.368826 137274321021824 utils.py:1231] [31050] img/sec/core = 165.0753226963538
+I1130 04:25:27.368893 137274321021824 utils.py:1231] [31050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.55366321542528
+I1130 04:25:27.368954 137274321021824 utils.py:1231] [31050] core_hours = 54.55366321542528
+I1130 04:25:27.369015 137274321021824 train.py:125] NOTE: Steps:31050/112603 [27.6%]
+Walltime:2d6h35m (0s eval)
+ETA:5d23h17m
+Total train time:8d5h51m
+I1130 04:30:39.143726 137274321021824 utils.py:1231] [31100] l2_params = 331.2601304448526
+I1130 04:30:39.143989 137274321021824 utils.py:1231] [31100] train/loss = 3.253839910030365
+I1130 04:30:39.144121 137274321021824 utils.py:1231] [31100] l2_grads = 1.2293241024017334
+I1130 04:30:39.144241 137274321021824 utils.py:1231] [31100] lr = 0.0008992405023471972
+I1130 04:30:39.144341 137274321021824 utils.py:1231] [31100] uptime = 196828.50669984298
+I1130 04:30:39.144431 137274321021824 utils.py:1231] [31100] examples_seen = 31846400.0
+I1130 04:30:39.144522 137274321021824 utils.py:1231] [31100] progress = 0.27619157571290287
+I1130 04:30:39.144602 137274321021824 utils.py:1231] [31100] epoch = 24.85733709969114
+I1130 04:30:39.144676 137274321021824 utils.py:1231] [31100] img/sec/core = 164.22059458737542
+I1130 04:30:39.144761 137274321021824 utils.py:1231] [31100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.64026759204416
+I1130 04:30:39.144852 137274321021824 utils.py:1231] [31100] core_hours = 54.64026759204416
+I1130 04:30:39.144963 137274321021824 train.py:125] NOTE: Steps:31100/112603 [27.6%]
+Walltime:2d6h40m (0s eval)
+ETA:5d23h12m
+Total train time:8d5h50m
+I1130 04:35:48.379824 137274321021824 utils.py:1231] [31150] l2_params = 331.267777727559
+I1130 04:35:48.380033 137274321021824 utils.py:1231] [31150] train/loss = 2.884968250989914
+I1130 04:35:48.380141 137274321021824 utils.py:1231] [31150] l2_grads = 1.3134326934814453
+I1130 04:35:48.380209 137274321021824 utils.py:1231] [31150] lr = 0.000898779204657835
+I1130 04:35:48.380269 137274321021824 utils.py:1231] [31150] uptime = 197137.742630521
+I1130 04:35:48.380331 137274321021824 utils.py:1231] [31150] examples_seen = 31897600.0
+I1130 04:35:48.380387 137274321021824 utils.py:1231] [31150] progress = 0.27663561361597827
+I1130 04:35:48.380455 137274321021824 utils.py:1231] [31150] epoch = 24.8973006641601
+I1130 04:35:48.380511 137274321021824 utils.py:1231] [31150] img/sec/core = 165.56937574409375
+I1130 04:35:48.380574 137274321021824 utils.py:1231] [31150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.72616646167694
+I1130 04:35:48.380629 137274321021824 utils.py:1231] [31150] core_hours = 54.72616646167694
+I1130 04:35:48.380691 137274321021824 train.py:125] NOTE: Steps:31150/112603 [27.7%]
+Walltime:2d6h45m (0s eval)
+ETA:5d23h6m
+Total train time:8d5h50m
+I1130 04:41:00.152762 137274321021824 utils.py:1231] [31200] l2_params = 331.2778037216461
+I1130 04:41:00.152993 137274321021824 utils.py:1231] [31200] train/loss = 2.7238121032714844
+I1130 04:41:00.153093 137274321021824 utils.py:1231] [31200] l2_grads = 1.3261231184005737
+I1130 04:41:00.153186 137274321021824 utils.py:1231] [31200] lr = 0.0008983169723119233
+I1130 04:41:00.153247 137274321021824 utils.py:1231] [31200] uptime = 197449.515602794
+I1130 04:41:00.153316 137274321021824 utils.py:1231] [31200] examples_seen = 31948800.0
+I1130 04:41:00.153377 137274321021824 utils.py:1231] [31200] progress = 0.27707965151905367
+I1130 04:41:00.153442 137274321021824 utils.py:1231] [31200] epoch = 24.937264228629054
+I1130 04:41:00.153505 137274321021824 utils.py:1231] [31200] img/sec/core = 164.22206077301456
+I1130 04:41:00.153585 137274321021824 utils.py:1231] [31200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.8127700650861
+I1130 04:41:00.153636 137274321021824 utils.py:1231] [31200] core_hours = 54.8127700650861
+I1130 04:41:00.153720 137274321021824 train.py:125] NOTE: Steps:31200/112603 [27.7%]
+Walltime:2d6h50m (0s eval)
+ETA:5d23h1m
+Total train time:8d5h50m
+I1130 04:46:08.118723 137274321021824 utils.py:1231] [31250] l2_params = 331.24253365910124
+I1130 04:46:08.119049 137274321021824 utils.py:1231] [31250] train/loss = 5.089210569858551
+I1130 04:46:08.119299 137274321021824 utils.py:1231] [31250] l2_grads = 0.9953965544700623
+I1130 04:46:08.119426 137274321021824 utils.py:1231] [31250] lr = 0.00089785380639284
+I1130 04:46:08.119526 137274321021824 utils.py:1231] [31250] uptime = 197757.481879148
+I1130 04:46:08.119633 137274321021824 utils.py:1231] [31250] examples_seen = 32000000.0
+I1130 04:46:08.119746 137274321021824 utils.py:1231] [31250] progress = 0.27752368942212907
+I1130 04:46:08.119828 137274321021824 utils.py:1231] [31250] epoch = 24.977227793098013
+I1130 04:46:08.119917 137274321021824 utils.py:1231] [31250] img/sec/core = 166.25196955378644
+I1130 04:46:08.120003 137274321021824 utils.py:1231] [31250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.89831625296222
+I1130 04:46:08.120085 137274321021824 utils.py:1231] [31250] core_hours = 54.89831625296222
+I1130 04:46:08.120189 137274321021824 train.py:125] NOTE: Steps:31250/112603 [27.8%]
+Walltime:2d6h55m (0s eval)
+ETA:5d22h55m
+Total train time:8d5h49m
+I1130 04:51:17.963909 137274321021824 utils.py:1231] [31300] l2_params = 331.1839878547909
+I1130 04:51:17.964162 137274321021824 utils.py:1231] [31300] train/loss = 2.8013043999671936
+I1130 04:51:17.964280 137274321021824 utils.py:1231] [31300] l2_grads = 1.3824365139007568
+I1130 04:51:17.964373 137274321021824 utils.py:1231] [31300] lr = 0.0008973897079861508
+I1130 04:51:17.964452 137274321021824 utils.py:1231] [31300] uptime = 198067.326813403
+I1130 04:51:17.964531 137274321021824 utils.py:1231] [31300] examples_seen = 32051200.0
+I1130 04:51:17.964592 137274321021824 utils.py:1231] [31300] progress = 0.27796772732520447
+I1130 04:51:17.964665 137274321021824 utils.py:1231] [31300] epoch = 25.01719135756697
+I1130 04:51:17.964748 137274321021824 utils.py:1231] [31300] img/sec/core = 165.24394734129365
+I1130 04:51:17.964837 137274321021824 utils.py:1231] [31300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 54.98438429025528
+I1130 04:51:17.964921 137274321021824 utils.py:1231] [31300] core_hours = 54.98438429025528
+I1130 04:51:17.965012 137274321021824 train.py:125] NOTE: Steps:31300/112603 [27.8%]
+Walltime:2d7h1m (0s eval)
+ETA:5d22h49m
+Total train time:8d5h49m
+I1130 04:56:26.299167 137274321021824 utils.py:1231] [31350] l2_params = 331.1822230692896
+I1130 04:56:26.299419 137274321021824 utils.py:1231] [31350] train/loss = 2.746083587408066
+I1130 04:56:26.299558 137274321021824 utils.py:1231] [31350] l2_grads = 1.3371707201004028
+I1130 04:56:26.299646 137274321021824 utils.py:1231] [31350] lr = 0.0008969246781796072
+I1130 04:56:26.299715 137274321021824 utils.py:1231] [31350] uptime = 198375.662076396
+I1130 04:56:26.299775 137274321021824 utils.py:1231] [31350] examples_seen = 32102400.0
+I1130 04:56:26.299839 137274321021824 utils.py:1231] [31350] progress = 0.2784117652282799
+I1130 04:56:26.299915 137274321021824 utils.py:1231] [31350] epoch = 25.057154922035924
+I1130 04:56:26.299976 137274321021824 utils.py:1231] [31350] img/sec/core = 166.05301483522547
+I1130 04:56:26.300038 137274321021824 utils.py:1231] [31350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.070032974419995
+I1130 04:56:26.300088 137274321021824 utils.py:1231] [31350] core_hours = 55.070032974419995
+I1130 04:56:26.300160 137274321021824 train.py:125] NOTE: Steps:31350/112603 [27.8%]
+Walltime:2d7h6m (0s eval)
+ETA:5d22h44m
+Total train time:8d5h48m
+I1130 05:01:34.851192 137274321021824 utils.py:1231] [31400] l2_params = 331.18713151598473
+I1130 05:01:34.851432 137274321021824 utils.py:1231] [31400] train/loss = 4.335947215557098
+I1130 05:01:34.851563 137274321021824 utils.py:1231] [31400] l2_grads = 1.0474603176116943
+I1130 05:01:34.851653 137274321021824 utils.py:1231] [31400] lr = 0.0008964587180631436
+I1130 05:01:34.851731 137274321021824 utils.py:1231] [31400] uptime = 198684.21409305
+I1130 05:01:34.851804 137274321021824 utils.py:1231] [31400] examples_seen = 32153600.0
+I1130 05:01:34.851863 137274321021824 utils.py:1231] [31400] progress = 0.2788558031313553
+I1130 05:01:34.851924 137274321021824 utils.py:1231] [31400] epoch = 25.097118486504883
+I1130 05:01:34.851999 137274321021824 utils.py:1231] [31400] img/sec/core = 165.9363648153207
+I1130 05:01:34.852059 137274321021824 utils.py:1231] [31400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.155741867935
+I1130 05:01:34.852110 137274321021824 utils.py:1231] [31400] core_hours = 55.155741867935
+I1130 05:01:34.852189 137274321021824 train.py:125] NOTE: Steps:31400/112603 [27.9%]
+Walltime:2d7h11m (0s eval)
+ETA:5d22h38m
+Total train time:8d5h48m
+I1130 05:06:43.058660 137274321021824 utils.py:1231] [31450] l2_params = 331.1733758975622
+I1130 05:06:43.058863 137274321021824 utils.py:1231] [31450] train/loss = 3.453758627176285
+I1130 05:06:43.058972 137274321021824 utils.py:1231] [31450] l2_grads = 1.1704152822494507
+I1130 05:06:43.059032 137274321021824 utils.py:1231] [31450] lr = 0.0008959918287288739
+I1130 05:06:43.059082 137274321021824 utils.py:1231] [31450] uptime = 198992.42144475598
+I1130 05:06:43.059135 137274321021824 utils.py:1231] [31450] examples_seen = 32204800.0
+I1130 05:06:43.059189 137274321021824 utils.py:1231] [31450] progress = 0.2792998410344307
+I1130 05:06:43.059237 137274321021824 utils.py:1231] [31450] epoch = 25.137082050973838
+I1130 05:06:43.059287 137274321021824 utils.py:1231] [31450] img/sec/core = 166.12192965741255
+I1130 05:06:43.059343 137274321021824 utils.py:1231] [31450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.241355021186656
+I1130 05:06:43.059393 137274321021824 utils.py:1231] [31450] core_hours = 55.241355021186656
+I1130 05:06:43.059459 137274321021824 train.py:125] NOTE: Steps:31450/112603 [27.9%]
+Walltime:2d7h16m (0s eval)
+ETA:5d22h33m
+Total train time:8d5h47m
+I1130 05:11:50.810758 137274321021824 utils.py:1231] [31500] l2_params = 331.15293748033463
+I1130 05:11:50.810978 137274321021824 utils.py:1231] [31500] train/loss = 2.6629887521266937
+I1130 05:11:50.811078 137274321021824 utils.py:1231] [31500] l2_grads = 1.2785760164260864
+I1130 05:11:50.811139 137274321021824 utils.py:1231] [31500] lr = 0.000895524011271092
+I1130 05:11:50.811199 137274321021824 utils.py:1231] [31500] uptime = 199300.17356090102
+I1130 05:11:50.811252 137274321021824 utils.py:1231] [31500] examples_seen = 32256000.0
+I1130 05:11:50.811302 137274321021824 utils.py:1231] [31500] progress = 0.27974387893750613
+I1130 05:11:50.811350 137274321021824 utils.py:1231] [31500] epoch = 25.177045615442797
+I1130 05:11:50.811400 137274321021824 utils.py:1231] [31500] img/sec/core = 166.36766187457994
+I1130 05:11:50.811461 137274321021824 utils.py:1231] [31500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.32684172011582
+I1130 05:11:50.811511 137274321021824 utils.py:1231] [31500] core_hours = 55.32684172011582
+I1130 05:11:50.811572 137274321021824 train.py:125] NOTE: Steps:31500/112603 [28.0%]
+Walltime:2d7h21m (0s eval)
+ETA:5d22h27m
+Total train time:8d5h47m
+I1130 05:17:02.591279 137274321021824 utils.py:1231] [31550] l2_params = 331.1634748274068
+I1130 05:17:02.591502 137274321021824 utils.py:1231] [31550] train/loss = 2.799906462430954
+I1130 05:17:02.591599 137274321021824 utils.py:1231] [31550] l2_grads = 1.345585823059082
+I1130 05:17:02.591658 137274321021824 utils.py:1231] [31550] lr = 0.0008950552667862644
+I1130 05:17:02.591706 137274321021824 utils.py:1231] [31550] uptime = 199611.954069275
+I1130 05:17:02.591767 137274321021824 utils.py:1231] [31550] examples_seen = 32307200.0
+I1130 05:17:02.591815 137274321021824 utils.py:1231] [31550] progress = 0.28018791684058153
+I1130 05:17:02.591862 137274321021824 utils.py:1231] [31550] epoch = 25.217009179911752
+I1130 05:17:02.591917 137274321021824 utils.py:1231] [31550] img/sec/core = 164.21809133296065
+I1130 05:17:02.591971 137274321021824 utils.py:1231] [31550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.413447416886385
+I1130 05:17:02.592017 137274321021824 utils.py:1231] [31550] core_hours = 55.413447416886385
+I1130 05:17:02.592072 137274321021824 train.py:125] NOTE: Steps:31550/112603 [28.0%]
+Walltime:2d7h26m (0s eval)
+ETA:5d22h22m
+Total train time:8d5h47m
+I1130 05:22:14.367631 137274321021824 utils.py:1231] [31600] l2_params = 331.12169075293656
+I1130 05:22:14.367834 137274321021824 utils.py:1231] [31600] train/loss = 2.692262649536133
+I1130 05:22:14.367942 137274321021824 utils.py:1231] [31600] l2_grads = 1.3117351531982422
+I1130 05:22:14.368004 137274321021824 utils.py:1231] [31600] lr = 0.000894585596373033
+I1130 05:22:14.368067 137274321021824 utils.py:1231] [31600] uptime = 199923.730429739
+I1130 05:22:14.368125 137274321021824 utils.py:1231] [31600] examples_seen = 32358400.0
+I1130 05:22:14.368169 137274321021824 utils.py:1231] [31600] progress = 0.28063195474365693
+I1130 05:22:14.368212 137274321021824 utils.py:1231] [31600] epoch = 25.256972744380707
+I1130 05:22:14.368258 137274321021824 utils.py:1231] [31600] img/sec/core = 164.22027611010412
+I1130 05:22:14.368309 137274321021824 utils.py:1231] [31600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.50005196145972
+I1130 05:22:14.368355 137274321021824 utils.py:1231] [31600] core_hours = 55.50005196145972
+I1130 05:22:14.368409 137274321021824 train.py:125] NOTE: Steps:31600/112603 [28.1%]
+Walltime:2d7h32m (0s eval)
+ETA:5d22h16m
+Total train time:8d5h46m
+I1130 05:27:24.767926 137274321021824 utils.py:1231] [31650] l2_params = 331.1393745041276
+I1130 05:27:24.768169 137274321021824 utils.py:1231] [31650] train/loss = 5.213304936885834
+I1130 05:27:24.768285 137274321021824 utils.py:1231] [31650] l2_grads = 1.0749173164367676
+I1130 05:27:24.768362 137274321021824 utils.py:1231] [31650] lr = 0.0008941150011322088
+I1130 05:27:24.768446 137274321021824 utils.py:1231] [31650] uptime = 200234.130801739
+I1130 05:27:24.768517 137274321021824 utils.py:1231] [31650] examples_seen = 32409600.0
+I1130 05:27:24.768573 137274321021824 utils.py:1231] [31650] progress = 0.28107599264673233
+I1130 05:27:24.768625 137274321021824 utils.py:1231] [31650] epoch = 25.296936308849666
+I1130 05:27:24.768681 137274321021824 utils.py:1231] [31650] img/sec/core = 164.94825592541298
+I1130 05:27:24.768736 137274321021824 utils.py:1231] [31650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.58627428701527
+I1130 05:27:24.768793 137274321021824 utils.py:1231] [31650] core_hours = 55.58627428701527
+I1130 05:27:24.768857 137274321021824 train.py:125] NOTE: Steps:31650/112603 [28.1%]
+Walltime:2d7h37m (0s eval)
+ETA:5d22h11m
+Total train time:8d5h46m
+I1130 05:32:31.488981 137274321021824 utils.py:1231] [31700] l2_params = 331.1457777149365
+I1130 05:32:31.489198 137274321021824 utils.py:1231] [31700] train/loss = 2.58091801404953
+I1130 05:32:31.489339 137274321021824 utils.py:1231] [31700] l2_grads = 1.357260823249817
+I1130 05:32:31.489420 137274321021824 utils.py:1231] [31700] lr = 0.0008936434821667703
+I1130 05:32:31.489485 137274321021824 utils.py:1231] [31700] uptime = 200540.851847184
+I1130 05:32:31.489544 137274321021824 utils.py:1231] [31700] examples_seen = 32460800.0
+I1130 05:32:31.489592 137274321021824 utils.py:1231] [31700] progress = 0.28152003054980773
+I1130 05:32:31.489639 137274321021824 utils.py:1231] [31700] epoch = 25.33689987331862
+I1130 05:32:31.489689 137274321021824 utils.py:1231] [31700] img/sec/core = 166.92692190625726
+I1130 05:32:31.489744 137274321021824 utils.py:1231] [31700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.67147457741666
+I1130 05:32:31.489795 137274321021824 utils.py:1231] [31700] core_hours = 55.67147457741666
+I1130 05:32:31.489853 137274321021824 train.py:125] NOTE: Steps:31700/112603 [28.2%]
+Walltime:2d7h42m (0s eval)
+ETA:5d22h5m
+Total train time:8d5h45m
+I1130 05:37:39.400417 137274321021824 utils.py:1231] [31750] l2_params = 331.1472077021178
+I1130 05:37:39.400690 137274321021824 utils.py:1231] [31750] train/loss = 5.41215443611145
+I1130 05:37:39.400833 137274321021824 utils.py:1231] [31750] l2_grads = 1.1862152814865112
+I1130 05:37:39.400935 137274321021824 utils.py:1231] [31750] lr = 0.0008931710405818614
+I1130 05:37:39.400995 137274321021824 utils.py:1231] [31750] uptime = 200848.76335683902
+I1130 05:37:39.401058 137274321021824 utils.py:1231] [31750] examples_seen = 32512000.0
+I1130 05:37:39.401139 137274321021824 utils.py:1231] [31750] progress = 0.28196406845288313
+I1130 05:37:39.401190 137274321021824 utils.py:1231] [31750] epoch = 25.37686343778758
+I1130 05:37:39.401251 137274321021824 utils.py:1231] [31750] img/sec/core = 166.2815399702522
+I1130 05:37:39.401313 137274321021824 utils.py:1231] [31750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.75700555232083
+I1130 05:37:39.401375 137274321021824 utils.py:1231] [31750] core_hours = 55.75700555232083
+I1130 05:37:39.401444 137274321021824 train.py:125] NOTE: Steps:31750/112603 [28.2%]
+Walltime:2d7h47m (0s eval)
+ETA:5d21h59m
+Total train time:8d5h45m
+I1130 05:42:45.592505 137274321021824 utils.py:1231] [31800] l2_params = 331.1010511492522
+I1130 05:42:45.592761 137274321021824 utils.py:1231] [31800] train/loss = 2.8544051945209503
+I1130 05:42:45.592875 137274321021824 utils.py:1231] [31800] l2_grads = 1.3750741481781006
+I1130 05:42:45.592952 137274321021824 utils.py:1231] [31800] lr = 0.0008926976774847883
+I1130 05:42:45.593011 137274321021824 utils.py:1231] [31800] uptime = 201154.95537389402
+I1130 05:42:45.593063 137274321021824 utils.py:1231] [31800] examples_seen = 32563200.0
+I1130 05:42:45.593111 137274321021824 utils.py:1231] [31800] progress = 0.28240810635595853
+I1130 05:42:45.593158 137274321021824 utils.py:1231] [31800] epoch = 25.416827002256536
+I1130 05:42:45.593207 137274321021824 utils.py:1231] [31800] img/sec/core = 167.2153326936774
+I1130 05:42:45.593261 137274321021824 utils.py:1231] [31800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.842058890391655
+I1130 05:42:45.593311 137274321021824 utils.py:1231] [31800] core_hours = 55.842058890391655
+I1130 05:42:45.593369 137274321021824 train.py:125] NOTE: Steps:31800/112603 [28.2%]
+Walltime:2d7h52m (0s eval)
+ETA:5d21h54m
+Total train time:8d5h44m
+I1130 05:47:50.551290 137274321021824 utils.py:1231] [31850] l2_params = 331.10715876013046
+I1130 05:47:50.551491 137274321021824 utils.py:1231] [31850] train/loss = 2.7841385900974274
+I1130 05:47:50.551586 137274321021824 utils.py:1231] [31850] l2_grads = 1.3481030464172363
+I1130 05:47:50.551653 137274321021824 utils.py:1231] [31850] lr = 0.0008922233939850159
+I1130 05:47:50.551703 137274321021824 utils.py:1231] [31850] uptime = 201459.914065297
+I1130 05:47:50.551754 137274321021824 utils.py:1231] [31850] examples_seen = 32614400.0
+I1130 05:47:50.551802 137274321021824 utils.py:1231] [31850] progress = 0.28285214425903393
+I1130 05:47:50.551851 137274321021824 utils.py:1231] [31850] epoch = 25.456790566725495
+I1130 05:47:50.551909 137274321021824 utils.py:1231] [31850] img/sec/core = 167.8915913642165
+I1130 05:47:50.551964 137274321021824 utils.py:1231] [31850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 55.926769638003606
+I1130 05:47:50.552011 137274321021824 utils.py:1231] [31850] core_hours = 55.926769638003606
+I1130 05:47:50.552071 137274321021824 train.py:125] NOTE: Steps:31850/112603 [28.3%]
+Walltime:2d7h57m (0s eval)
+ETA:5d21h48m
+Total train time:8d5h44m
+I1130 05:52:55.052281 137274321021824 utils.py:1231] [31900] l2_params = 331.08677469397617
+I1130 05:52:55.052554 137274321021824 utils.py:1231] [31900] train/loss = 2.824831873178482
+I1130 05:52:55.052732 137274321021824 utils.py:1231] [31900] l2_grads = 1.3760968446731567
+I1130 05:52:55.052805 137274321021824 utils.py:1231] [31900] lr = 0.000891748191194167
+I1130 05:52:55.052868 137274321021824 utils.py:1231] [31900] uptime = 201764.41522509098
+I1130 05:52:55.052939 137274321021824 utils.py:1231] [31900] examples_seen = 32665600.0
+I1130 05:52:55.052989 137274321021824 utils.py:1231] [31900] progress = 0.28329618216210933
+I1130 05:52:55.053038 137274321021824 utils.py:1231] [31900] epoch = 25.49675413119445
+I1130 05:52:55.053088 137274321021824 utils.py:1231] [31900] img/sec/core = 168.14385874470756
+I1130 05:52:55.053144 137274321021824 utils.py:1231] [31900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.011353293501934
+I1130 05:52:55.053193 137274321021824 utils.py:1231] [31900] core_hours = 56.011353293501934
+I1130 05:52:55.053253 137274321021824 train.py:125] NOTE: Steps:31900/112603 [28.3%]
+Walltime:2d8h2m (0s eval)
+ETA:5d21h42m
+Total train time:8d5h43m
+I1130 05:57:59.655720 137274321021824 utils.py:1231] [31950] l2_params = 331.06450747528896
+I1130 05:57:59.655986 137274321021824 utils.py:1231] [31950] train/loss = 5.33327579498291
+I1130 05:57:59.656218 137274321021824 utils.py:1231] [31950] l2_grads = 1.06822669506073
+I1130 05:57:59.656322 137274321021824 utils.py:1231] [31950] lr = 0.0008912720702260207
+I1130 05:57:59.656406 137274321021824 utils.py:1231] [31950] uptime = 202069.01876384503
+I1130 05:57:59.656519 137274321021824 utils.py:1231] [31950] examples_seen = 32716800.0
+I1130 05:57:59.656607 137274321021824 utils.py:1231] [31950] progress = 0.2837402200651848
+I1130 05:57:59.656710 137274321021824 utils.py:1231] [31950] epoch = 25.536717695663405
+I1130 05:57:59.656796 137274321021824 utils.py:1231] [31950] img/sec/core = 168.08734464948714
+I1130 05:57:59.656904 137274321021824 utils.py:1231] [31950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.09596538760027
+I1130 05:57:59.657003 137274321021824 utils.py:1231] [31950] core_hours = 56.09596538760027
+I1130 05:57:59.657106 137274321021824 train.py:125] NOTE: Steps:31950/112603 [28.4%]
+Walltime:2d8h7m (0s eval)
+ETA:5d21h36m
+Total train time:8d5h42m
+I1130 06:03:06.091436 137274321021824 utils.py:1231] [32000] l2_params = 331.00296389711025
+I1130 06:03:06.091648 137274321021824 utils.py:1231] [32000] train/loss = 4.816031277179718
+I1130 06:03:06.091767 137274321021824 utils.py:1231] [32000] l2_grads = 1.0446780920028687
+I1130 06:03:06.091842 137274321021824 utils.py:1231] [32000] lr = 0.0008907950321965062
+I1130 06:03:06.091927 137274321021824 utils.py:1231] [32000] uptime = 202375.45428457198
+I1130 06:03:06.091989 137274321021824 utils.py:1231] [32000] examples_seen = 32768000.0
+I1130 06:03:06.092044 137274321021824 utils.py:1231] [32000] progress = 0.2841842579682602
+I1130 06:03:06.092100 137274321021824 utils.py:1231] [32000] epoch = 25.576681260132364
+I1130 06:03:06.092158 137274321021824 utils.py:1231] [32000] img/sec/core = 167.0824579295957
+I1130 06:03:06.092216 137274321021824 utils.py:1231] [32000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.18108636557999
+I1130 06:03:06.092271 137274321021824 utils.py:1231] [32000] core_hours = 56.18108636557999
+I1130 06:03:06.092337 137274321021824 train.py:125] NOTE: Steps:32000/112603 [28.4%]
+Walltime:2d8h12m (0s eval)
+ETA:5d21h31m
+Total train time:8d5h42m
+I1130 06:08:14.068912 137274321021824 utils.py:1231] [32050] l2_params = 330.96143508083827
+I1130 06:08:14.069136 137274321021824 utils.py:1231] [32050] train/loss = 5.377943515777588
+I1130 06:08:14.069226 137274321021824 utils.py:1231] [32050] l2_grads = 1.169028401374817
+I1130 06:08:14.069293 137274321021824 utils.py:1231] [32050] lr = 0.0008903170782237021
+I1130 06:08:14.069341 137274321021824 utils.py:1231] [32050] uptime = 202683.43170327903
+I1130 06:08:14.069389 137274321021824 utils.py:1231] [32050] examples_seen = 32819200.0
+I1130 06:08:14.069437 137274321021824 utils.py:1231] [32050] progress = 0.2846282958713356
+I1130 06:08:14.069482 137274321021824 utils.py:1231] [32050] epoch = 25.61664482460132
+I1130 06:08:14.069531 137274321021824 utils.py:1231] [32050] img/sec/core = 166.24595470328197
+I1130 06:08:14.069584 137274321021824 utils.py:1231] [32050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.266635648554164
+I1130 06:08:14.069630 137274321021824 utils.py:1231] [32050] core_hours = 56.266635648554164
+I1130 06:08:14.069687 137274321021824 train.py:125] NOTE: Steps:32050/112603 [28.5%]
+Walltime:2d8h18m (0s eval)
+ETA:5d21h25m
+Total train time:8d5h41m
+I1130 06:13:22.843026 137274321021824 utils.py:1231] [32100] l2_params = 330.95522259670224
+I1130 06:13:22.843330 137274321021824 utils.py:1231] [32100] train/loss = 2.7316154837608337
+I1130 06:13:22.843556 137274321021824 utils.py:1231] [32100] l2_grads = 1.293140172958374
+I1130 06:13:22.843646 137274321021824 utils.py:1231] [32100] lr = 0.0008898382094278355
+I1130 06:13:22.843726 137274321021824 utils.py:1231] [32100] uptime = 202992.20608494902
+I1130 06:13:22.843819 137274321021824 utils.py:1231] [32100] examples_seen = 32870400.0
+I1130 06:13:22.843900 137274321021824 utils.py:1231] [32100] progress = 0.285072333774411
+I1130 06:13:22.844002 137274321021824 utils.py:1231] [32100] epoch = 25.65660838907028
+I1130 06:13:22.844098 137274321021824 utils.py:1231] [32100] img/sec/core = 165.81686512684985
+I1130 06:13:22.844184 137274321021824 utils.py:1231] [32100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.35240631012916
+I1130 06:13:22.844260 137274321021824 utils.py:1231] [32100] core_hours = 56.35240631012916
+I1130 06:13:22.844368 137274321021824 train.py:125] NOTE: Steps:32100/112603 [28.5%]
+Walltime:2d8h23m (0s eval)
+ETA:5d21h20m
+Total train time:8d5h41m
+I1130 06:18:29.847104 137274321021824 utils.py:1231] [32150] l2_params = 330.91199931486284
+I1130 06:18:29.847363 137274321021824 utils.py:1231] [32150] train/loss = 2.8729051053524017
+I1130 06:18:29.847485 137274321021824 utils.py:1231] [32150] l2_grads = 1.3143357038497925
+I1130 06:18:29.847576 137274321021824 utils.py:1231] [32150] lr = 0.0008893584269312756
+I1130 06:18:29.847639 137274321021824 utils.py:1231] [32150] uptime = 203299.210000057
+I1130 06:18:29.847698 137274321021824 utils.py:1231] [32150] examples_seen = 32921600.0
+I1130 06:18:29.847753 137274321021824 utils.py:1231] [32150] progress = 0.2855163716774864
+I1130 06:18:29.847812 137274321021824 utils.py:1231] [32150] epoch = 25.696571953539234
+I1130 06:18:29.847867 137274321021824 utils.py:1231] [32150] img/sec/core = 166.7731174763398
+I1130 06:18:29.847941 137274321021824 utils.py:1231] [32150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.43768517543694
+I1130 06:18:29.848003 137274321021824 utils.py:1231] [32150] core_hours = 56.43768517543694
+I1130 06:18:29.848083 137274321021824 train.py:125] NOTE: Steps:32150/112603 [28.6%]
+Walltime:2d8h28m (0s eval)
+ETA:5d21h14m
+Total train time:8d5h40m
+I1130 06:23:37.158623 137274321021824 utils.py:1231] [32200] l2_params = 330.89758571118386
+I1130 06:23:37.158878 137274321021824 utils.py:1231] [32200] train/loss = 3.4554033875465393
+I1130 06:23:37.159015 137274321021824 utils.py:1231] [32200] l2_grads = 1.241214394569397
+I1130 06:23:37.159102 137274321021824 utils.py:1231] [32200] lr = 0.0008888777318585339
+I1130 06:23:37.159169 137274321021824 utils.py:1231] [32200] uptime = 203606.521530509
+I1130 06:23:37.159234 137274321021824 utils.py:1231] [32200] examples_seen = 32972800.0
+I1130 06:23:37.159297 137274321021824 utils.py:1231] [32200] progress = 0.2859604095805618
+I1130 06:23:37.159358 137274321021824 utils.py:1231] [32200] epoch = 25.736535518008193
+I1130 06:23:37.159415 137274321021824 utils.py:1231] [32200] img/sec/core = 166.60617948402725
+I1130 06:23:37.159481 137274321021824 utils.py:1231] [32200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.52304948945139
+I1130 06:23:37.159536 137274321021824 utils.py:1231] [32200] core_hours = 56.52304948945139
+I1130 06:23:37.159613 137274321021824 train.py:125] NOTE: Steps:32200/112603 [28.6%]
+Walltime:2d8h33m (0s eval)
+ETA:5d21h8m
+Total train time:8d5h40m
+I1130 06:28:43.813928 137274321021824 utils.py:1231] [32250] l2_params = 330.89816292661413
+I1130 06:28:43.814201 137274321021824 utils.py:1231] [32250] train/loss = 4.226140022277832
+I1130 06:28:43.814358 137274321021824 utils.py:1231] [32250] l2_grads = 1.023229718208313
+I1130 06:28:43.814436 137274321021824 utils.py:1231] [32250] lr = 0.0008883961253362617
+I1130 06:28:43.814500 137274321021824 utils.py:1231] [32250] uptime = 203913.17686225602
+I1130 06:28:43.814552 137274321021824 utils.py:1231] [32250] examples_seen = 33024000.0
+I1130 06:28:43.814604 137274321021824 utils.py:1231] [32250] progress = 0.2864044474836372
+I1130 06:28:43.814656 137274321021824 utils.py:1231] [32250] epoch = 25.776499082477148
+I1130 06:28:43.814719 137274321021824 utils.py:1231] [32250] img/sec/core = 166.96269296319414
+I1130 06:28:43.814775 137274321021824 utils.py:1231] [32250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.608231526047774
+I1130 06:28:43.814834 137274321021824 utils.py:1231] [32250] core_hours = 56.608231526047774
+I1130 06:28:43.814919 137274321021824 train.py:125] NOTE: Steps:32250/112603 [28.6%]
+Walltime:2d8h38m (0s eval)
+ETA:5d21h3m
+Total train time:8d5h39m
+I1130 06:33:51.144031 137274321021824 utils.py:1231] [32300] l2_params = 330.8276312641554
+I1130 06:33:51.144230 137274321021824 utils.py:1231] [32300] train/loss = 3.017395853996277
+I1130 06:33:51.144331 137274321021824 utils.py:1231] [32300] l2_grads = 1.2357022762298584
+I1130 06:33:51.144389 137274321021824 utils.py:1231] [32300] lr = 0.0008879136084932451
+I1130 06:33:51.144439 137274321021824 utils.py:1231] [32300] uptime = 204220.50680120703
+I1130 06:33:51.144490 137274321021824 utils.py:1231] [32300] examples_seen = 33075200.0
+I1130 06:33:51.144541 137274321021824 utils.py:1231] [32300] progress = 0.2868484853867126
+I1130 06:33:51.144588 137274321021824 utils.py:1231] [32300] epoch = 25.816462646946103
+I1130 06:33:51.144636 137274321021824 utils.py:1231] [32300] img/sec/core = 166.59620007981752
+I1130 06:33:51.144690 137274321021824 utils.py:1231] [32300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.693600953534165
+I1130 06:33:51.144738 137274321021824 utils.py:1231] [32300] core_hours = 56.693600953534165
+I1130 06:33:51.144797 137274321021824 train.py:125] NOTE: Steps:32300/112603 [28.7%]
+Walltime:2d8h43m (0s eval)
+ETA:5d20h57m
+Total train time:8d5h39m
+I1130 06:38:58.207838 137274321021824 utils.py:1231] [32350] l2_params = 330.83809609562127
+I1130 06:38:58.208124 137274321021824 utils.py:1231] [32350] train/loss = 4.689834296703339
+I1130 06:38:58.208364 137274321021824 utils.py:1231] [32350] l2_grads = 1.1036878824234009
+I1130 06:38:58.208484 137274321021824 utils.py:1231] [32350] lr = 0.0008874301824604047
+I1130 06:38:58.208575 137274321021824 utils.py:1231] [32350] uptime = 204527.57093227003
+I1130 06:38:58.208667 137274321021824 utils.py:1231] [32350] examples_seen = 33126400.0
+I1130 06:38:58.208778 137274321021824 utils.py:1231] [32350] progress = 0.287292523289788
+I1130 06:38:58.208861 137274321021824 utils.py:1231] [32350] epoch = 25.856426211415062
+I1130 06:38:58.208954 137274321021824 utils.py:1231] [32350] img/sec/core = 166.7404128992685
+I1130 06:38:58.209034 137274321021824 utils.py:1231] [32350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.7788965454961
+I1130 06:38:58.209111 137274321021824 utils.py:1231] [32350] core_hours = 56.7788965454961
+I1130 06:38:58.209209 137274321021824 train.py:125] NOTE: Steps:32350/112603 [28.7%]
+Walltime:2d8h48m (0s eval)
+ETA:5d20h51m
+Total train time:8d5h38m
+I1130 06:44:08.745333 137274321021824 utils.py:1231] [32400] l2_params = 330.79634070942257
+I1130 06:44:08.745596 137274321021824 utils.py:1231] [32400] train/loss = 2.4776268005371094
+I1130 06:44:08.745727 137274321021824 utils.py:1231] [32400] l2_grads = 1.3224387168884277
+I1130 06:44:08.745815 137274321021824 utils.py:1231] [32400] lr = 0.0008869458483707925
+I1130 06:44:08.745879 137274321021824 utils.py:1231] [32400] uptime = 204838.10824021703
+I1130 06:44:08.745948 137274321021824 utils.py:1231] [32400] examples_seen = 33177600.0
+I1130 06:44:08.746003 137274321021824 utils.py:1231] [32400] progress = 0.28773656119286345
+I1130 06:44:08.746059 137274321021824 utils.py:1231] [32400] epoch = 25.896389775884018
+I1130 06:44:08.746116 137274321021824 utils.py:1231] [32400] img/sec/core = 164.87551959051004
+I1130 06:44:08.746176 137274321021824 utils.py:1231] [32400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.86515690881472
+I1130 06:44:08.746226 137274321021824 utils.py:1231] [32400] core_hours = 56.86515690881472
+I1130 06:44:08.746290 137274321021824 train.py:125] NOTE: Steps:32400/112603 [28.8%]
+Walltime:2d8h53m (0s eval)
+ETA:5d20h46m
+Total train time:8d5h38m
+I1130 06:49:17.930286 137274321021824 utils.py:1231] [32450] l2_params = 330.7971320172681
+I1130 06:49:17.930520 137274321021824 utils.py:1231] [32450] train/loss = 2.874951511621475
+I1130 06:49:17.930623 137274321021824 utils.py:1231] [32450] l2_grads = 1.3721907138824463
+I1130 06:49:17.930693 137274321021824 utils.py:1231] [32450] lr = 0.0008864606073595869
+I1130 06:49:17.930762 137274321021824 utils.py:1231] [32450] uptime = 205147.29311378498
+I1130 06:49:17.930836 137274321021824 utils.py:1231] [32450] examples_seen = 33228800.0
+I1130 06:49:17.930902 137274321021824 utils.py:1231] [32450] progress = 0.28818059909593885
+I1130 06:49:17.930958 137274321021824 utils.py:1231] [32450] epoch = 25.936353340352976
+I1130 06:49:17.931015 137274321021824 utils.py:1231] [32450] img/sec/core = 165.59671697115706
+I1130 06:49:17.931075 137274321021824 utils.py:1231] [32450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 56.95104159591694
+I1130 06:49:17.931127 137274321021824 utils.py:1231] [32450] core_hours = 56.95104159591694
+I1130 06:49:17.931186 137274321021824 train.py:125] NOTE: Steps:32450/112603 [28.8%]
+Walltime:2d8h59m (0s eval)
+ETA:5d20h40m
+Total train time:8d5h38m
+I1130 06:54:29.693867 137274321021824 utils.py:1231] [32500] l2_params = 330.74743439601275
+I1130 06:54:29.694079 137274321021824 utils.py:1231] [32500] train/loss = 2.7328735291957855
+I1130 06:54:29.694174 137274321021824 utils.py:1231] [32500] l2_grads = 1.4029566049575806
+I1130 06:54:29.694244 137274321021824 utils.py:1231] [32500] lr = 0.000885974460564094
+I1130 06:54:29.694307 137274321021824 utils.py:1231] [32500] uptime = 205459.05666884803
+I1130 06:54:29.694365 137274321021824 utils.py:1231] [32500] examples_seen = 33280000.0
+I1130 06:54:29.694420 137274321021824 utils.py:1231] [32500] progress = 0.28862463699901425
+I1130 06:54:29.694474 137274321021824 utils.py:1231] [32500] epoch = 25.976316904821932
+I1130 06:54:29.694530 137274321021824 utils.py:1231] [32500] img/sec/core = 164.2270213067402
+I1130 06:54:29.694592 137274321021824 utils.py:1231] [32500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.03764258343445
+I1130 06:54:29.694647 137274321021824 utils.py:1231] [32500] core_hours = 57.03764258343445
+I1130 06:54:29.694713 137274321021824 train.py:125] NOTE: Steps:32500/112603 [28.9%]
+Walltime:2d9h4m (0s eval)
+ETA:5d20h35m
+Total train time:8d5h37m
+I1130 06:54:29.694814 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:32500/112603 [28.9%]
+Walltime:2d9h4m (0s eval)
+ETA:5d20h35m
+Total train time:8d5h37m
+I1130 06:55:59.662457 137274321021824 utils.py:1231] [32500] val/acc@1 = 0.5947066326530612
+I1130 06:55:59.662737 137274321021824 utils.py:1231] [32500] val/loss = 1.6807682273947462
+I1130 06:55:59.662917 137274321021824 utils.py:1231] [32500] z/secs/eval/val = 89.96802899197792
+I1130 06:55:59.662992 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 89.96802899197792
+I1130 07:01:04.988916 137274321021824 utils.py:1231] [32550] l2_params = 330.70783287174044
+I1130 07:01:04.989132 137274321021824 utils.py:1231] [32550] train/loss = 2.622967451810837
+I1130 07:01:04.989235 137274321021824 utils.py:1231] [32550] l2_grads = 1.4517711400985718
+I1130 07:01:04.989305 137274321021824 utils.py:1231] [32550] lr = 0.0008854874091237414
+I1130 07:01:04.989385 137274321021824 utils.py:1231] [32550] uptime = 205854.35174281098
+I1130 07:01:04.989448 137274321021824 utils.py:1231] [32550] examples_seen = 33331200.0
+I1130 07:01:04.989508 137274321021824 utils.py:1231] [32550] progress = 0.28906867490208965
+I1130 07:01:04.989567 137274321021824 utils.py:1231] [32550] epoch = 26.016280469290887
+I1130 07:01:04.989638 137274321021824 utils.py:1231] [32550] img/sec/core = 129.52349617388788
+I1130 07:01:04.989699 137274321021824 utils.py:1231] [32550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.14744677064638
+I1130 07:01:04.989761 137274321021824 utils.py:1231] [32550] core_hours = 57.14744677064638
+I1130 07:01:04.989835 137274321021824 train.py:125] NOTE: Steps:32550/112603 [28.9%]
+Walltime:2d9h10m (0s eval)
+ETA:5d20h33m
+Total train time:8d5h42m
+I1130 07:06:15.210806 137274321021824 utils.py:1231] [32600] l2_params = 330.6649742045398
+I1130 07:06:15.211034 137274321021824 utils.py:1231] [32600] train/loss = 4.893142998218536
+I1130 07:06:15.211127 137274321021824 utils.py:1231] [32600] l2_grads = 1.2682693004608154
+I1130 07:06:15.211187 137274321021824 utils.py:1231] [32600] lr = 0.0008849994541800784
+I1130 07:06:15.211241 137274321021824 utils.py:1231] [32600] uptime = 206164.57360138802
+I1130 07:06:15.211295 137274321021824 utils.py:1231] [32600] examples_seen = 33382400.0
+I1130 07:06:15.211343 137274321021824 utils.py:1231] [32600] progress = 0.28951271280516505
+I1130 07:06:15.211389 137274321021824 utils.py:1231] [32600] epoch = 26.056244033759846
+I1130 07:06:15.211439 137274321021824 utils.py:1231] [32600] img/sec/core = 165.04317340774642
+I1130 07:06:15.211494 137274321021824 utils.py:1231] [32600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.23361950914
+I1130 07:06:15.431197 137274321021824 utils.py:1231] [32600] core_hours = 57.23361950914
+I1130 07:06:15.431406 137274321021824 train.py:125] NOTE: Steps:32600/112603 [29.0%]
+Walltime:2d9h16m (0s eval)
+ETA:5d20h27m
+Total train time:8d5h42m
+I1130 07:11:18.229310 137274321021824 utils.py:1231] [32650] l2_params = 330.65234020592374
+I1130 07:11:18.229520 137274321021824 utils.py:1231] [32650] train/loss = 2.6891352236270905
+I1130 07:11:18.229638 137274321021824 utils.py:1231] [32650] l2_grads = 1.3287220001220703
+I1130 07:11:18.229726 137274321021824 utils.py:1231] [32650] lr = 0.0008845105968767709
+I1130 07:11:18.229784 137274321021824 utils.py:1231] [32650] uptime = 206467.59214588703
+I1130 07:11:18.229849 137274321021824 utils.py:1231] [32650] examples_seen = 33433600.0
+I1130 07:11:18.229911 137274321021824 utils.py:1231] [32650] progress = 0.28995675070824045
+I1130 07:11:18.229968 137274321021824 utils.py:1231] [32650] epoch = 26.0962075982288
+I1130 07:11:18.230034 137274321021824 utils.py:1231] [32650] img/sec/core = 168.96655643518878
+I1130 07:11:18.230089 137274321021824 utils.py:1231] [32650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.31779132705638
+I1130 07:11:18.230141 137274321021824 utils.py:1231] [32650] core_hours = 57.31779132705638
+I1130 07:11:18.230224 137274321021824 train.py:125] NOTE: Steps:32650/112603 [29.0%]
+Walltime:2d9h21m (0s eval)
+ETA:5d20h22m
+Total train time:8d5h41m
+I1130 07:16:30.008609 137274321021824 utils.py:1231] [32700] l2_params = 330.61100999775914
+I1130 07:16:30.008875 137274321021824 utils.py:1231] [32700] train/loss = 2.669571876525879
+I1130 07:16:30.009010 137274321021824 utils.py:1231] [32700] l2_grads = 1.224936842918396
+I1130 07:16:30.009097 137274321021824 utils.py:1231] [32700] lr = 0.0008840208383595995
+I1130 07:16:30.009172 137274321021824 utils.py:1231] [32700] uptime = 206779.371534102
+I1130 07:16:30.009248 137274321021824 utils.py:1231] [32700] examples_seen = 33484800.0
+I1130 07:16:30.009317 137274321021824 utils.py:1231] [32700] progress = 0.29040078861131585
+I1130 07:16:30.009374 137274321021824 utils.py:1231] [32700] epoch = 26.13617116269776
+I1130 07:16:30.009429 137274321021824 utils.py:1231] [32700] img/sec/core = 164.21868133468587
+I1130 07:16:30.009489 137274321021824 utils.py:1231] [32700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.40439671267166
+I1130 07:16:30.009547 137274321021824 utils.py:1231] [32700] core_hours = 57.40439671267166
+I1130 07:16:30.009610 137274321021824 train.py:125] NOTE: Steps:32700/112603 [29.0%]
+Walltime:2d9h26m (0s eval)
+ETA:5d20h16m
+Total train time:8d5h41m
+I1130 07:21:35.997267 137274321021824 utils.py:1231] [32750] l2_params = 330.60114353505946
+I1130 07:21:35.997499 137274321021824 utils.py:1231] [32750] train/loss = 3.0139715671539307
+I1130 07:21:35.997657 137274321021824 utils.py:1231] [32750] l2_grads = 1.2587696313858032
+I1130 07:21:35.997759 137274321021824 utils.py:1231] [32750] lr = 0.0008835301797764589
+I1130 07:21:35.997851 137274321021824 utils.py:1231] [32750] uptime = 207085.360208133
+I1130 07:21:35.997942 137274321021824 utils.py:1231] [32750] examples_seen = 33536000.0
+I1130 07:21:35.998026 137274321021824 utils.py:1231] [32750] progress = 0.29084482651439125
+I1130 07:21:35.998118 137274321021824 utils.py:1231] [32750] epoch = 26.176134727166716
+I1130 07:21:35.998211 137274321021824 utils.py:1231] [32750] img/sec/core = 167.32645468704484
+I1130 07:21:35.998296 137274321021824 utils.py:1231] [32750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.48939356656917
+I1130 07:21:35.998373 137274321021824 utils.py:1231] [32750] core_hours = 57.48939356656917
+I1130 07:21:35.998477 137274321021824 train.py:125] NOTE: Steps:32750/112603 [29.1%]
+Walltime:2d9h31m (0s eval)
+ETA:5d20h10m
+Total train time:8d5h40m
+I1130 07:26:45.777899 137274321021824 utils.py:1231] [32800] l2_params = 330.59529963590904
+I1130 07:26:45.778100 137274321021824 utils.py:1231] [32800] train/loss = 3.041784852743149
+I1130 07:26:45.778197 137274321021824 utils.py:1231] [32800] l2_grads = 1.2633261680603027
+I1130 07:26:45.778273 137274321021824 utils.py:1231] [32800] lr = 0.0008830386222773504
+I1130 07:26:45.778343 137274321021824 utils.py:1231] [32800] uptime = 207395.140691092
+I1130 07:26:45.778402 137274321021824 utils.py:1231] [32800] examples_seen = 33587200.0
+I1130 07:26:45.778459 137274321021824 utils.py:1231] [32800] progress = 0.29128886441746665
+I1130 07:26:45.778517 137274321021824 utils.py:1231] [32800] epoch = 26.216098291635674
+I1130 07:26:45.778568 137274321021824 utils.py:1231] [32800] img/sec/core = 165.27832712682823
+I1130 07:26:45.778623 137274321021824 utils.py:1231] [32800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.575443700724435
+I1130 07:26:45.778674 137274321021824 utils.py:1231] [32800] core_hours = 57.575443700724435
+I1130 07:26:45.778736 137274321021824 train.py:125] NOTE: Steps:32800/112603 [29.1%]
+Walltime:2d9h36m (0s eval)
+ETA:5d20h5m
+Total train time:8d5h40m
+I1130 07:31:53.326816 137274321021824 utils.py:1231] [32850] l2_params = 330.5834788023109
+I1130 07:31:53.327075 137274321021824 utils.py:1231] [32850] train/loss = 2.6853718161582947
+I1130 07:31:53.327195 137274321021824 utils.py:1231] [32850] l2_grads = 1.430335283279419
+I1130 07:31:53.327268 137274321021824 utils.py:1231] [32850] lr = 0.0008825461670143845
+I1130 07:31:53.327362 137274321021824 utils.py:1231] [32850] uptime = 207702.689724561
+I1130 07:31:53.327427 137274321021824 utils.py:1231] [32850] examples_seen = 33638400.0
+I1130 07:31:53.327479 137274321021824 utils.py:1231] [32850] progress = 0.2917329023205421
+I1130 07:31:53.327527 137274321021824 utils.py:1231] [32850] epoch = 26.25606185610463
+I1130 07:31:53.327577 137274321021824 utils.py:1231] [32850] img/sec/core = 166.47751879589802
+I1130 07:31:53.327631 137274321021824 utils.py:1231] [32850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.660873987799164
+I1130 07:31:53.327680 137274321021824 utils.py:1231] [32850] core_hours = 57.660873987799164
+I1130 07:31:53.327738 137274321021824 train.py:125] NOTE: Steps:32850/112603 [29.2%]
+Walltime:2d9h41m (0s eval)
+ETA:5d19h59m
+Total train time:8d5h39m
+I1130 07:37:05.104684 137274321021824 utils.py:1231] [32900] l2_params = 330.5311861406019
+I1130 07:37:05.104900 137274321021824 utils.py:1231] [32900] train/loss = 5.273788571357727
+I1130 07:37:05.105028 137274321021824 utils.py:1231] [32900] l2_grads = 1.2214559316635132
+I1130 07:37:05.105136 137274321021824 utils.py:1231] [32900] lr = 0.0008820528151417752
+I1130 07:37:05.105229 137274321021824 utils.py:1231] [32900] uptime = 208014.467583413
+I1130 07:37:05.105323 137274321021824 utils.py:1231] [32900] examples_seen = 33689600.0
+I1130 07:37:05.105399 137274321021824 utils.py:1231] [32900] progress = 0.2921769402236175
+I1130 07:37:05.105466 137274321021824 utils.py:1231] [32900] epoch = 26.296025420573585
+I1130 07:37:05.105550 137274321021824 utils.py:1231] [32900] img/sec/core = 164.2194868760846
+I1130 07:37:05.105637 137274321021824 utils.py:1231] [32900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.747478948591386
+I1130 07:37:05.105700 137274321021824 utils.py:1231] [32900] core_hours = 57.747478948591386
+I1130 07:37:05.105772 137274321021824 train.py:125] NOTE: Steps:32900/112603 [29.2%]
+Walltime:2d9h46m (0s eval)
+ETA:5d19h54m
+Total train time:8d5h39m
+I1130 07:42:10.414951 137274321021824 utils.py:1231] [32950] l2_params = 330.53415795821763
+I1130 07:42:10.415256 137274321021824 utils.py:1231] [32950] train/loss = 4.517104625701904
+I1130 07:42:10.415417 137274321021824 utils.py:1231] [32950] l2_grads = 1.0527074337005615
+I1130 07:42:10.415506 137274321021824 utils.py:1231] [32950] lr = 0.000881558567815838
+I1130 07:42:10.415566 137274321021824 utils.py:1231] [32950] uptime = 208319.777927562
+I1130 07:42:10.415633 137274321021824 utils.py:1231] [32950] examples_seen = 33740800.0
+I1130 07:42:10.415691 137274321021824 utils.py:1231] [32950] progress = 0.2926209781266929
+I1130 07:42:10.415747 137274321021824 utils.py:1231] [32950] epoch = 26.335988985042544
+I1130 07:42:10.415807 137274321021824 utils.py:1231] [32950] img/sec/core = 167.6982158685599
+I1130 07:42:10.415866 137274321021824 utils.py:1231] [32950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.83228737752166
+I1130 07:42:10.415930 137274321021824 utils.py:1231] [32950] core_hours = 57.83228737752166
+I1130 07:42:10.415999 137274321021824 train.py:125] NOTE: Steps:32950/112603 [29.3%]
+Walltime:2d9h51m (0s eval)
+ETA:5d19h48m
+Total train time:8d5h38m
+I1130 07:47:15.230892 137274321021824 utils.py:1231] [33000] l2_params = 330.51384758848945
+I1130 07:47:15.231113 137274321021824 utils.py:1231] [33000] train/loss = 4.421171009540558
+I1130 07:47:15.231210 137274321021824 utils.py:1231] [33000] l2_grads = 1.0448721647262573
+I1130 07:47:15.231266 137274321021824 utils.py:1231] [33000] lr = 0.0008810634261949867
+I1130 07:47:15.231329 137274321021824 utils.py:1231] [33000] uptime = 208624.593683045
+I1130 07:47:15.231379 137274321021824 utils.py:1231] [33000] examples_seen = 33792000.0
+I1130 07:47:15.231423 137274321021824 utils.py:1231] [33000] progress = 0.2930650160297683
+I1130 07:47:15.231480 137274321021824 utils.py:1231] [33000] epoch = 26.3759525495115
+I1130 07:47:15.231531 137274321021824 utils.py:1231] [33000] img/sec/core = 167.97032003437954
+I1130 07:47:15.231584 137274321021824 utils.py:1231] [33000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 57.91695842071138
+I1130 07:47:15.231629 137274321021824 utils.py:1231] [33000] core_hours = 57.91695842071138
+I1130 07:47:15.231686 137274321021824 train.py:125] NOTE: Steps:33000/112603 [29.3%]
+Walltime:2d9h57m (0s eval)
+ETA:5d19h42m
+Total train time:8d5h38m
+I1130 07:52:25.843942 137274321021824 utils.py:1231] [33050] l2_params = 330.5040917743808
+I1130 07:52:25.844205 137274321021824 utils.py:1231] [33050] train/loss = 2.7095682621002197
+I1130 07:52:25.844326 137274321021824 utils.py:1231] [33050] l2_grads = 1.4848899841308594
+I1130 07:52:25.844415 137274321021824 utils.py:1231] [33050] lr = 0.0008805673914397323
+I1130 07:52:25.844474 137274321021824 utils.py:1231] [33050] uptime = 208935.20683174703
+I1130 07:52:25.844528 137274321021824 utils.py:1231] [33050] examples_seen = 33843200.0
+I1130 07:52:25.844576 137274321021824 utils.py:1231] [33050] progress = 0.2935090539328437
+I1130 07:52:25.844626 137274321021824 utils.py:1231] [33050] epoch = 26.415916113980458
+I1130 07:52:25.844676 137274321021824 utils.py:1231] [33050] img/sec/core = 164.8352628147047
+I1130 07:52:25.844742 137274321021824 utils.py:1231] [33050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.00323985090639
+I1130 07:52:25.844797 137274321021824 utils.py:1231] [33050] core_hours = 58.00323985090639
+I1130 07:52:25.844856 137274321021824 train.py:125] NOTE: Steps:33050/112603 [29.4%]
+Walltime:2d10h2m (0s eval)
+ETA:5d19h37m
+Total train time:8d5h37m
+I1130 07:57:32.224672 137274321021824 utils.py:1231] [33100] l2_params = 330.4809230607752
+I1130 07:57:32.224921 137274321021824 utils.py:1231] [33100] train/loss = 2.74266716837883
+I1130 07:57:32.225048 137274321021824 utils.py:1231] [33100] l2_grads = 1.3551338911056519
+I1130 07:57:32.225118 137274321021824 utils.py:1231] [33100] lr = 0.0008800704647126781
+I1130 07:57:32.225187 137274321021824 utils.py:1231] [33100] uptime = 209241.58754342998
+I1130 07:57:32.225237 137274321021824 utils.py:1231] [33100] examples_seen = 33894400.0
+I1130 07:57:32.225293 137274321021824 utils.py:1231] [33100] progress = 0.2939530918359191
+I1130 07:57:32.225344 137274321021824 utils.py:1231] [33100] epoch = 26.455879678449413
+I1130 07:57:32.225401 137274321021824 utils.py:1231] [33100] img/sec/core = 167.11234763687457
+I1130 07:57:32.225459 137274321021824 utils.py:1231] [33100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.088345604151655
+I1130 07:57:32.225519 137274321021824 utils.py:1231] [33100] core_hours = 58.088345604151655
+I1130 07:57:32.225578 137274321021824 train.py:125] NOTE: Steps:33100/112603 [29.4%]
+Walltime:2d10h7m (0s eval)
+ETA:5d19h31m
+Total train time:8d5h37m
+I1130 08:02:39.248928 137274321021824 utils.py:1231] [33150] l2_params = 330.4618944219033
+I1130 08:02:39.249140 137274321021824 utils.py:1231] [33150] train/loss = 2.7718847393989563
+I1130 08:02:39.249246 137274321021824 utils.py:1231] [33150] l2_grads = 1.355901837348938
+I1130 08:02:39.249317 137274321021824 utils.py:1231] [33150] lr = 0.0008795726471785187
+I1130 08:02:39.249376 137274321021824 utils.py:1231] [33150] uptime = 209548.61173743202
+I1130 08:02:39.249434 137274321021824 utils.py:1231] [33150] examples_seen = 33945600.0
+I1130 08:02:39.249488 137274321021824 utils.py:1231] [33150] progress = 0.2943971297389945
+I1130 08:02:39.249543 137274321021824 utils.py:1231] [33150] epoch = 26.49584324291837
+I1130 08:02:39.249601 137274321021824 utils.py:1231] [33150] img/sec/core = 166.76210214125823
+I1130 08:02:39.249668 137274321021824 utils.py:1231] [33150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.173630102485546
+I1130 08:02:39.249724 137274321021824 utils.py:1231] [33150] core_hours = 58.173630102485546
+I1130 08:02:39.249788 137274321021824 train.py:125] NOTE: Steps:33150/112603 [29.4%]
+Walltime:2d10h12m (0s eval)
+ETA:5d19h26m
+Total train time:8d5h36m
+I1130 08:07:47.571996 137274321021824 utils.py:1231] [33200] l2_params = 330.43011660883406
+I1130 08:07:47.572200 137274321021824 utils.py:1231] [33200] train/loss = 2.947022706270218
+I1130 08:07:47.572329 137274321021824 utils.py:1231] [33200] l2_grads = 1.3439340591430664
+I1130 08:07:47.572422 137274321021824 utils.py:1231] [33200] lr = 0.0008790739400040357
+I1130 08:07:47.572498 137274321021824 utils.py:1231] [33200] uptime = 209856.934856218
+I1130 08:07:47.572586 137274321021824 utils.py:1231] [33200] examples_seen = 33996800.0
+I1130 08:07:47.572676 137274321021824 utils.py:1231] [33200] progress = 0.2948411676420699
+I1130 08:07:47.572770 137274321021824 utils.py:1231] [33200] epoch = 26.535806807387328
+I1130 08:07:47.572879 137274321021824 utils.py:1231] [33200] img/sec/core = 166.05955531844342
+I1130 08:07:47.572992 137274321021824 utils.py:1231] [33200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.25927541325944
+I1130 08:07:47.573072 137274321021824 utils.py:1231] [33200] core_hours = 58.25927541325944
+I1130 08:07:47.573169 137274321021824 train.py:125] NOTE: Steps:33200/112603 [29.5%]
+Walltime:2d10h17m (0s eval)
+ETA:5d19h20m
+Total train time:8d5h36m
+I1130 08:12:56.326128 137274321021824 utils.py:1231] [33250] l2_params = 330.3505078764743
+I1130 08:12:56.326408 137274321021824 utils.py:1231] [33250] train/loss = 2.989220976829529
+I1130 08:12:56.326522 137274321021824 utils.py:1231] [33250] l2_grads = 1.2770878076553345
+I1130 08:12:56.326598 137274321021824 utils.py:1231] [33250] lr = 0.0008785743443580957
+I1130 08:12:56.326666 137274321021824 utils.py:1231] [33250] uptime = 210165.68902598
+I1130 08:12:56.326724 137274321021824 utils.py:1231] [33250] examples_seen = 34048000.0
+I1130 08:12:56.326772 137274321021824 utils.py:1231] [33250] progress = 0.2952852055451453
+I1130 08:12:56.326820 137274321021824 utils.py:1231] [33250] epoch = 26.575770371856283
+I1130 08:12:56.326869 137274321021824 utils.py:1231] [33250] img/sec/core = 165.82771996073387
+I1130 08:12:56.326929 137274321021824 utils.py:1231] [33250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.345040460415554
+I1130 08:12:56.326977 137274321021824 utils.py:1231] [33250] core_hours = 58.345040460415554
+I1130 08:12:56.327035 137274321021824 train.py:125] NOTE: Steps:33250/112603 [29.5%]
+Walltime:2d10h22m (0s eval)
+ETA:5d19h15m
+Total train time:8d5h36m
+I1130 08:18:03.010772 137274321021824 utils.py:1231] [33300] l2_params = 330.33444516636274
+I1130 08:18:03.010988 137274321021824 utils.py:1231] [33300] train/loss = 3.5332559049129486
+I1130 08:18:03.011085 137274321021824 utils.py:1231] [33300] l2_grads = 1.2151157855987549
+I1130 08:18:03.011149 137274321021824 utils.py:1231] [33300] lr = 0.0008780738614116489
+I1130 08:18:03.011200 137274321021824 utils.py:1231] [33300] uptime = 210472.37356233702
+I1130 08:18:03.011268 137274321021824 utils.py:1231] [33300] examples_seen = 34099200.0
+I1130 08:18:03.011315 137274321021824 utils.py:1231] [33300] progress = 0.2957292434482207
+I1130 08:18:03.011366 137274321021824 utils.py:1231] [33300] epoch = 26.615733936325242
+I1130 08:18:03.011413 137274321021824 utils.py:1231] [33300] img/sec/core = 166.94679362770194
+I1130 08:18:03.011465 137274321021824 utils.py:1231] [33300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.4302306094036
+I1130 08:18:03.011511 137274321021824 utils.py:1231] [33300] core_hours = 58.4302306094036
+I1130 08:18:03.011566 137274321021824 train.py:125] NOTE: Steps:33300/112603 [29.6%]
+Walltime:2d10h27m (0s eval)
+ETA:5d19h9m
+Total train time:8d5h35m
+I1130 08:23:09.271279 137274321021824 utils.py:1231] [33350] l2_params = 330.294394098734
+I1130 08:23:09.271495 137274321021824 utils.py:1231] [33350] train/loss = 2.8168781995773315
+I1130 08:23:09.271603 137274321021824 utils.py:1231] [33350] l2_grads = 1.3270419836044312
+I1130 08:23:09.271671 137274321021824 utils.py:1231] [33350] lr = 0.0008775724923377239
+I1130 08:23:09.271729 137274321021824 utils.py:1231] [33350] uptime = 210778.63409040798
+I1130 08:23:09.271786 137274321021824 utils.py:1231] [33350] examples_seen = 34150400.0
+I1130 08:23:09.271843 137274321021824 utils.py:1231] [33350] progress = 0.29617328135129617
+I1130 08:23:09.271903 137274321021824 utils.py:1231] [33350] epoch = 26.655697500794197
+I1130 08:23:09.271975 137274321021824 utils.py:1231] [33350] img/sec/core = 167.177926331175
+I1130 08:23:09.272033 137274321021824 utils.py:1231] [33350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.515302978312214
+I1130 08:23:09.272085 137274321021824 utils.py:1231] [33350] core_hours = 58.515302978312214
+I1130 08:23:09.272149 137274321021824 train.py:125] NOTE: Steps:33350/112603 [29.6%]
+Walltime:2d10h32m (0s eval)
+ETA:5d19h3m
+Total train time:8d5h34m
+I1130 08:28:21.050054 137274321021824 utils.py:1231] [33400] l2_params = 330.25427186541765
+I1130 08:28:21.050267 137274321021824 utils.py:1231] [33400] train/loss = 2.414106398820877
+I1130 08:28:21.050372 137274321021824 utils.py:1231] [33400] l2_grads = 1.2821097373962402
+I1130 08:28:21.050437 137274321021824 utils.py:1231] [33400] lr = 0.0008770702383114287
+I1130 08:28:21.050514 137274321021824 utils.py:1231] [33400] uptime = 211090.412870194
+I1130 08:28:21.050577 137274321021824 utils.py:1231] [33400] examples_seen = 34201600.0
+I1130 08:28:21.050631 137274321021824 utils.py:1231] [33400] progress = 0.29661731925437157
+I1130 08:28:21.050680 137274321021824 utils.py:1231] [33400] epoch = 26.695661065263156
+I1130 08:28:21.050730 137274321021824 utils.py:1231] [33400] img/sec/core = 164.2190018035995
+I1130 08:28:21.050789 137274321021824 utils.py:1231] [33400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.60190819491944
+I1130 08:28:21.050840 137274321021824 utils.py:1231] [33400] core_hours = 58.60190819491944
+I1130 08:28:21.050922 137274321021824 train.py:125] NOTE: Steps:33400/112603 [29.7%]
+Walltime:2d10h38m (0s eval)
+ETA:5d18h58m
+Total train time:8d5h34m
+I1130 08:33:27.247616 137274321021824 utils.py:1231] [33450] l2_params = 330.23146976296715
+I1130 08:33:27.247862 137274321021824 utils.py:1231] [33450] train/loss = 2.717869848012924
+I1130 08:33:27.247976 137274321021824 utils.py:1231] [33450] l2_grads = 1.3051941394805908
+I1130 08:33:27.248063 137274321021824 utils.py:1231] [33450] lr = 0.0008765671005099422
+I1130 08:33:27.248129 137274321021824 utils.py:1231] [33450] uptime = 211396.61049004702
+I1130 08:33:27.248204 137274321021824 utils.py:1231] [33450] examples_seen = 34252800.0
+I1130 08:33:27.248265 137274321021824 utils.py:1231] [33450] progress = 0.29706135715744697
+I1130 08:33:27.248320 137274321021824 utils.py:1231] [33450] epoch = 26.73562462973211
+I1130 08:33:27.248373 137274321021824 utils.py:1231] [33450] img/sec/core = 167.21227299081323
+I1130 08:33:27.248460 137274321021824 utils.py:1231] [33450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.68696308932305
+I1130 08:33:27.248513 137274321021824 utils.py:1231] [33450] core_hours = 58.68696308932305
+I1130 08:33:27.248593 137274321021824 train.py:125] NOTE: Steps:33450/112603 [29.7%]
+Walltime:2d10h43m (0s eval)
+ETA:5d18h52m
+Total train time:8d5h34m
+I1130 08:38:35.570242 137274321021824 utils.py:1231] [33500] l2_params = 330.19021370124796
+I1130 08:38:35.570455 137274321021824 utils.py:1231] [33500] train/loss = 2.768832206726074
+I1130 08:38:35.570554 137274321021824 utils.py:1231] [33500] l2_grads = 1.3895106315612793
+I1130 08:38:35.570620 137274321021824 utils.py:1231] [33500] lr = 0.0008760630801125159
+I1130 08:38:35.570693 137274321021824 utils.py:1231] [33500] uptime = 211704.933055347
+I1130 08:38:35.570765 137274321021824 utils.py:1231] [33500] examples_seen = 34304000.0
+I1130 08:38:35.570822 137274321021824 utils.py:1231] [33500] progress = 0.29750539506052237
+I1130 08:38:35.570897 137274321021824 utils.py:1231] [33500] epoch = 26.775588194201067
+I1130 08:38:35.570952 137274321021824 utils.py:1231] [33500] img/sec/core = 166.05985342066975
+I1130 08:38:35.571026 137274321021824 utils.py:1231] [33500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.77260824635083
+I1130 08:38:35.571084 137274321021824 utils.py:1231] [33500] core_hours = 58.77260824635083
+I1130 08:38:35.571169 137274321021824 train.py:125] NOTE: Steps:33500/112603 [29.8%]
+Walltime:2d10h48m (0s eval)
+ETA:5d18h47m
+Total train time:8d5h33m
+I1130 08:43:47.341159 137274321021824 utils.py:1231] [33550] l2_params = 330.15203823539036
+I1130 08:43:47.341398 137274321021824 utils.py:1231] [33550] train/loss = 3.4501229226589203
+I1130 08:43:47.341522 137274321021824 utils.py:1231] [33550] l2_grads = 1.1575167179107666
+I1130 08:43:47.341627 137274321021824 utils.py:1231] [33550] lr = 0.0008755581783004698
+I1130 08:43:47.341697 137274321021824 utils.py:1231] [33550] uptime = 212016.704054847
+I1130 08:43:47.341768 137274321021824 utils.py:1231] [33550] examples_seen = 34355200.0
+I1130 08:43:47.341827 137274321021824 utils.py:1231] [33550] progress = 0.29794943296359777
+I1130 08:43:47.341900 137274321021824 utils.py:1231] [33550] epoch = 26.815551758670026
+I1130 08:43:47.341957 137274321021824 utils.py:1231] [33550] img/sec/core = 164.2230999102254
+I1130 08:43:47.342016 137274321021824 utils.py:1231] [33550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.85921130176749
+I1130 08:43:47.342066 137274321021824 utils.py:1231] [33550] core_hours = 58.85921130176749
+I1130 08:43:47.342129 137274321021824 train.py:125] NOTE: Steps:33550/112603 [29.8%]
+Walltime:2d10h53m (0s eval)
+ETA:5d18h41m
+Total train time:8d5h33m
+I1130 08:48:51.820433 137274321021824 utils.py:1231] [33600] l2_params = 330.10413252655883
+I1130 08:48:51.820642 137274321021824 utils.py:1231] [33600] train/loss = 2.675183117389679
+I1130 08:48:51.820755 137274321021824 utils.py:1231] [33600] l2_grads = 1.4321136474609375
+I1130 08:48:51.820822 137274321021824 utils.py:1231] [33600] lr = 0.0008750523962571909
+I1130 08:48:51.820897 137274321021824 utils.py:1231] [33600] uptime = 212321.18324225402
+I1130 08:48:51.820949 137274321021824 utils.py:1231] [33600] examples_seen = 34406400.0
+I1130 08:48:51.820996 137274321021824 utils.py:1231] [33600] progress = 0.29839347086667317
+I1130 08:48:51.821043 137274321021824 utils.py:1231] [33600] epoch = 26.85551532313898
+I1130 08:48:51.821093 137274321021824 utils.py:1231] [33600] img/sec/core = 168.15599265101605
+I1130 08:48:51.821149 137274321021824 utils.py:1231] [33600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 58.943788853825
+I1130 08:48:51.821198 137274321021824 utils.py:1231] [33600] core_hours = 58.943788853825
+I1130 08:48:51.821256 137274321021824 train.py:125] NOTE: Steps:33600/112603 [29.8%]
+Walltime:2d10h58m (0s eval)
+ETA:5d18h36m
+Total train time:8d5h32m
+I1130 08:53:52.630631 137274321021824 utils.py:1231] [33650] l2_params = 330.08826959006507
+I1130 08:53:52.630826 137274321021824 utils.py:1231] [33650] train/loss = 2.707103967666626
+I1130 08:53:52.630920 137274321021824 utils.py:1231] [33650] l2_grads = 1.3792681694030762
+I1130 08:53:52.630982 137274321021824 utils.py:1231] [33650] lr = 0.0008745457351681281
+I1130 08:53:52.631041 137274321021824 utils.py:1231] [33650] uptime = 212621.99340303702
+I1130 08:53:52.631091 137274321021824 utils.py:1231] [33650] examples_seen = 34457600.0
+I1130 08:53:52.631144 137274321021824 utils.py:1231] [33650] progress = 0.29883750876974857
+I1130 08:53:52.631191 137274321021824 utils.py:1231] [33650] epoch = 26.89547888760794
+I1130 08:53:52.631240 137274321021824 utils.py:1231] [33650] img/sec/core = 170.20701650079698
+I1130 08:53:52.631296 137274321021824 utils.py:1231] [33650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.02734723182027
+I1130 08:53:52.631345 137274321021824 utils.py:1231] [33650] core_hours = 59.02734723182027
+I1130 08:53:52.631405 137274321021824 train.py:125] NOTE: Steps:33650/112603 [29.9%]
+Walltime:2d11h3m (0s eval)
+ETA:5d18h30m
+Total train time:8d5h32m
+I1130 08:58:58.179665 137274321021824 utils.py:1231] [33700] l2_params = 330.0645985798467
+I1130 08:58:58.179888 137274321021824 utils.py:1231] [33700] train/loss = 4.403348863124847
+I1130 08:58:58.179996 137274321021824 utils.py:1231] [33700] l2_grads = 1.066469430923462
+I1130 08:58:58.180074 137274321021824 utils.py:1231] [33700] lr = 0.0008740381962207908
+I1130 08:58:58.180140 137274321021824 utils.py:1231] [33700] uptime = 212927.542500881
+I1130 08:58:58.180204 137274321021824 utils.py:1231] [33700] examples_seen = 34508800.0
+I1130 08:58:58.180265 137274321021824 utils.py:1231] [33700] progress = 0.299281546672824
+I1130 08:58:58.180328 137274321021824 utils.py:1231] [33700] epoch = 26.935442452076895
+I1130 08:58:58.180389 137274321021824 utils.py:1231] [33700] img/sec/core = 167.5671777834599
+I1130 08:58:58.180455 137274321021824 utils.py:1231] [33700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.11222198122138
+I1130 08:58:58.180514 137274321021824 utils.py:1231] [33700] core_hours = 59.11222198122138
+I1130 08:58:58.180599 137274321021824 train.py:125] NOTE: Steps:33700/112603 [29.9%]
+Walltime:2d11h8m (0s eval)
+ETA:5d18h24m
+Total train time:8d5h31m
+I1130 09:03:58.130722 137274321021824 utils.py:1231] [33750] l2_params = 330.059412206629
+I1130 09:03:58.130928 137274321021824 utils.py:1231] [33750] train/loss = 3.1850563287734985
+I1130 09:03:58.131044 137274321021824 utils.py:1231] [33750] l2_grads = 1.1522135734558105
+I1130 09:03:58.131119 137274321021824 utils.py:1231] [33750] lr = 0.0008735297806047466
+I1130 09:03:58.131183 137274321021824 utils.py:1231] [33750] uptime = 213227.49354409403
+I1130 09:03:58.131244 137274321021824 utils.py:1231] [33750] examples_seen = 34560000.0
+I1130 09:03:58.131307 137274321021824 utils.py:1231] [33750] progress = 0.2997255845758994
+I1130 09:03:58.131381 137274321021824 utils.py:1231] [33750] epoch = 26.975406016545854
+I1130 09:03:58.131442 137274321021824 utils.py:1231] [33750] img/sec/core = 170.69452218453796
+I1130 09:03:58.131506 137274321021824 utils.py:1231] [33750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.19554171544722
+I1130 09:03:58.131569 137274321021824 utils.py:1231] [33750] core_hours = 59.19554171544722
+I1130 09:03:58.131637 137274321021824 train.py:125] NOTE: Steps:33750/112603 [30.0%]
+Walltime:2d11h13m (0s eval)
+ETA:5d18h18m
+Total train time:8d5h30m
+I1130 09:09:06.124366 137274321021824 utils.py:1231] [33800] l2_params = 330.0452836353231
+I1130 09:09:06.124597 137274321021824 utils.py:1231] [33800] train/loss = 2.604268193244934
+I1130 09:09:06.124712 137274321021824 utils.py:1231] [33800] l2_grads = 1.277426838874817
+I1130 09:09:06.124783 137274321021824 utils.py:1231] [33800] lr = 0.0008730204895116172
+I1130 09:09:06.124846 137274321021824 utils.py:1231] [33800] uptime = 213535.48720733798
+I1130 09:09:06.124913 137274321021824 utils.py:1231] [33800] examples_seen = 34611200.0
+I1130 09:09:06.124971 137274321021824 utils.py:1231] [33800] progress = 0.30016962247897483
+I1130 09:09:06.125027 137274321021824 utils.py:1231] [33800] epoch = 27.01536958101481
+I1130 09:09:06.125084 137274321021824 utils.py:1231] [33800] img/sec/core = 166.23718637822006
+I1130 09:09:06.125145 137274321021824 utils.py:1231] [33800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.28109551079277
+I1130 09:09:06.125206 137274321021824 utils.py:1231] [33800] core_hours = 59.28109551079277
+I1130 09:09:06.125272 137274321021824 train.py:125] NOTE: Steps:33800/112603 [30.0%]
+Walltime:2d11h18m (0s eval)
+ETA:5d18h13m
+Total train time:8d5h30m
+I1130 09:14:13.977282 137274321021824 utils.py:1231] [33850] l2_params = 330.01256309576365
+I1130 09:14:13.977618 137274321021824 utils.py:1231] [33850] train/loss = 5.141309380531311
+I1130 09:14:13.977807 137274321021824 utils.py:1231] [33850] l2_grads = 1.246063470840454
+I1130 09:14:13.977876 137274321021824 utils.py:1231] [33850] lr = 0.000872510324135076
+I1130 09:14:13.977972 137274321021824 utils.py:1231] [33850] uptime = 213843.34033394099
+I1130 09:14:13.978023 137274321021824 utils.py:1231] [33850] examples_seen = 34662400.0
+I1130 09:14:13.978071 137274321021824 utils.py:1231] [33850] progress = 0.30061366038205023
+I1130 09:14:13.978118 137274321021824 utils.py:1231] [33850] epoch = 27.055333145483765
+I1130 09:14:13.978167 137274321021824 utils.py:1231] [33850] img/sec/core = 166.3130745656698
+I1130 09:14:13.978222 137274321021824 utils.py:1231] [33850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.36661026818249
+I1130 09:14:13.978272 137274321021824 utils.py:1231] [33850] core_hours = 59.36661026818249
+I1130 09:14:13.978331 137274321021824 train.py:125] NOTE: Steps:33850/112603 [30.1%]
+Walltime:2d11h24m (0s eval)
+ETA:5d18h7m
+Total train time:8d5h29m
+I1130 09:19:25.751208 137274321021824 utils.py:1231] [33900] l2_params = 329.9781142517426
+I1130 09:19:25.751426 137274321021824 utils.py:1231] [33900] train/loss = 2.9263998568058014
+I1130 09:19:25.751514 137274321021824 utils.py:1231] [33900] l2_grads = 1.2421318292617798
+I1130 09:19:25.751572 137274321021824 utils.py:1231] [33900] lr = 0.0008719992856708454
+I1130 09:19:25.751622 137274321021824 utils.py:1231] [33900] uptime = 214155.11398198403
+I1130 09:19:25.751674 137274321021824 utils.py:1231] [33900] examples_seen = 34713600.0
+I1130 09:19:25.751719 137274321021824 utils.py:1231] [33900] progress = 0.30105769828512563
+I1130 09:19:25.751764 137274321021824 utils.py:1231] [33900] epoch = 27.095296709952724
+I1130 09:19:25.751810 137274321021824 utils.py:1231] [33900] img/sec/core = 164.22170482136315
+I1130 09:19:25.751860 137274321021824 utils.py:1231] [33900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.45321405930555
+I1130 09:19:25.751911 137274321021824 utils.py:1231] [33900] core_hours = 59.45321405930555
+I1130 09:19:25.751965 137274321021824 train.py:125] NOTE: Steps:33900/112603 [30.1%]
+Walltime:2d11h29m (0s eval)
+ETA:5d18h2m
+Total train time:8d5h29m
+I1130 09:24:34.407018 137274321021824 utils.py:1231] [33950] l2_params = 329.9525487133958
+I1130 09:24:34.407199 137274321021824 utils.py:1231] [33950] train/loss = 3.7300906777381897
+I1130 09:24:34.407293 137274321021824 utils.py:1231] [33950] l2_grads = 1.2344324588775635
+I1130 09:24:34.407356 137274321021824 utils.py:1231] [33950] lr = 0.000871487375316695
+I1130 09:24:34.407415 137274321021824 utils.py:1231] [33950] uptime = 214463.76977677102
+I1130 09:24:34.407469 137274321021824 utils.py:1231] [33950] examples_seen = 34764800.0
+I1130 09:24:34.407539 137274321021824 utils.py:1231] [33950] progress = 0.30150173618820103
+I1130 09:24:34.407589 137274321021824 utils.py:1231] [33950] epoch = 27.13526027442168
+I1130 09:24:34.407650 137274321021824 utils.py:1231] [33950] img/sec/core = 165.88057267913788
+I1130 09:24:34.407707 137274321021824 utils.py:1231] [33950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.53895178007972
+I1130 09:24:34.407761 137274321021824 utils.py:1231] [33950] core_hours = 59.53895178007972
+I1130 09:24:34.407820 137274321021824 train.py:125] NOTE: Steps:33950/112603 [30.2%]
+Walltime:2d11h34m (0s eval)
+ETA:5d17h56m
+Total train time:8d5h29m
+I1130 09:29:43.035495 137274321021824 utils.py:1231] [34000] l2_params = 329.95271245324835
+I1130 09:29:43.035736 137274321021824 utils.py:1231] [34000] train/loss = 2.650960773229599
+I1130 09:29:43.035859 137274321021824 utils.py:1231] [34000] l2_grads = 1.4109383821487427
+I1130 09:29:43.035921 137274321021824 utils.py:1231] [34000] lr = 0.0008709745942724376
+I1130 09:29:43.035992 137274321021824 utils.py:1231] [34000] uptime = 214772.39835396002
+I1130 09:29:43.036041 137274321021824 utils.py:1231] [34000] examples_seen = 34816000.0
+I1130 09:29:43.036087 137274321021824 utils.py:1231] [34000] progress = 0.30194577409127643
+I1130 09:29:43.036133 137274321021824 utils.py:1231] [34000] epoch = 27.175223838890638
+I1130 09:29:43.036182 137274321021824 utils.py:1231] [34000] img/sec/core = 165.8952014953769
+I1130 09:29:43.036235 137274321021824 utils.py:1231] [34000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.62468194040999
+I1130 09:29:43.036284 137274321021824 utils.py:1231] [34000] core_hours = 59.62468194040999
+I1130 09:29:43.036341 137274321021824 train.py:125] NOTE: Steps:34000/112603 [30.2%]
+Walltime:2d11h39m (0s eval)
+ETA:5d17h51m
+Total train time:8d5h28m
+I1130 09:34:53.261468 137274321021824 utils.py:1231] [34050] l2_params = 329.90707212296724
+I1130 09:34:53.261755 137274321021824 utils.py:1231] [34050] train/loss = 2.6636302769184113
+I1130 09:34:53.261945 137274321021824 utils.py:1231] [34050] l2_grads = 1.3998708724975586
+I1130 09:34:53.262042 137274321021824 utils.py:1231] [34050] lr = 0.0008704609437399269
+I1130 09:34:53.262131 137274321021824 utils.py:1231] [34050] uptime = 215082.624486304
+I1130 09:34:53.262216 137274321021824 utils.py:1231] [34050] examples_seen = 34867200.0
+I1130 09:34:53.262319 137274321021824 utils.py:1231] [34050] progress = 0.30238981199435183
+I1130 09:34:53.262414 137274321021824 utils.py:1231] [34050] epoch = 27.215187403359593
+I1130 09:34:53.262488 137274321021824 utils.py:1231] [34050] img/sec/core = 165.04089972416563
+I1130 09:34:53.262578 137274321021824 utils.py:1231] [34050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.7108558660611
+I1130 09:34:53.262651 137274321021824 utils.py:1231] [34050] core_hours = 59.7108558660611
+I1130 09:34:53.262740 137274321021824 train.py:125] NOTE: Steps:34050/112603 [30.2%]
+Walltime:2d11h44m (0s eval)
+ETA:5d17h45m
+Total train time:8d5h28m
+I1130 09:40:01.894836 137274321021824 utils.py:1231] [34100] l2_params = 329.86432310517273
+I1130 09:40:01.895068 137274321021824 utils.py:1231] [34100] train/loss = 2.5639908611774445
+I1130 09:40:01.895191 137274321021824 utils.py:1231] [34100] l2_grads = 1.303537368774414
+I1130 09:40:01.895268 137274321021824 utils.py:1231] [34100] lr = 0.0008699464249230544
+I1130 09:40:01.895325 137274321021824 utils.py:1231] [34100] uptime = 215391.25768751698
+I1130 09:40:01.895397 137274321021824 utils.py:1231] [34100] examples_seen = 34918400.0
+I1130 09:40:01.895446 137274321021824 utils.py:1231] [34100] progress = 0.30283384989742723
+I1130 09:40:01.895494 137274321021824 utils.py:1231] [34100] epoch = 27.25515096782855
+I1130 09:40:01.895547 137274321021824 utils.py:1231] [34100] img/sec/core = 165.892716009723
+I1130 09:40:01.895602 137274321021824 utils.py:1231] [34100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.79658731084249
+I1130 09:40:01.895651 137274321021824 utils.py:1231] [34100] core_hours = 59.79658731084249
+I1130 09:40:01.895709 137274321021824 train.py:125] NOTE: Steps:34100/112603 [30.3%]
+Walltime:2d11h49m (0s eval)
+ETA:5d17h40m
+Total train time:8d5h28m
+I1130 09:45:10.388763 137274321021824 utils.py:1231] [34150] l2_params = 329.8524578397926
+I1130 09:45:10.389007 137274321021824 utils.py:1231] [34150] train/loss = 2.587774157524109
+I1130 09:45:10.389129 137274321021824 utils.py:1231] [34150] l2_grads = 1.379525065422058
+I1130 09:45:10.389207 137274321021824 utils.py:1231] [34150] lr = 0.0008694310390277452
+I1130 09:45:10.389269 137274321021824 utils.py:1231] [34150] uptime = 215699.75163143902
+I1130 09:45:10.389324 137274321021824 utils.py:1231] [34150] examples_seen = 34969600.0
+I1130 09:45:10.389386 137274321021824 utils.py:1231] [34150] progress = 0.30327788780050263
+I1130 09:45:10.389436 137274321021824 utils.py:1231] [34150] epoch = 27.295114532297507
+I1130 09:45:10.389488 137274321021824 utils.py:1231] [34150] img/sec/core = 165.96760166203617
+I1130 09:45:10.389551 137274321021824 utils.py:1231] [34150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.88228007304305
+I1130 09:45:10.389607 137274321021824 utils.py:1231] [34150] core_hours = 59.88228007304305
+I1130 09:45:10.389668 137274321021824 train.py:125] NOTE: Steps:34150/112603 [30.3%]
+Walltime:2d11h54m (0s eval)
+ETA:5d17h34m
+Total train time:8d5h27m
+I1130 09:50:15.642608 137274321021824 utils.py:1231] [34200] l2_params = 329.8106677978744
+I1130 09:50:15.642865 137274321021824 utils.py:1231] [34200] train/loss = 2.6349380910396576
+I1130 09:50:15.642962 137274321021824 utils.py:1231] [34200] l2_grads = 1.3976775407791138
+I1130 09:50:15.643022 137274321021824 utils.py:1231] [34200] lr = 0.0008689147872619597
+I1130 09:50:15.643072 137274321021824 utils.py:1231] [34200] uptime = 216005.00543459802
+I1130 09:50:15.643125 137274321021824 utils.py:1231] [34200] examples_seen = 35020800.0
+I1130 09:50:15.643173 137274321021824 utils.py:1231] [34200] progress = 0.30372192570357803
+I1130 09:50:15.643220 137274321021824 utils.py:1231] [34200] epoch = 27.335078096766463
+I1130 09:50:15.643269 137274321021824 utils.py:1231] [34200] img/sec/core = 167.72927796523499
+I1130 09:50:15.643323 137274321021824 utils.py:1231] [34200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 59.96707279614277
+I1130 09:50:15.643372 137274321021824 utils.py:1231] [34200] core_hours = 59.96707279614277
+I1130 09:50:15.643432 137274321021824 train.py:125] NOTE: Steps:34200/112603 [30.4%]
+Walltime:2d12h0m (0s eval)
+ETA:5d17h28m
+Total train time:8d5h27m
+I1130 09:55:21.092502 137274321021824 utils.py:1231] [34250] l2_params = 329.7906183218705
+I1130 09:55:21.092719 137274321021824 utils.py:1231] [34250] train/loss = 2.7314326763153076
+I1130 09:55:21.092834 137274321021824 utils.py:1231] [34250] l2_grads = 1.3114240169525146
+I1130 09:55:21.092915 137274321021824 utils.py:1231] [34250] lr = 0.0008683976708356841
+I1130 09:55:21.093003 137274321021824 utils.py:1231] [34250] uptime = 216310.455361403
+I1130 09:55:21.093106 137274321021824 utils.py:1231] [34250] examples_seen = 35072000.0
+I1130 09:55:21.093201 137274321021824 utils.py:1231] [34250] progress = 0.3041659636066535
+I1130 09:55:21.093293 137274321021824 utils.py:1231] [34250] epoch = 27.37504166123542
+I1130 09:55:21.093394 137274321021824 utils.py:1231] [34250] img/sec/core = 167.62158215439356
+I1130 09:55:21.093476 137274321021824 utils.py:1231] [34250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.05191999803306
+I1130 09:55:21.093544 137274321021824 utils.py:1231] [34250] core_hours = 60.05191999803306
+I1130 09:55:21.093640 137274321021824 train.py:125] NOTE: Steps:34250/112603 [30.4%]
+Walltime:2d12h5m (0s eval)
+ETA:5d17h23m
+Total train time:8d5h26m
+I1130 10:00:27.608058 137274321021824 utils.py:1231] [34300] l2_params = 329.7581463285446
+I1130 10:00:27.608299 137274321021824 utils.py:1231] [34300] train/loss = 2.6346271634101868
+I1130 10:00:27.608431 137274321021824 utils.py:1231] [34300] l2_grads = 1.4278557300567627
+I1130 10:00:27.608521 137274321021824 utils.py:1231] [34300] lr = 0.0008678796909609348
+I1130 10:00:27.608588 137274321021824 utils.py:1231] [34300] uptime = 216616.970949437
+I1130 10:00:27.608657 137274321021824 utils.py:1231] [34300] examples_seen = 35123200.0
+I1130 10:00:27.608722 137274321021824 utils.py:1231] [34300] progress = 0.3046100015097289
+I1130 10:00:27.608782 137274321021824 utils.py:1231] [34300] epoch = 27.415005225704377
+I1130 10:00:27.608847 137274321021824 utils.py:1231] [34300] img/sec/core = 167.03881302873612
+I1130 10:00:27.608913 137274321021824 utils.py:1231] [34300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.137063216931374
+I1130 10:00:27.608969 137274321021824 utils.py:1231] [34300] core_hours = 60.137063216931374
+I1130 10:00:27.609034 137274321021824 train.py:125] NOTE: Steps:34300/112603 [30.5%]
+Walltime:2d12h10m (0s eval)
+ETA:5d17h17m
+Total train time:8d5h26m
+I1130 10:05:33.882819 137274321021824 utils.py:1231] [34350] l2_params = 329.7128758922729
+I1130 10:05:33.883021 137274321021824 utils.py:1231] [34350] train/loss = 4.388948619365692
+I1130 10:05:33.883119 137274321021824 utils.py:1231] [34350] l2_grads = 1.1087031364440918
+I1130 10:05:33.883181 137274321021824 utils.py:1231] [34350] lr = 0.0008673608488517499
+I1130 10:05:33.883242 137274321021824 utils.py:1231] [34350] uptime = 216923.245604106
+I1130 10:05:33.883295 137274321021824 utils.py:1231] [34350] examples_seen = 35174400.0
+I1130 10:05:33.883346 137274321021824 utils.py:1231] [34350] progress = 0.3050540394128043
+I1130 10:05:33.883394 137274321021824 utils.py:1231] [34350] epoch = 27.454968790173336
+I1130 10:05:33.883444 137274321021824 utils.py:1231] [34350] img/sec/core = 167.1702154242256
+I1130 10:05:33.883497 137274321021824 utils.py:1231] [34350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.222139509894994
+I1130 10:05:33.883546 137274321021824 utils.py:1231] [34350] core_hours = 60.222139509894994
+I1130 10:05:33.883605 137274321021824 train.py:125] NOTE: Steps:34350/112603 [30.5%]
+Walltime:2d12h15m (0s eval)
+ETA:5d17h12m
+Total train time:8d5h25m
+I1130 10:10:41.766610 137274321021824 utils.py:1231] [34400] l2_params = 329.67472052597026
+I1130 10:10:41.766837 137274321021824 utils.py:1231] [34400] train/loss = 2.6807615756988525
+I1130 10:10:41.766971 137274321021824 utils.py:1231] [34400] l2_grads = 1.511217474937439
+I1130 10:10:41.767047 137274321021824 utils.py:1231] [34400] lr = 0.0008668411457241881
+I1130 10:10:41.767111 137274321021824 utils.py:1231] [34400] uptime = 217231.12946737098
+I1130 10:10:41.767168 137274321021824 utils.py:1231] [34400] examples_seen = 35225600.0
+I1130 10:10:41.767217 137274321021824 utils.py:1231] [34400] progress = 0.3054980773158797
+I1130 10:10:41.767267 137274321021824 utils.py:1231] [34400] epoch = 27.49493235464229
+I1130 10:10:41.767322 137274321021824 utils.py:1231] [34400] img/sec/core = 166.2964711987318
+I1130 10:10:41.767382 137274321021824 utils.py:1231] [34400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.30766280524638
+I1130 10:10:41.767431 137274321021824 utils.py:1231] [34400] core_hours = 60.30766280524638
+I1130 10:10:41.767506 137274321021824 train.py:125] NOTE: Steps:34400/112603 [30.5%]
+Walltime:2d12h20m (0s eval)
+ETA:5d17h6m
+Total train time:8d5h25m
+I1130 10:15:53.538621 137274321021824 utils.py:1231] [34450] l2_params = 329.6313101954161
+I1130 10:15:53.538830 137274321021824 utils.py:1231] [34450] train/loss = 2.6114572286605835
+I1130 10:15:53.538930 137274321021824 utils.py:1231] [34450] l2_grads = 1.344495177268982
+I1130 10:15:53.538993 137274321021824 utils.py:1231] [34450] lr = 0.0008663205827963272
+I1130 10:15:53.539047 137274321021824 utils.py:1231] [34450] uptime = 217542.901409277
+I1130 10:15:53.539099 137274321021824 utils.py:1231] [34450] examples_seen = 35276800.0
+I1130 10:15:53.539146 137274321021824 utils.py:1231] [34450] progress = 0.3059421152189551
+I1130 10:15:53.539192 137274321021824 utils.py:1231] [34450] epoch = 27.534895919111246
+I1130 10:15:53.539242 137274321021824 utils.py:1231] [34450] img/sec/core = 164.22260350623316
+I1130 10:15:53.539295 137274321021824 utils.py:1231] [34450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.394266122442495
+I1130 10:15:53.539344 137274321021824 utils.py:1231] [34450] core_hours = 60.394266122442495
+I1130 10:15:53.539402 137274321021824 train.py:125] NOTE: Steps:34450/112603 [30.6%]
+Walltime:2d12h25m (0s eval)
+ETA:5d17h1m
+Total train time:8d5h24m
+I1130 10:20:58.534199 137274321021824 utils.py:1231] [34500] l2_params = 329.61584744062696
+I1130 10:20:58.534413 137274321021824 utils.py:1231] [34500] train/loss = 2.885935068130493
+I1130 10:20:58.534522 137274321021824 utils.py:1231] [34500] l2_grads = 1.3624422550201416
+I1130 10:20:58.534610 137274321021824 utils.py:1231] [34500] lr = 0.0008657991612882597
+I1130 10:20:58.534679 137274321021824 utils.py:1231] [34500] uptime = 217847.89703681698
+I1130 10:20:58.534754 137274321021824 utils.py:1231] [34500] examples_seen = 35328000.0
+I1130 10:20:58.534841 137274321021824 utils.py:1231] [34500] progress = 0.3063861531220305
+I1130 10:20:58.534916 137274321021824 utils.py:1231] [34500] epoch = 27.574859483580205
+I1130 10:20:58.535000 137274321021824 utils.py:1231] [34500] img/sec/core = 167.8712590503842
+I1130 10:20:58.535068 137274321021824 utils.py:1231] [34500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.47898713009249
+I1130 10:20:58.535126 137274321021824 utils.py:1231] [34500] core_hours = 60.47898713009249
+I1130 10:20:58.535192 137274321021824 train.py:125] NOTE: Steps:34500/112603 [30.6%]
+Walltime:2d12h30m (0s eval)
+ETA:5d16h55m
+Total train time:8d5h24m
+I1130 10:26:05.610190 137274321021824 utils.py:1231] [34550] l2_params = 329.5868267306213
+I1130 10:26:05.610478 137274321021824 utils.py:1231] [34550] train/loss = 2.941830337047577
+I1130 10:26:05.610647 137274321021824 utils.py:1231] [34550] l2_grads = 1.3272345066070557
+I1130 10:26:05.610719 137274321021824 utils.py:1231] [34550] lr = 0.0008652768824220903
+I1130 10:26:05.610779 137274321021824 utils.py:1231] [34550] uptime = 218154.973140657
+I1130 10:26:05.610838 137274321021824 utils.py:1231] [34550] examples_seen = 35379200.0
+I1130 10:26:05.610905 137274321021824 utils.py:1231] [34550] progress = 0.3068301910251059
+I1130 10:26:05.610973 137274321021824 utils.py:1231] [34550] epoch = 27.61482304804916
+I1130 10:26:05.611036 137274321021824 utils.py:1231] [34550] img/sec/core = 166.73391175588966
+I1130 10:26:05.611105 137274321021824 utils.py:1231] [34550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.564286047825824
+I1130 10:26:05.611163 137274321021824 utils.py:1231] [34550] core_hours = 60.564286047825824
+I1130 10:26:05.611252 137274321021824 train.py:125] NOTE: Steps:34550/112603 [30.7%]
+Walltime:2d12h35m (0s eval)
+ETA:5d16h49m
+Total train time:8d5h23m
+I1130 10:31:17.391739 137274321021824 utils.py:1231] [34600] l2_params = 329.51900236522636
+I1130 10:31:17.392006 137274321021824 utils.py:1231] [34600] train/loss = 2.952417403459549
+I1130 10:31:17.392132 137274321021824 utils.py:1231] [34600] l2_grads = 1.2107348442077637
+I1130 10:31:17.392232 137274321021824 utils.py:1231] [34600] lr = 0.0008647537474219342
+I1130 10:31:17.392316 137274321021824 utils.py:1231] [34600] uptime = 218466.75467770203
+I1130 10:31:17.392387 137274321021824 utils.py:1231] [34600] examples_seen = 35430400.0
+I1130 10:31:17.392441 137274321021824 utils.py:1231] [34600] progress = 0.3072742289281813
+I1130 10:31:17.392494 137274321021824 utils.py:1231] [34600] epoch = 27.65478661251812
+I1130 10:31:17.392549 137274321021824 utils.py:1231] [34600] img/sec/core = 164.21754952284397
+I1130 10:31:17.392611 137274321021824 utils.py:1231] [34600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.65089203033833
+I1130 10:31:17.392663 137274321021824 utils.py:1231] [34600] core_hours = 60.65089203033833
+I1130 10:31:17.392750 137274321021824 train.py:125] NOTE: Steps:34600/112603 [30.7%]
+Walltime:2d12h41m (0s eval)
+ETA:5d16h44m
+Total train time:8d5h23m
+I1130 10:36:20.897929 137274321021824 utils.py:1231] [34650] l2_params = 329.4731108977394
+I1130 10:36:20.898119 137274321021824 utils.py:1231] [34650] train/loss = 2.803847759962082
+I1130 10:36:20.898210 137274321021824 utils.py:1231] [34650] l2_grads = 1.341939091682434
+I1130 10:36:20.898265 137274321021824 utils.py:1231] [34650] lr = 0.0008642297575139114
+I1130 10:36:20.898317 137274321021824 utils.py:1231] [34650] uptime = 218770.260679785
+I1130 10:36:20.898368 137274321021824 utils.py:1231] [34650] examples_seen = 35481600.0
+I1130 10:36:20.898416 137274321021824 utils.py:1231] [34650] progress = 0.3077182668312567
+I1130 10:36:20.898463 137274321021824 utils.py:1231] [34650] epoch = 27.694750176987075
+I1130 10:36:20.898511 137274321021824 utils.py:1231] [34650] img/sec/core = 168.69518114505348
+I1130 10:36:20.898565 137274321021824 utils.py:1231] [34650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.735199253139164
+I1130 10:36:20.898614 137274321021824 utils.py:1231] [34650] core_hours = 60.735199253139164
+I1130 10:36:20.898673 137274321021824 train.py:125] NOTE: Steps:34650/112603 [30.8%]
+Walltime:2d12h46m (0s eval)
+ETA:5d16h38m
+Total train time:8d5h23m
+I1130 10:41:26.264488 137274321021824 utils.py:1231] [34700] l2_params = 329.4347347224158
+I1130 10:41:26.264720 137274321021824 utils.py:1231] [34700] train/loss = 2.70942884683609
+I1130 10:41:26.264811 137274321021824 utils.py:1231] [34700] l2_grads = 1.316693902015686
+I1130 10:41:26.264872 137274321021824 utils.py:1231] [34700] lr = 0.0008637049139261466
+I1130 10:41:26.264927 137274321021824 utils.py:1231] [34700] uptime = 219075.627289431
+I1130 10:41:26.264978 137274321021824 utils.py:1231] [34700] examples_seen = 35532800.0
+I1130 10:41:26.265026 137274321021824 utils.py:1231] [34700] progress = 0.30816230473433215
+I1130 10:41:26.265073 137274321021824 utils.py:1231] [34700] epoch = 27.734713741456034
+I1130 10:41:26.265122 137274321021824 utils.py:1231] [34700] img/sec/core = 167.6673165391363
+I1130 10:41:26.265176 137274321021824 utils.py:1231] [34700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.820023311374165
+I1130 10:41:26.265225 137274321021824 utils.py:1231] [34700] core_hours = 60.820023311374165
+I1130 10:41:26.265290 137274321021824 train.py:125] NOTE: Steps:34700/112603 [30.8%]
+Walltime:2d12h51m (0s eval)
+ETA:5d16h33m
+Total train time:8d5h22m
+I1130 10:46:31.532645 137274321021824 utils.py:1231] [34750] l2_params = 329.3989908350218
+I1130 10:46:31.532861 137274321021824 utils.py:1231] [34750] train/loss = 4.4241843819618225
+I1130 10:46:31.532995 137274321021824 utils.py:1231] [34750] l2_grads = 1.2151802778244019
+I1130 10:46:31.533075 137274321021824 utils.py:1231] [34750] lr = 0.0008631792178887662
+I1130 10:46:31.533157 137274321021824 utils.py:1231] [34750] uptime = 219380.89551270002
+I1130 10:46:31.533209 137274321021824 utils.py:1231] [34750] examples_seen = 35584000.0
+I1130 10:46:31.533266 137274321021824 utils.py:1231] [34750] progress = 0.30860634263740755
+I1130 10:46:31.533316 137274321021824 utils.py:1231] [34750] epoch = 27.77467730592499
+I1130 10:46:31.533369 137274321021824 utils.py:1231] [34750] img/sec/core = 167.72135485219894
+I1130 10:46:31.533426 137274321021824 utils.py:1231] [34750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.904820040059995
+I1130 10:46:31.533478 137274321021824 utils.py:1231] [34750] core_hours = 60.904820040059995
+I1130 10:46:31.533566 137274321021824 train.py:125] NOTE: Steps:34750/112603 [30.9%]
+Walltime:2d12h56m (0s eval)
+ETA:5d16h27m
+Total train time:8d5h21m
+I1130 10:51:35.613713 137274321021824 utils.py:1231] [34800] l2_params = 329.35868058085526
+I1130 10:51:35.613931 137274321021824 utils.py:1231] [34800] train/loss = 2.6374533474445343
+I1130 10:51:35.614033 137274321021824 utils.py:1231] [34800] l2_grads = 1.3459970951080322
+I1130 10:51:35.614099 137274321021824 utils.py:1231] [34800] lr = 0.0008626526706338938
+I1130 10:51:35.614169 137274321021824 utils.py:1231] [34800] uptime = 219684.976517544
+I1130 10:51:35.614227 137274321021824 utils.py:1231] [34800] examples_seen = 35635200.0
+I1130 10:51:35.614281 137274321021824 utils.py:1231] [34800] progress = 0.30905038054048295
+I1130 10:51:35.614345 137274321021824 utils.py:1231] [34800] epoch = 27.814640870393944
+I1130 10:51:35.614404 137274321021824 utils.py:1231] [34800] img/sec/core = 168.37618655683065
+I1130 10:51:35.614465 137274321021824 utils.py:1231] [34800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 60.98928698584999
+I1130 10:51:35.614518 137274321021824 utils.py:1231] [34800] core_hours = 60.98928698584999
+I1130 10:51:35.614581 137274321021824 train.py:125] NOTE: Steps:34800/112603 [30.9%]
+Walltime:2d13h1m (0s eval)
+ETA:5d16h21m
+Total train time:8d5h21m
+I1130 10:56:46.972308 137274321021824 utils.py:1231] [34850] l2_params = 329.31937849783696
+I1130 10:56:46.972553 137274321021824 utils.py:1231] [34850] train/loss = 3.2319483160972595
+I1130 10:56:46.972684 137274321021824 utils.py:1231] [34850] l2_grads = 1.1795594692230225
+I1130 10:56:46.972780 137274321021824 utils.py:1231] [34850] lr = 0.0008621252733956471
+I1130 10:56:46.972861 137274321021824 utils.py:1231] [34850] uptime = 219996.33521814202
+I1130 10:56:46.972946 137274321021824 utils.py:1231] [34850] examples_seen = 35686400.0
+I1130 10:56:46.973014 137274321021824 utils.py:1231] [34850] progress = 0.30949441844355835
+I1130 10:56:46.973073 137274321021824 utils.py:1231] [34850] epoch = 27.854604434862903
+I1130 10:56:46.973141 137274321021824 utils.py:1231] [34850] img/sec/core = 164.44056293163874
+I1130 10:56:46.973205 137274321021824 utils.py:1231] [34850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.075775513793886
+I1130 10:56:46.973260 137274321021824 utils.py:1231] [34850] core_hours = 61.075775513793886
+I1130 10:56:46.973325 137274321021824 train.py:125] NOTE: Steps:34850/112603 [30.9%]
+Walltime:2d13h6m (0s eval)
+ETA:5d16h16m
+Total train time:8d5h21m
+I1130 11:01:50.530753 137274321021824 utils.py:1231] [34900] l2_params = 329.29503549580147
+I1130 11:01:50.530986 137274321021824 utils.py:1231] [34900] train/loss = 5.0164636969566345
+I1130 11:01:50.531097 137274321021824 utils.py:1231] [34900] l2_grads = 1.077697515487671
+I1130 11:01:50.531175 137274321021824 utils.py:1231] [34900] lr = 0.000861597027410137
+I1130 11:01:50.531233 137274321021824 utils.py:1231] [34900] uptime = 220299.893594952
+I1130 11:01:50.531291 137274321021824 utils.py:1231] [34900] examples_seen = 35737600.0
+I1130 11:01:50.531346 137274321021824 utils.py:1231] [34900] progress = 0.30993845634663375
+I1130 11:01:50.531401 137274321021824 utils.py:1231] [34900] epoch = 27.89456799933186
+I1130 11:01:50.531491 137274321021824 utils.py:1231] [34900] img/sec/core = 168.66607516500298
+I1130 11:01:50.531562 137274321021824 utils.py:1231] [34900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.160097285129986
+I1130 11:01:50.531620 137274321021824 utils.py:1231] [34900] core_hours = 61.160097285129986
+I1130 11:01:50.531683 137274321021824 train.py:125] NOTE: Steps:34900/112603 [31.0%]
+Walltime:2d13h11m (0s eval)
+ETA:5d16h10m
+Total train time:8d5h20m
+I1130 11:06:51.558291 137274321021824 utils.py:1231] [34950] l2_params = 329.2390202081474
+I1130 11:06:51.558534 137274321021824 utils.py:1231] [34950] train/loss = 2.580211967229843
+I1130 11:06:51.558631 137274321021824 utils.py:1231] [34950] l2_grads = 1.3125890493392944
+I1130 11:06:51.558697 137274321021824 utils.py:1231] [34950] lr = 0.0008610679339154646
+I1130 11:06:51.558749 137274321021824 utils.py:1231] [34950] uptime = 220600.92111152702
+I1130 11:06:51.558800 137274321021824 utils.py:1231] [34950] examples_seen = 35788800.0
+I1130 11:06:51.558849 137274321021824 utils.py:1231] [34950] progress = 0.31038249424970915
+I1130 11:06:51.558902 137274321021824 utils.py:1231] [34950] epoch = 27.934531563800817
+I1130 11:06:51.558956 137274321021824 utils.py:1231] [34950] img/sec/core = 170.08411916138775
+I1130 11:06:51.559014 137274321021824 utils.py:1231] [34950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.24371603973416
+I1130 11:06:51.559062 137274321021824 utils.py:1231] [34950] core_hours = 61.24371603973416
+I1130 11:06:51.559121 137274321021824 train.py:125] NOTE: Steps:34950/112603 [31.0%]
+Walltime:2d13h16m (0s eval)
+ETA:5d16h4m
+Total train time:8d5h19m
+I1130 11:12:03.348534 137274321021824 utils.py:1231] [35000] l2_params = 329.1957485127193
+I1130 11:12:03.348752 137274321021824 utils.py:1231] [35000] train/loss = 3.6430477499961853
+I1130 11:12:03.348854 137274321021824 utils.py:1231] [35000] l2_grads = 1.3020535707473755
+I1130 11:12:03.348921 137274321021824 utils.py:1231] [35000] lr = 0.0008605379941517158
+I1130 11:12:03.348972 137274321021824 utils.py:1231] [35000] uptime = 220912.711334454
+I1130 11:12:03.349024 137274321021824 utils.py:1231] [35000] examples_seen = 35840000.0
+I1130 11:12:03.349073 137274321021824 utils.py:1231] [35000] progress = 0.31082653215278455
+I1130 11:12:03.349120 137274321021824 utils.py:1231] [35000] epoch = 27.974495128269773
+I1130 11:12:03.349171 137274321021824 utils.py:1231] [35000] img/sec/core = 164.21297473458011
+I1130 11:12:03.349229 137274321021824 utils.py:1231] [35000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.33032443499166
+I1130 11:12:03.349280 137274321021824 utils.py:1231] [35000] core_hours = 61.33032443499166
+I1130 11:12:03.349341 137274321021824 train.py:125] NOTE: Steps:35000/112603 [31.1%]
+Walltime:2d13h21m (0s eval)
+ETA:5d15h59m
+Total train time:8d5h19m
+I1130 11:12:03.663853 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:35000/112603 [31.1%]
+Walltime:2d13h21m (0s eval)
+ETA:5d15h59m
+Total train time:8d5h19m
+I1130 11:13:33.244050 137274321021824 utils.py:1231] [35000] val/acc@1 = 0.5967195471938775
+I1130 11:13:33.244326 137274321021824 utils.py:1231] [35000] val/loss = 1.6753505775514914
+I1130 11:13:33.244499 137274321021824 utils.py:1231] [35000] z/secs/eval/val = 89.58039790598559
+I1130 11:13:33.244572 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 89.58039790598559
+I1130 11:18:36.510317 137274321021824 utils.py:1231] [35050] l2_params = 329.16561289532103
+I1130 11:18:36.510533 137274321021824 utils.py:1231] [35050] train/loss = 2.701222836971283
+I1130 11:18:36.510643 137274321021824 utils.py:1231] [35050] l2_grads = 1.327290654182434
+I1130 11:18:36.510712 137274321021824 utils.py:1231] [35050] lr = 0.0008600072093609602
+I1130 11:18:36.510770 137274321021824 utils.py:1231] [35050] uptime = 221305.873131415
+I1130 11:18:36.510836 137274321021824 utils.py:1231] [35050] examples_seen = 35891200.0
+I1130 11:18:36.510907 137274321021824 utils.py:1231] [35050] progress = 0.31127057005585995
+I1130 11:18:36.510988 137274321021824 utils.py:1231] [35050] epoch = 28.014458692738728
+I1130 11:18:36.511044 137274321021824 utils.py:1231] [35050] img/sec/core = 130.226284434949
+I1130 11:18:36.511103 137274321021824 utils.py:1231] [35050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.439536045258606
+I1130 11:18:36.511157 137274321021824 utils.py:1231] [35050] core_hours = 61.439536045258606
+I1130 11:18:36.511222 137274321021824 train.py:125] NOTE: Steps:35050/112603 [31.1%]
+Walltime:2d13h28m (0s eval)
+ETA:5d15h57m
+Total train time:8d5h23m
+I1130 11:23:43.250717 137274321021824 utils.py:1231] [35100] l2_params = 329.11533980412673
+I1130 11:23:43.250966 137274321021824 utils.py:1231] [35100] train/loss = 4.867147624492645
+I1130 11:23:43.251094 137274321021824 utils.py:1231] [35100] l2_grads = 1.0517940521240234
+I1130 11:23:43.251177 137274321021824 utils.py:1231] [35100] lr = 0.0008594755807872491
+I1130 11:23:43.251237 137274321021824 utils.py:1231] [35100] uptime = 221612.61359899398
+I1130 11:23:43.251288 137274321021824 utils.py:1231] [35100] examples_seen = 35942400.0
+I1130 11:23:43.251335 137274321021824 utils.py:1231] [35100] progress = 0.31171460795893535
+I1130 11:23:43.251384 137274321021824 utils.py:1231] [35100] epoch = 28.054422257207687
+I1130 11:23:43.251434 137274321021824 utils.py:1231] [35100] img/sec/core = 166.91635245948004
+I1130 11:23:43.251488 137274321021824 utils.py:1231] [35100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.524741730697215
+I1130 11:23:43.251538 137274321021824 utils.py:1231] [35100] core_hours = 61.524741730697215
+I1130 11:23:43.251595 137274321021824 train.py:125] NOTE: Steps:35100/112603 [31.2%]
+Walltime:2d13h33m (0s eval)
+ETA:5d15h51m
+Total train time:8d5h23m
+I1130 11:28:55.034883 137274321021824 utils.py:1231] [35150] l2_params = 329.10683604611546
+I1130 11:28:55.035150 137274321021824 utils.py:1231] [35150] train/loss = 3.047129362821579
+I1130 11:28:55.035252 137274321021824 utils.py:1231] [35150] l2_grads = 1.2417570352554321
+I1130 11:28:55.035320 137274321021824 utils.py:1231] [35150] lr = 0.0008589431096766096
+I1130 11:28:55.035376 137274321021824 utils.py:1231] [35150] uptime = 221924.397737718
+I1130 11:28:55.035437 137274321021824 utils.py:1231] [35150] examples_seen = 35993600.0
+I1130 11:28:55.035487 137274321021824 utils.py:1231] [35150] progress = 0.3121586458620108
+I1130 11:28:55.035535 137274321021824 utils.py:1231] [35150] epoch = 28.094385821676642
+I1130 11:28:55.035586 137274321021824 utils.py:1231] [35150] img/sec/core = 164.2161792114864
+I1130 11:28:55.035641 137274321021824 utils.py:1231] [35150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.61134843589833
+I1130 11:28:55.035694 137274321021824 utils.py:1231] [35150] core_hours = 61.61134843589833
+I1130 11:28:55.035754 137274321021824 train.py:125] NOTE: Steps:35150/112603 [31.2%]
+Walltime:2d13h38m (0s eval)
+ETA:5d15h46m
+Total train time:8d5h22m
+I1130 11:33:59.388128 137274321021824 utils.py:1231] [35200] l2_params = 329.0918127738618
+I1130 11:33:59.388338 137274321021824 utils.py:1231] [35200] train/loss = 2.5667761862277985
+I1130 11:33:59.388433 137274321021824 utils.py:1231] [35200] l2_grads = 1.3387150764465332
+I1130 11:33:59.388505 137274321021824 utils.py:1231] [35200] lr = 0.0008584097972770451
+I1130 11:33:59.388570 137274321021824 utils.py:1231] [35200] uptime = 222228.75092577998
+I1130 11:33:59.388643 137274321021824 utils.py:1231] [35200] examples_seen = 36044800.0
+I1130 11:33:59.388706 137274321021824 utils.py:1231] [35200] progress = 0.3126026837650862
+I1130 11:33:59.388780 137274321021824 utils.py:1231] [35200] epoch = 28.1343493861456
+I1130 11:33:59.388839 137274321021824 utils.py:1231] [35200] img/sec/core = 168.2256076436108
+I1130 11:33:59.388909 137274321021824 utils.py:1231] [35200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.69589098813777
+I1130 11:33:59.388963 137274321021824 utils.py:1231] [35200] core_hours = 61.69589098813777
+I1130 11:33:59.389028 137274321021824 train.py:125] NOTE: Steps:35200/112603 [31.3%]
+Walltime:2d13h43m (0s eval)
+ETA:5d15h40m
+Total train time:8d5h22m
+I1130 11:39:10.258926 137274321021824 utils.py:1231] [35250] l2_params = 329.0383687496944
+I1130 11:39:10.259196 137274321021824 utils.py:1231] [35250] train/loss = 2.7163407504558563
+I1130 11:39:10.259352 137274321021824 utils.py:1231] [35250] l2_grads = 1.3138172626495361
+I1130 11:39:10.259439 137274321021824 utils.py:1231] [35250] lr = 0.0008578756448385308
+I1130 11:39:10.259514 137274321021824 utils.py:1231] [35250] uptime = 222539.62187484198
+I1130 11:39:10.259581 137274321021824 utils.py:1231] [35250] examples_seen = 36096000.0
+I1130 11:39:10.259637 137274321021824 utils.py:1231] [35250] progress = 0.3130467216681616
+I1130 11:39:10.259693 137274321021824 utils.py:1231] [35250] epoch = 28.174312950614556
+I1130 11:39:10.259747 137274321021824 utils.py:1231] [35250] img/sec/core = 164.698567539
+I1130 11:39:10.259806 137274321021824 utils.py:1231] [35250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.78224402954388
+I1130 11:39:10.259856 137274321021824 utils.py:1231] [35250] core_hours = 61.78224402954388
+I1130 11:39:10.259928 137274321021824 train.py:125] NOTE: Steps:35250/112603 [31.3%]
+Walltime:2d13h48m (0s eval)
+ETA:5d15h35m
+Total train time:8d5h22m
+I1130 11:44:20.414542 137274321021824 utils.py:1231] [35300] l2_params = 329.021265923495
+I1130 11:44:20.414767 137274321021824 utils.py:1231] [35300] train/loss = 2.7239713966846466
+I1130 11:44:20.414918 137274321021824 utils.py:1231] [35300] l2_grads = 1.3990488052368164
+I1130 11:44:20.415024 137274321021824 utils.py:1231] [35300] lr = 0.0008573406536130098
+I1130 11:44:20.415112 137274321021824 utils.py:1231] [35300] uptime = 222849.77746969502
+I1130 11:44:20.415189 137274321021824 utils.py:1231] [35300] examples_seen = 36147200.0
+I1130 11:44:20.415275 137274321021824 utils.py:1231] [35300] progress = 0.313490759571237
+I1130 11:44:20.415360 137274321021824 utils.py:1231] [35300] epoch = 28.214276515083515
+I1130 11:44:20.415439 137274321021824 utils.py:1231] [35300] img/sec/core = 165.07843433959033
+I1130 11:44:20.415520 137274321021824 utils.py:1231] [35300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.8683983614475
+I1130 11:44:20.415599 137274321021824 utils.py:1231] [35300] core_hours = 61.8683983614475
+I1130 11:44:20.415695 137274321021824 train.py:125] NOTE: Steps:35300/112603 [31.3%]
+Walltime:2d13h54m (0s eval)
+ETA:5d15h29m
+Total train time:8d5h21m
+I1130 11:49:29.984796 137274321021824 utils.py:1231] [35350] l2_params = 328.9889560021786
+I1130 11:49:29.985052 137274321021824 utils.py:1231] [35350] train/loss = 2.761715829372406
+I1130 11:49:29.985197 137274321021824 utils.py:1231] [35350] l2_grads = 1.378290057182312
+I1130 11:49:29.985295 137274321021824 utils.py:1231] [35350] lr = 0.0008568048248543921
+I1130 11:49:29.985358 137274321021824 utils.py:1231] [35350] uptime = 223159.34771942702
+I1130 11:49:29.985423 137274321021824 utils.py:1231] [35350] examples_seen = 36198400.0
+I1130 11:49:29.985478 137274321021824 utils.py:1231] [35350] progress = 0.3139347974743124
+I1130 11:49:29.985544 137274321021824 utils.py:1231] [35350] epoch = 28.25424007955247
+I1130 11:49:29.985600 137274321021824 utils.py:1231] [35350] img/sec/core = 165.39056981193895
+I1130 11:49:29.985661 137274321021824 utils.py:1231] [35350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 61.954390097484165
+I1130 11:49:29.985715 137274321021824 utils.py:1231] [35350] core_hours = 61.954390097484165
+I1130 11:49:29.985800 137274321021824 train.py:125] NOTE: Steps:35350/112603 [31.4%]
+Walltime:2d13h59m (0s eval)
+ETA:5d15h24m
+Total train time:8d5h21m
+I1130 11:54:36.935127 137274321021824 utils.py:1231] [35400] l2_params = 328.95273642484636
+I1130 11:54:36.935349 137274321021824 utils.py:1231] [35400] train/loss = 2.73575896024704
+I1130 11:54:36.935453 137274321021824 utils.py:1231] [35400] l2_grads = 1.4218006134033203
+I1130 11:54:36.935525 137274321021824 utils.py:1231] [35400] lr = 0.0008562681598185502
+I1130 11:54:36.935590 137274321021824 utils.py:1231] [35400] uptime = 223466.29795153998
+I1130 11:54:36.935657 137274321021824 utils.py:1231] [35400] examples_seen = 36249600.0
+I1130 11:54:36.935714 137274321021824 utils.py:1231] [35400] progress = 0.3143788353773878
+I1130 11:54:36.935782 137274321021824 utils.py:1231] [35400] epoch = 28.294203644021426
+I1130 11:54:36.935840 137274321021824 utils.py:1231] [35400] img/sec/core = 166.80228468162233
+I1130 11:54:36.935908 137274321021824 utils.py:1231] [35400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.03965405084888
+I1130 11:54:36.935966 137274321021824 utils.py:1231] [35400] core_hours = 62.03965405084888
+I1130 11:54:36.936054 137274321021824 train.py:125] NOTE: Steps:35400/112603 [31.4%]
+Walltime:2d14h4m (0s eval)
+ETA:5d15h18m
+Total train time:8d5h21m
+I1130 11:59:45.450732 137274321021824 utils.py:1231] [35450] l2_params = 328.9173140425014
+I1130 11:59:45.450966 137274321021824 utils.py:1231] [35450] train/loss = 3.046400338411331
+I1130 11:59:45.451070 137274321021824 utils.py:1231] [35450] l2_grads = 1.250544786453247
+I1130 11:59:45.451137 137274321021824 utils.py:1231] [35450] lr = 0.0008557306597633163
+I1130 11:59:45.451196 137274321021824 utils.py:1231] [35450] uptime = 223774.81355832802
+I1130 11:59:45.451259 137274321021824 utils.py:1231] [35450] examples_seen = 36300800.0
+I1130 11:59:45.451312 137274321021824 utils.py:1231] [35450] progress = 0.3148228732804632
+I1130 11:59:45.451361 137274321021824 utils.py:1231] [35450] epoch = 28.334167208490385
+I1130 11:59:45.451411 137274321021824 utils.py:1231] [35450] img/sec/core = 165.9559480087501
+I1130 11:59:45.451465 137274321021824 utils.py:1231] [35450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.12535283051222
+I1130 11:59:45.451514 137274321021824 utils.py:1231] [35450] core_hours = 62.12535283051222
+I1130 11:59:45.451572 137274321021824 train.py:125] NOTE: Steps:35450/112603 [31.5%]
+Walltime:2d14h9m (0s eval)
+ETA:5d15h12m
+Total train time:8d5h20m
+I1130 12:04:50.765449 137274321021824 utils.py:1231] [35500] l2_params = 328.872789300925
+I1130 12:04:50.765712 137274321021824 utils.py:1231] [35500] train/loss = 2.5771273374557495
+I1130 12:04:50.765820 137274321021824 utils.py:1231] [35500] l2_grads = 1.4601508378982544
+I1130 12:04:50.765899 137274321021824 utils.py:1231] [35500] lr = 0.0008551923259484803
+I1130 12:04:50.765964 137274321021824 utils.py:1231] [35500] uptime = 224080.12832283397
+I1130 12:04:50.766023 137274321021824 utils.py:1231] [35500] examples_seen = 36352000.0
+I1130 12:04:50.766083 137274321021824 utils.py:1231] [35500] progress = 0.3152669111835386
+I1130 12:04:50.766141 137274321021824 utils.py:1231] [35500] epoch = 28.37413077295934
+I1130 12:04:50.766204 137274321021824 utils.py:1231] [35500] img/sec/core = 167.6957879283863
+I1130 12:04:50.766266 137274321021824 utils.py:1231] [35500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.21016248731943
+I1130 12:04:50.766322 137274321021824 utils.py:1231] [35500] core_hours = 62.21016248731943
+I1130 12:04:50.766390 137274321021824 train.py:125] NOTE: Steps:35500/112603 [31.5%]
+Walltime:2d14h14m (0s eval)
+ETA:5d15h7m
+Total train time:8d5h20m
+I1130 12:09:59.513861 137274321021824 utils.py:1231] [35550] l2_params = 328.8583836569765
+I1130 12:09:59.514163 137274321021824 utils.py:1231] [35550] train/loss = 2.76730740070343
+I1130 12:09:59.514345 137274321021824 utils.py:1231] [35550] l2_grads = 1.3957303762435913
+I1130 12:09:59.514443 137274321021824 utils.py:1231] [35550] lr = 0.0008546531596357863
+I1130 12:09:59.514512 137274321021824 utils.py:1231] [35550] uptime = 224388.87687289598
+I1130 12:09:59.514572 137274321021824 utils.py:1231] [35550] examples_seen = 36403200.0
+I1130 12:09:59.514635 137274321021824 utils.py:1231] [35550] progress = 0.315710949086614
+I1130 12:09:59.514703 137274321021824 utils.py:1231] [35550] epoch = 28.4140943374283
+I1130 12:09:59.514761 137274321021824 utils.py:1231] [35550] img/sec/core = 165.83073828109283
+I1130 12:09:59.514830 137274321021824 utils.py:1231] [35550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.29592597344777
+I1130 12:09:59.514902 137274321021824 utils.py:1231] [35550] core_hours = 62.29592597344777
+I1130 12:09:59.514969 137274321021824 train.py:125] NOTE: Steps:35550/112603 [31.6%]
+Walltime:2d14h19m (0s eval)
+ETA:5d15h1m
+Total train time:8d5h19m
+I1130 12:15:11.293599 137274321021824 utils.py:1231] [35600] l2_params = 328.7868491281501
+I1130 12:15:11.293802 137274321021824 utils.py:1231] [35600] train/loss = 2.6367018818855286
+I1130 12:15:11.293913 137274321021824 utils.py:1231] [35600] l2_grads = 1.362985610961914
+I1130 12:15:11.293974 137274321021824 utils.py:1231] [35600] lr = 0.0008541131620889297
+I1130 12:15:11.294026 137274321021824 utils.py:1231] [35600] uptime = 224700.656388122
+I1130 12:15:11.294078 137274321021824 utils.py:1231] [35600] examples_seen = 36454400.0
+I1130 12:15:11.294128 137274321021824 utils.py:1231] [35600] progress = 0.3161549869896894
+I1130 12:15:11.294178 137274321021824 utils.py:1231] [35600] epoch = 28.454057901897254
+I1130 12:15:11.294229 137274321021824 utils.py:1231] [35600] img/sec/core = 164.21861443619125
+I1130 12:15:11.294293 137274321021824 utils.py:1231] [35600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.38253139434388
+I1130 12:15:11.294345 137274321021824 utils.py:1231] [35600] core_hours = 62.38253139434388
+I1130 12:15:11.294407 137274321021824 train.py:125] NOTE: Steps:35600/112603 [31.6%]
+Walltime:2d14h25m (0s eval)
+ETA:5d14h56m
+Total train time:8d5h19m
+I1130 12:20:23.070546 137274321021824 utils.py:1231] [35650] l2_params = 328.7429480823434
+I1130 12:20:23.070779 137274321021824 utils.py:1231] [35650] train/loss = 3.841162919998169
+I1130 12:20:23.070924 137274321021824 utils.py:1231] [35650] l2_grads = 1.183448076248169
+I1130 12:20:23.071023 137274321021824 utils.py:1231] [35650] lr = 0.0008535723345735527
+I1130 12:20:23.071100 137274321021824 utils.py:1231] [35650] uptime = 225012.43345757702
+I1130 12:20:23.071199 137274321021824 utils.py:1231] [35650] examples_seen = 36505600.0
+I1130 12:20:23.071283 137274321021824 utils.py:1231] [35650] progress = 0.31659902489276487
+I1130 12:20:23.071358 137274321021824 utils.py:1231] [35650] epoch = 28.49402146636621
+I1130 12:20:23.071440 137274321021824 utils.py:1231] [35650] img/sec/core = 164.21990266794722
+I1130 12:20:23.071522 137274321021824 utils.py:1231] [35650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.46913613585916
+I1130 12:20:23.071593 137274321021824 utils.py:1231] [35650] core_hours = 62.46913613585916
+I1130 12:20:23.071675 137274321021824 train.py:125] NOTE: Steps:35650/112603 [31.7%]
+Walltime:2d14h30m (0s eval)
+ETA:5d14h51m
+Total train time:8d5h19m
+I1130 12:25:29.930574 137274321021824 utils.py:1231] [35700] l2_params = 328.7093087039152
+I1130 12:25:29.930815 137274321021824 utils.py:1231] [35700] train/loss = 5.121833860874176
+I1130 12:25:29.930920 137274321021824 utils.py:1231] [35700] l2_grads = 1.2279984951019287
+I1130 12:25:29.930992 137274321021824 utils.py:1231] [35700] lr = 0.0008530306783572445
+I1130 12:25:29.931054 137274321021824 utils.py:1231] [35700] uptime = 225319.293416136
+I1130 12:25:29.931123 137274321021824 utils.py:1231] [35700] examples_seen = 36556800.0
+I1130 12:25:29.931191 137274321021824 utils.py:1231] [35700] progress = 0.31704306279584027
+I1130 12:25:29.931246 137274321021824 utils.py:1231] [35700] epoch = 28.53398503083517
+I1130 12:25:29.931301 137274321021824 utils.py:1231] [35700] img/sec/core = 166.85135538840152
+I1130 12:25:29.931361 137274321021824 utils.py:1231] [35700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.55437501323666
+I1130 12:25:29.931416 137274321021824 utils.py:1231] [35700] core_hours = 62.55437501323666
+I1130 12:25:29.931487 137274321021824 train.py:125] NOTE: Steps:35700/112603 [31.7%]
+Walltime:2d14h35m (0s eval)
+ETA:5d14h45m
+Total train time:8d5h18m
+I1130 12:30:41.703255 137274321021824 utils.py:1231] [35750] l2_params = 328.66829873839896
+I1130 12:30:41.703495 137274321021824 utils.py:1231] [35750] train/loss = 2.7446843683719635
+I1130 12:30:41.703600 137274321021824 utils.py:1231] [35750] l2_grads = 1.3652992248535156
+I1130 12:30:41.703665 137274321021824 utils.py:1231] [35750] lr = 0.000852488194709536
+I1130 12:30:41.703719 137274321021824 utils.py:1231] [35750] uptime = 225631.06608137698
+I1130 12:30:41.703773 137274321021824 utils.py:1231] [35750] examples_seen = 36608000.0
+I1130 12:30:41.703825 137274321021824 utils.py:1231] [35750] progress = 0.31748710069891567
+I1130 12:30:41.703875 137274321021824 utils.py:1231] [35750] epoch = 28.573948595304124
+I1130 12:30:41.703938 137274321021824 utils.py:1231] [35750] img/sec/core = 164.22222249800606
+I1130 12:30:41.703999 137274321021824 utils.py:1231] [35750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.64097853135915
+I1130 12:30:41.704052 137274321021824 utils.py:1231] [35750] core_hours = 62.64097853135915
+I1130 12:30:41.704115 137274321021824 train.py:125] NOTE: Steps:35750/112603 [31.7%]
+Walltime:2d14h40m (0s eval)
+ETA:5d14h40m
+Total train time:8d5h18m
+I1130 12:35:47.836177 137274321021824 utils.py:1231] [35800] l2_params = 328.6549901088466
+I1130 12:35:47.836472 137274321021824 utils.py:1231] [35800] train/loss = 4.368241727352142
+I1130 12:35:47.836650 137274321021824 utils.py:1231] [35800] l2_grads = 1.0575599670410156
+I1130 12:35:47.836730 137274321021824 utils.py:1231] [35800] lr = 0.0008519448849018982
+I1130 12:35:47.836799 137274321021824 utils.py:1231] [35800] uptime = 225937.19915996998
+I1130 12:35:47.836873 137274321021824 utils.py:1231] [35800] examples_seen = 36659200.0
+I1130 12:35:47.836968 137274321021824 utils.py:1231] [35800] progress = 0.31793113860199107
+I1130 12:35:47.837049 137274321021824 utils.py:1231] [35800] epoch = 28.613912159773083
+I1130 12:35:47.837110 137274321021824 utils.py:1231] [35800] img/sec/core = 167.24752592995372
+I1130 12:35:47.837182 137274321021824 utils.py:1231] [35800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.72601549763499
+I1130 12:35:47.837248 137274321021824 utils.py:1231] [35800] core_hours = 62.72601549763499
+I1130 12:35:47.837326 137274321021824 train.py:125] NOTE: Steps:35800/112603 [31.8%]
+Walltime:2d14h45m (0s eval)
+ETA:5d14h34m
+Total train time:8d5h18m
+I1130 12:40:59.698228 137274321021824 utils.py:1231] [35850] l2_params = 328.63005335230037
+I1130 12:40:59.698465 137274321021824 utils.py:1231] [35850] train/loss = 2.782854914665222
+I1130 12:40:59.698590 137274321021824 utils.py:1231] [35850] l2_grads = 1.3285499811172485
+I1130 12:40:59.698705 137274321021824 utils.py:1231] [35850] lr = 0.0008514007502077368
+I1130 12:40:59.698766 137274321021824 utils.py:1231] [35850] uptime = 226249.06112778396
+I1130 12:40:59.698824 137274321021824 utils.py:1231] [35850] examples_seen = 36710400.0
+I1130 12:40:59.698878 137274321021824 utils.py:1231] [35850] progress = 0.31837517650506647
+I1130 12:40:59.698934 137274321021824 utils.py:1231] [35850] epoch = 28.653875724242038
+I1130 12:40:59.698998 137274321021824 utils.py:1231] [35850] img/sec/core = 164.17519699144535
+I1130 12:40:59.699064 137274321021824 utils.py:1231] [35850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.812643822027766
+I1130 12:40:59.699142 137274321021824 utils.py:1231] [35850] core_hours = 62.812643822027766
+I1130 12:40:59.699215 137274321021824 train.py:125] NOTE: Steps:35850/112603 [31.8%]
+Walltime:2d14h50m (0s eval)
+ETA:5d14h29m
+Total train time:8d5h18m
+I1130 12:46:11.469925 137274321021824 utils.py:1231] [35900] l2_params = 328.58615771196264
+I1130 12:46:11.470126 137274321021824 utils.py:1231] [35900] train/loss = 4.103965252637863
+I1130 12:46:11.470226 137274321021824 utils.py:1231] [35900] l2_grads = 1.2428059577941895
+I1130 12:46:11.470289 137274321021824 utils.py:1231] [35900] lr = 0.0008508557919023919
+I1130 12:46:11.470354 137274321021824 utils.py:1231] [35900] uptime = 226560.83271600597
+I1130 12:46:11.470405 137274321021824 utils.py:1231] [35900] examples_seen = 36761600.0
+I1130 12:46:11.470453 137274321021824 utils.py:1231] [35900] progress = 0.31881921440814187
+I1130 12:46:11.470500 137274321021824 utils.py:1231] [35900] epoch = 28.693839288710997
+I1130 12:46:11.470550 137274321021824 utils.py:1231] [35900] img/sec/core = 164.22278980578997
+I1130 12:46:11.470604 137274321021824 utils.py:1231] [35900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.899247040978324
+I1130 12:46:11.470653 137274321021824 utils.py:1231] [35900] core_hours = 62.899247040978324
+I1130 12:46:11.470711 137274321021824 train.py:125] NOTE: Steps:35900/112603 [31.9%]
+Walltime:2d14h56m (0s eval)
+ETA:5d14h23m
+Total train time:8d5h17m
+I1130 12:51:20.855419 137274321021824 utils.py:1231] [35950] l2_params = 328.52469188493075
+I1130 12:51:20.855612 137274321021824 utils.py:1231] [35950] train/loss = 5.0119739174842834
+I1130 12:51:20.855703 137274321021824 utils.py:1231] [35950] l2_grads = 1.2915757894515991
+I1130 12:51:20.855762 137274321021824 utils.py:1231] [35950] lr = 0.0008503100112631335
+I1130 12:51:20.855811 137274321021824 utils.py:1231] [35950] uptime = 226870.21817307099
+I1130 12:51:20.855861 137274321021824 utils.py:1231] [35950] examples_seen = 36812800.0
+I1130 12:51:20.855913 137274321021824 utils.py:1231] [35950] progress = 0.31926325231121727
+I1130 12:51:20.855959 137274321021824 utils.py:1231] [35950] epoch = 28.733802853179952
+I1130 12:51:20.856008 137274321021824 utils.py:1231] [35950] img/sec/core = 165.48935585307996
+I1130 12:51:20.856061 137274321021824 utils.py:1231] [35950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 62.9851874457186
+I1130 12:51:20.856109 137274321021824 utils.py:1231] [35950] core_hours = 62.9851874457186
+I1130 12:51:20.856168 137274321021824 train.py:125] NOTE: Steps:35950/112603 [31.9%]
+Walltime:2d15h1m (0s eval)
+ETA:5d14h18m
+Total train time:8d5h17m
+I1130 12:56:32.644879 137274321021824 utils.py:1231] [36000] l2_params = 328.4422596216364
+I1130 12:56:32.645079 137274321021824 utils.py:1231] [36000] train/loss = 5.252499461174011
+I1130 12:56:32.645187 137274321021824 utils.py:1231] [36000] l2_grads = 1.2588540315628052
+I1130 12:56:32.645257 137274321021824 utils.py:1231] [36000] lr = 0.00084976340956916
+I1130 12:56:32.645318 137274321021824 utils.py:1231] [36000] uptime = 227182.007680976
+I1130 12:56:32.645382 137274321021824 utils.py:1231] [36000] examples_seen = 36864000.0
+I1130 12:56:32.645430 137274321021824 utils.py:1231] [36000] progress = 0.31970729021429267
+I1130 12:56:32.645483 137274321021824 utils.py:1231] [36000] epoch = 28.773766417648908
+I1130 12:56:32.645547 137274321021824 utils.py:1231] [36000] img/sec/core = 164.2133513216127
+I1130 12:56:32.645604 137274321021824 utils.py:1231] [36000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.071795642358886
+I1130 12:56:32.645648 137274321021824 utils.py:1231] [36000] core_hours = 63.071795642358886
+I1130 12:56:32.645703 137274321021824 train.py:125] NOTE: Steps:36000/112603 [32.0%]
+Walltime:2d15h6m (0s eval)
+ETA:5d14h12m
+Total train time:8d5h17m
+I1130 13:01:42.632777 137274321021824 utils.py:1231] [36050] l2_params = 328.40018254979526
+I1130 13:01:42.633016 137274321021824 utils.py:1231] [36050] train/loss = 5.014599025249481
+I1130 13:01:42.633151 137274321021824 utils.py:1231] [36050] l2_grads = 1.1076533794403076
+I1130 13:01:42.633222 137274321021824 utils.py:1231] [36050] lr = 0.0008492159881015932
+I1130 13:01:42.633279 137274321021824 utils.py:1231] [36050] uptime = 227491.995640883
+I1130 13:01:42.633341 137274321021824 utils.py:1231] [36050] examples_seen = 36915200.0
+I1130 13:01:42.633391 137274321021824 utils.py:1231] [36050] progress = 0.3201513281173681
+I1130 13:01:42.633441 137274321021824 utils.py:1231] [36050] epoch = 28.813729982117867
+I1130 13:01:42.633499 137274321021824 utils.py:1231] [36050] img/sec/core = 165.16770527269333
+I1130 13:01:42.633567 137274321021824 utils.py:1231] [36050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.157903408999715
+I1130 13:01:42.633623 137274321021824 utils.py:1231] [36050] core_hours = 63.157903408999715
+I1130 13:01:42.633683 137274321021824 train.py:125] NOTE: Steps:36050/112603 [32.0%]
+Walltime:2d15h11m (0s eval)
+ETA:5d14h7m
+Total train time:8d5h17m
+I1130 13:06:54.423875 137274321021824 utils.py:1231] [36100] l2_params = 328.34956321315485
+I1130 13:06:54.424140 137274321021824 utils.py:1231] [36100] train/loss = 2.8445582687854767
+I1130 13:06:54.424265 137274321021824 utils.py:1231] [36100] l2_grads = 1.3345226049423218
+I1130 13:06:54.424352 137274321021824 utils.py:1231] [36100] lr = 0.0008486677481434762
+I1130 13:06:54.424401 137274321021824 utils.py:1231] [36100] uptime = 227803.78676375496
+I1130 13:06:54.424449 137274321021824 utils.py:1231] [36100] examples_seen = 36966400.0
+I1130 13:06:54.424495 137274321021824 utils.py:1231] [36100] progress = 0.32059536602044353
+I1130 13:06:54.424539 137274321021824 utils.py:1231] [36100] epoch = 28.853693546586822
+I1130 13:06:54.424586 137274321021824 utils.py:1231] [36100] img/sec/core = 164.21250075494714
+I1130 13:06:54.424637 137274321021824 utils.py:1231] [36100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.24451205424193
+I1130 13:06:54.424684 137274321021824 utils.py:1231] [36100] core_hours = 63.24451205424193
+I1130 13:06:54.424740 137274321021824 train.py:125] NOTE: Steps:36100/112603 [32.1%]
+Walltime:2d15h16m (0s eval)
+ETA:5d14h2m
+Total train time:8d5h16m
+I1130 13:12:04.347756 137274321021824 utils.py:1231] [36150] l2_params = 328.3133800792717
+I1130 13:12:04.347978 137274321021824 utils.py:1231] [36150] train/loss = 3.927648901939392
+I1130 13:12:04.348087 137274321021824 utils.py:1231] [36150] l2_grads = 1.1293779611587524
+I1130 13:12:04.348153 137274321021824 utils.py:1231] [36150] lr = 0.0008481186909797721
+I1130 13:12:04.348226 137274321021824 utils.py:1231] [36150] uptime = 228113.71058628202
+I1130 13:12:04.348307 137274321021824 utils.py:1231] [36150] examples_seen = 37017600.0
+I1130 13:12:04.348362 137274321021824 utils.py:1231] [36150] progress = 0.32103940392351893
+I1130 13:12:04.348463 137274321021824 utils.py:1231] [36150] epoch = 28.89365711105578
+I1130 13:12:04.348542 137274321021824 utils.py:1231] [36150] img/sec/core = 165.20188600710364
+I1130 13:12:04.348619 137274321021824 utils.py:1231] [36150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.33060200494389
+I1130 13:12:04.348691 137274321021824 utils.py:1231] [36150] core_hours = 63.33060200494389
+I1130 13:12:04.348785 137274321021824 train.py:125] NOTE: Steps:36150/112603 [32.1%]
+Walltime:2d15h21m (0s eval)
+ETA:5d13h56m
+Total train time:8d5h16m
+I1130 13:17:14.198758 137274321021824 utils.py:1231] [36200] l2_params = 328.2922553512251
+I1130 13:17:14.198984 137274321021824 utils.py:1231] [36200] train/loss = 2.5750844478607178
+I1130 13:17:14.199080 137274321021824 utils.py:1231] [36200] l2_grads = 1.41749107837677
+I1130 13:17:14.199142 137274321021824 utils.py:1231] [36200] lr = 0.0008475688178973571
+I1130 13:17:14.199210 137274321021824 utils.py:1231] [36200] uptime = 228423.561567472
+I1130 13:17:14.199265 137274321021824 utils.py:1231] [36200] examples_seen = 37068800.0
+I1130 13:17:14.199326 137274321021824 utils.py:1231] [36200] progress = 0.32148344182659433
+I1130 13:17:14.199385 137274321021824 utils.py:1231] [36200] epoch = 28.933620675524736
+I1130 13:17:14.199435 137274321021824 utils.py:1231] [36200] img/sec/core = 165.24072250268787
+I1130 13:17:14.199491 137274321021824 utils.py:1231] [36200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.416671721941114
+I1130 13:17:14.199538 137274321021824 utils.py:1231] [36200] core_hours = 63.416671721941114
+I1130 13:17:14.199596 137274321021824 train.py:125] NOTE: Steps:36200/112603 [32.1%]
+Walltime:2d15h27m (0s eval)
+ETA:5d13h51m
+Total train time:8d5h16m
+I1130 13:22:26.005622 137274321021824 utils.py:1231] [36250] l2_params = 328.26853763772164
+I1130 13:22:26.005893 137274321021824 utils.py:1231] [36250] train/loss = 2.641763836145401
+I1130 13:22:26.006023 137274321021824 utils.py:1231] [36250] l2_grads = 1.33711576461792
+I1130 13:22:26.006114 137274321021824 utils.py:1231] [36250] lr = 0.000847018130185021
+I1130 13:22:26.006180 137274321021824 utils.py:1231] [36250] uptime = 228735.36854122498
+I1130 13:22:26.006235 137274321021824 utils.py:1231] [36250] examples_seen = 37120000.0
+I1130 13:22:26.006286 137274321021824 utils.py:1231] [36250] progress = 0.32192747972966973
+I1130 13:22:26.006339 137274321021824 utils.py:1231] [36250] epoch = 28.973584239993695
+I1130 13:22:26.006394 137274321021824 utils.py:1231] [36250] img/sec/core = 164.20415292110445
+I1130 13:22:26.006459 137274321021824 utils.py:1231] [36250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.50328477020582
+I1130 13:22:26.006513 137274321021824 utils.py:1231] [36250] core_hours = 63.50328477020582
+I1130 13:22:26.006583 137274321021824 train.py:125] NOTE: Steps:36250/112603 [32.2%]
+Walltime:2d15h32m (0s eval)
+ETA:5d13h45m
+Total train time:8d5h16m
+I1130 13:27:37.787154 137274321021824 utils.py:1231] [36300] l2_params = 328.2057876580607
+I1130 13:27:37.787375 137274321021824 utils.py:1231] [36300] train/loss = 2.924872636795044
+I1130 13:27:37.787484 137274321021824 utils.py:1231] [36300] l2_grads = 1.2157849073410034
+I1130 13:27:37.787567 137274321021824 utils.py:1231] [36300] lr = 0.0008464666291334619
+I1130 13:27:37.787634 137274321021824 utils.py:1231] [36300] uptime = 229047.149995716
+I1130 13:27:37.787707 137274321021824 utils.py:1231] [36300] examples_seen = 37171200.0
+I1130 13:27:37.787759 137274321021824 utils.py:1231] [36300] progress = 0.32237151763274513
+I1130 13:27:37.787811 137274321021824 utils.py:1231] [36300] epoch = 29.01354780446265
+I1130 13:27:37.787864 137274321021824 utils.py:1231] [36300] img/sec/core = 164.21759300463592
+I1130 13:27:37.787935 137274321021824 utils.py:1231] [36300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.58989072978667
+I1130 13:27:37.788003 137274321021824 utils.py:1231] [36300] core_hours = 63.58989072978667
+I1130 13:27:37.788069 137274321021824 train.py:125] NOTE: Steps:36300/112603 [32.2%]
+Walltime:2d15h37m (0s eval)
+ETA:5d13h40m
+Total train time:8d5h16m
+I1130 13:32:49.563053 137274321021824 utils.py:1231] [36350] l2_params = 328.19919654294836
+I1130 13:32:49.563311 137274321021824 utils.py:1231] [36350] train/loss = 2.690680891275406
+I1130 13:32:49.563448 137274321021824 utils.py:1231] [36350] l2_grads = 1.4378758668899536
+I1130 13:32:49.563539 137274321021824 utils.py:1231] [36350] lr = 0.0008459143160352867
+I1130 13:32:49.563614 137274321021824 utils.py:1231] [36350] uptime = 229358.92597542197
+I1130 13:32:49.563687 137274321021824 utils.py:1231] [36350] examples_seen = 37222400.0
+I1130 13:32:49.563743 137274321021824 utils.py:1231] [36350] progress = 0.32281555553582053
+I1130 13:32:49.563797 137274321021824 utils.py:1231] [36350] epoch = 29.053511368931606
+I1130 13:32:49.563868 137274321021824 utils.py:1231] [36350] img/sec/core = 164.22047666496238
+I1130 13:32:49.563953 137274321021824 utils.py:1231] [36350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.676495168593874
+I1130 13:32:49.564012 137274321021824 utils.py:1231] [36350] core_hours = 63.676495168593874
+I1130 13:32:49.564075 137274321021824 train.py:125] NOTE: Steps:36350/112603 [32.3%]
+Walltime:2d15h42m (0s eval)
+ETA:5d13h35m
+Total train time:8d5h15m
+I1130 13:38:01.341528 137274321021824 utils.py:1231] [36400] l2_params = 328.14923300296005
+I1130 13:38:01.341769 137274321021824 utils.py:1231] [36400] train/loss = 2.591905117034912
+I1130 13:38:01.341866 137274321021824 utils.py:1231] [36400] l2_grads = 1.500071406364441
+I1130 13:38:01.341933 137274321021824 utils.py:1231] [36400] lr = 0.0008453611921850025
+I1130 13:38:01.341984 137274321021824 utils.py:1231] [36400] uptime = 229670.704346126
+I1130 13:38:01.342035 137274321021824 utils.py:1231] [36400] examples_seen = 37273600.0
+I1130 13:38:01.342083 137274321021824 utils.py:1231] [36400] progress = 0.32325959343889593
+I1130 13:38:01.342131 137274321021824 utils.py:1231] [36400] epoch = 29.093474933400564
+I1130 13:38:01.342182 137274321021824 utils.py:1231] [36400] img/sec/core = 164.21921727406468
+I1130 13:38:01.342238 137274321021824 utils.py:1231] [36400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.76310027156722
+I1130 13:38:01.342290 137274321021824 utils.py:1231] [36400] core_hours = 63.76310027156722
+I1130 13:38:01.342351 137274321021824 train.py:125] NOTE: Steps:36400/112603 [32.3%]
+Walltime:2d15h47m (0s eval)
+ETA:5d13h29m
+Total train time:8d5h15m
+I1130 13:43:13.117876 137274321021824 utils.py:1231] [36450] l2_params = 328.1225601282688
+I1130 13:43:13.118122 137274321021824 utils.py:1231] [36450] train/loss = 2.5135077834129333
+I1130 13:43:13.118242 137274321021824 utils.py:1231] [36450] l2_grads = 1.3331128358840942
+I1130 13:43:13.118309 137274321021824 utils.py:1231] [36450] lr = 0.0008448072588790184
+I1130 13:43:13.118361 137274321021824 utils.py:1231] [36450] uptime = 229982.480723775
+I1130 13:43:13.118417 137274321021824 utils.py:1231] [36450] examples_seen = 37324800.0
+I1130 13:43:13.118465 137274321021824 utils.py:1231] [36450] progress = 0.32370363134197133
+I1130 13:43:13.118513 137274321021824 utils.py:1231] [36450] epoch = 29.13343849786952
+I1130 13:43:13.118567 137274321021824 utils.py:1231] [36450] img/sec/core = 164.2202670583441
+I1130 13:43:13.118622 137274321021824 utils.py:1231] [36450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.84970482091416
+I1130 13:43:13.118670 137274321021824 utils.py:1231] [36450] core_hours = 63.84970482091416
+I1130 13:43:13.118730 137274321021824 train.py:125] NOTE: Steps:36450/112603 [32.4%]
+Walltime:2d15h53m (0s eval)
+ETA:5d13h24m
+Total train time:8d5h15m
+I1130 13:48:23.056274 137274321021824 utils.py:1231] [36500] l2_params = 328.09733330390077
+I1130 13:48:23.056539 137274321021824 utils.py:1231] [36500] train/loss = 3.2273674309253693
+I1130 13:48:23.056716 137274321021824 utils.py:1231] [36500] l2_grads = 1.2643028497695923
+I1130 13:48:23.056794 137274321021824 utils.py:1231] [36500] lr = 0.0008442525174156404
+I1130 13:48:23.056852 137274321021824 utils.py:1231] [36500] uptime = 230292.41921449598
+I1130 13:48:23.056913 137274321021824 utils.py:1231] [36500] examples_seen = 37376000.0
+I1130 13:48:23.056965 137274321021824 utils.py:1231] [36500] progress = 0.32414766924504673
+I1130 13:48:23.057014 137274321021824 utils.py:1231] [36500] epoch = 29.17340206233848
+I1130 13:48:23.057066 137274321021824 utils.py:1231] [36500] img/sec/core = 165.19406763870077
+I1130 13:48:23.057121 137274321021824 utils.py:1231] [36500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 63.93579884611443
+I1130 13:48:23.057169 137274321021824 utils.py:1231] [36500] core_hours = 63.93579884611443
+I1130 13:48:23.057228 137274321021824 train.py:125] NOTE: Steps:36500/112603 [32.4%]
+Walltime:2d15h58m (0s eval)
+ETA:5d13h18m
+Total train time:8d5h15m
+I1130 13:53:34.840032 137274321021824 utils.py:1231] [36550] l2_params = 328.0300860233048
+I1130 13:53:34.840259 137274321021824 utils.py:1231] [36550] train/loss = 2.711941570043564
+I1130 13:53:34.840387 137274321021824 utils.py:1231] [36550] l2_grads = 1.3880510330200195
+I1130 13:53:34.840471 137274321021824 utils.py:1231] [36550] lr = 0.0008436969690950684
+I1130 13:53:34.840535 137274321021824 utils.py:1231] [36550] uptime = 230604.202896144
+I1130 13:53:34.840603 137274321021824 utils.py:1231] [36550] examples_seen = 37427200.0
+I1130 13:53:34.840660 137274321021824 utils.py:1231] [36550] progress = 0.3245917071481222
+I1130 13:53:34.840717 137274321021824 utils.py:1231] [36550] epoch = 29.213365626807434
+I1130 13:53:34.840774 137274321021824 utils.py:1231] [36550] img/sec/core = 164.21641995298458
+I1130 13:53:34.840837 137274321021824 utils.py:1231] [36550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.02240542435
+I1130 13:53:34.840905 137274321021824 utils.py:1231] [36550] core_hours = 64.02240542435
+I1130 13:53:34.840975 137274321021824 train.py:125] NOTE: Steps:36550/112603 [32.5%]
+Walltime:2d16h3m (0s eval)
+ETA:5d13h13m
+Total train time:8d5h15m
+I1130 13:58:41.914011 137274321021824 utils.py:1231] [36600] l2_params = 328.0060081672173
+I1130 13:58:41.914302 137274321021824 utils.py:1231] [36600] train/loss = 3.766901880502701
+I1130 13:58:41.914465 137274321021824 utils.py:1231] [36600] l2_grads = 1.1904851198196411
+I1130 13:58:41.914524 137274321021824 utils.py:1231] [36600] lr = 0.0008431406152193937
+I1130 13:58:41.914572 137274321021824 utils.py:1231] [36600] uptime = 230911.276935224
+I1130 13:58:41.914623 137274321021824 utils.py:1231] [36600] examples_seen = 37478400.0
+I1130 13:58:41.914669 137274321021824 utils.py:1231] [36600] progress = 0.3250357450511976
+I1130 13:58:41.914719 137274321021824 utils.py:1231] [36600] epoch = 29.25332919127639
+I1130 13:58:41.914768 137274321021824 utils.py:1231] [36600] img/sec/core = 166.7350328715358
+I1130 13:58:41.914822 137274321021824 utils.py:1231] [36600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.1077037685389
+I1130 13:58:41.914869 137274321021824 utils.py:1231] [36600] core_hours = 64.1077037685389
+I1130 13:58:41.914935 137274321021824 train.py:125] NOTE: Steps:36600/112603 [32.5%]
+Walltime:2d16h8m (0s eval)
+ETA:5d13h7m
+Total train time:8d5h14m
+I1130 14:03:53.341637 137274321021824 utils.py:1231] [36650] l2_params = 327.94768537260916
+I1130 14:03:53.341886 137274321021824 utils.py:1231] [36650] train/loss = 2.588795632123947
+I1130 14:03:53.342024 137274321021824 utils.py:1231] [36650] l2_grads = 1.4760710000991821
+I1130 14:03:53.342113 137274321021824 utils.py:1231] [36650] lr = 0.0008425834570925954
+I1130 14:03:53.342182 137274321021824 utils.py:1231] [36650] uptime = 231222.704543702
+I1130 14:03:53.342245 137274321021824 utils.py:1231] [36650] examples_seen = 37529600.0
+I1130 14:03:53.342312 137274321021824 utils.py:1231] [36650] progress = 0.325479782954273
+I1130 14:03:53.342361 137274321021824 utils.py:1231] [36650] epoch = 29.293292755745348
+I1130 14:03:53.342412 137274321021824 utils.py:1231] [36650] img/sec/core = 164.4041780695785
+I1130 14:03:53.342476 137274321021824 utils.py:1231] [36650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.19421143756055
+I1130 14:03:53.342533 137274321021824 utils.py:1231] [36650] core_hours = 64.19421143756055
+I1130 14:03:53.342598 137274321021824 train.py:125] NOTE: Steps:36650/112603 [32.5%]
+Walltime:2d16h13m (0s eval)
+ETA:5d13h2m
+Total train time:8d5h14m
+I1130 14:09:02.078181 137274321021824 utils.py:1231] [36700] l2_params = 327.8966217001646
+I1130 14:09:02.078376 137274321021824 utils.py:1231] [36700] train/loss = 2.8599202930927277
+I1130 14:09:02.078468 137274321021824 utils.py:1231] [36700] l2_grads = 1.2634061574935913
+I1130 14:09:02.078543 137274321021824 utils.py:1231] [36700] lr = 0.0008420254960205381
+I1130 14:09:02.078592 137274321021824 utils.py:1231] [36700] uptime = 231531.440954799
+I1130 14:09:02.078641 137274321021824 utils.py:1231] [36700] examples_seen = 37580800.0
+I1130 14:09:02.078688 137274321021824 utils.py:1231] [36700] progress = 0.3259238208573484
+I1130 14:09:02.078735 137274321021824 utils.py:1231] [36700] epoch = 29.333256320214304
+I1130 14:09:02.078784 137274321021824 utils.py:1231] [36700] img/sec/core = 165.83725844994294
+I1130 14:09:02.078836 137274321021824 utils.py:1231] [36700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.27997155175416
+I1130 14:09:02.078887 137274321021824 utils.py:1231] [36700] core_hours = 64.27997155175416
+I1130 14:09:02.078947 137274321021824 train.py:125] NOTE: Steps:36700/112603 [32.6%]
+Walltime:2d16h18m (0s eval)
+ETA:5d12h57m
+Total train time:8d5h14m
+I1130 14:14:11.210897 137274321021824 utils.py:1231] [36750] l2_params = 327.84835025600404
+I1130 14:14:11.211162 137274321021824 utils.py:1231] [36750] train/loss = 3.2504441142082214
+I1130 14:14:11.211378 137274321021824 utils.py:1231] [36750] l2_grads = 1.225100040435791
+I1130 14:14:11.211483 137274321021824 utils.py:1231] [36750] lr = 0.000841466733310967
+I1130 14:14:11.211567 137274321021824 utils.py:1231] [36750] uptime = 231840.573924771
+I1130 14:14:11.211649 137274321021824 utils.py:1231] [36750] examples_seen = 37632000.0
+I1130 14:14:11.211730 137274321021824 utils.py:1231] [36750] progress = 0.3263678587604238
+I1130 14:14:11.211805 137274321021824 utils.py:1231] [36750] epoch = 29.373219884683262
+I1130 14:14:11.211890 137274321021824 utils.py:1231] [36750] img/sec/core = 165.62452075118665
+I1130 14:14:11.211999 137274321021824 utils.py:1231] [36750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.36584182119083
+I1130 14:14:11.212075 137274321021824 utils.py:1231] [36750] core_hours = 64.36584182119083
+I1130 14:14:11.212165 137274321021824 train.py:125] NOTE: Steps:36750/112603 [32.6%]
+Walltime:2d16h24m (0s eval)
+ETA:5d12h51m
+Total train time:8d5h13m
+I1130 14:19:23.001374 137274321021824 utils.py:1231] [36800] l2_params = 327.84505867371894
+I1130 14:19:23.001645 137274321021824 utils.py:1231] [36800] train/loss = 2.672068625688553
+I1130 14:19:23.001758 137274321021824 utils.py:1231] [36800] l2_grads = 1.4844287633895874
+I1130 14:19:23.001821 137274321021824 utils.py:1231] [36800] lr = 0.0008409071702735081
+I1130 14:19:23.001872 137274321021824 utils.py:1231] [36800] uptime = 232152.36423421296
+I1130 14:19:23.001942 137274321021824 utils.py:1231] [36800] examples_seen = 37683200.0
+I1130 14:19:23.001990 137274321021824 utils.py:1231] [36800] progress = 0.3268118966634992
+I1130 14:19:23.002037 137274321021824 utils.py:1231] [36800] epoch = 29.413183449152218
+I1130 14:19:23.002085 137274321021824 utils.py:1231] [36800] img/sec/core = 164.21292916907848
+I1130 14:19:23.002141 137274321021824 utils.py:1231] [36800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.45245024048026
+I1130 14:19:23.002191 137274321021824 utils.py:1231] [36800] core_hours = 64.45245024048026
+I1130 14:19:23.002267 137274321021824 train.py:125] NOTE: Steps:36800/112603 [32.7%]
+Walltime:2d16h29m (0s eval)
+ETA:5d12h46m
+Total train time:8d5h13m
+I1130 14:24:33.525178 137274321021824 utils.py:1231] [36850] l2_params = 327.8004671208464
+I1130 14:24:33.525566 137274321021824 utils.py:1231] [36850] train/loss = 3.615701347589493
+I1130 14:24:33.525741 137274321021824 utils.py:1231] [36850] l2_grads = 1.186658501625061
+I1130 14:24:33.525840 137274321021824 utils.py:1231] [36850] lr = 0.0008403468082196611
+I1130 14:24:33.525934 137274321021824 utils.py:1231] [36850] uptime = 232462.88829169597
+I1130 14:24:33.526021 137274321021824 utils.py:1231] [36850] examples_seen = 37734400.0
+I1130 14:24:33.526077 137274321021824 utils.py:1231] [36850] progress = 0.3272559345665746
+I1130 14:24:33.526140 137274321021824 utils.py:1231] [36850] epoch = 29.453147013621177
+I1130 14:24:33.526200 137274321021824 utils.py:1231] [36850] img/sec/core = 164.88255504261994
+I1130 14:24:33.526267 137274321021824 utils.py:1231] [36850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.53870692311443
+I1130 14:24:33.526326 137274321021824 utils.py:1231] [36850] core_hours = 64.53870692311443
+I1130 14:24:33.526391 137274321021824 train.py:125] NOTE: Steps:36850/112603 [32.7%]
+Walltime:2d16h34m (0s eval)
+ETA:5d12h40m
+Total train time:8d5h13m
+I1130 14:29:57.001828 137274321021824 utils.py:1231] [36900] l2_params = 327.76191164249934
+I1130 14:29:57.002035 137274321021824 utils.py:1231] [36900] train/loss = 3.0214388966560364
+I1130 14:29:57.002132 137274321021824 utils.py:1231] [36900] l2_grads = 1.248082160949707
+I1130 14:29:57.002217 137274321021824 utils.py:1231] [36900] lr = 0.0008397856484628017
+I1130 14:29:57.002323 137274321021824 utils.py:1231] [36900] uptime = 232786.36467254796
+I1130 14:29:57.002405 137274321021824 utils.py:1231] [36900] examples_seen = 37785600.0
+I1130 14:29:57.002471 137274321021824 utils.py:1231] [36900] progress = 0.32769997246965
+I1130 14:29:57.002535 137274321021824 utils.py:1231] [36900] epoch = 29.493110578090132
+I1130 14:29:57.002612 137274321021824 utils.py:1231] [36900] img/sec/core = 158.28048980004706
+I1130 14:29:57.002711 137274321021824 utils.py:1231] [36900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.6285614733511
+I1130 14:29:57.002799 137274321021824 utils.py:1231] [36900] core_hours = 64.6285614733511
+I1130 14:29:57.002923 137274321021824 train.py:125] NOTE: Steps:36900/112603 [32.8%]
+Walltime:2d16h39m (0s eval)
+ETA:5d12h35m
+Total train time:8d5h13m
+I1130 14:35:08.959574 137274321021824 utils.py:1231] [36950] l2_params = 327.7612113880969
+I1130 14:35:08.959818 137274321021824 utils.py:1231] [36950] train/loss = 2.810541421175003
+I1130 14:35:08.959941 137274321021824 utils.py:1231] [36950] l2_grads = 1.3063243627548218
+I1130 14:35:08.960036 137274321021824 utils.py:1231] [36950] lr = 0.000839223692318171
+I1130 14:35:08.960120 137274321021824 utils.py:1231] [36950] uptime = 233098.322474361
+I1130 14:35:08.960222 137274321021824 utils.py:1231] [36950] examples_seen = 37836800.0
+I1130 14:35:08.960286 137274321021824 utils.py:1231] [36950] progress = 0.3281440103727254
+I1130 14:35:08.960358 137274321021824 utils.py:1231] [36950] epoch = 29.533074142559087
+I1130 14:35:08.960446 137274321021824 utils.py:1231] [36950] img/sec/core = 164.12476207499026
+I1130 14:35:08.960552 137274321021824 utils.py:1231] [36950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.71521641829916
+I1130 14:35:08.960636 137274321021824 utils.py:1231] [36950] core_hours = 64.71521641829916
+I1130 14:35:08.960709 137274321021824 train.py:125] NOTE: Steps:36950/112603 [32.8%]
+Walltime:2d16h44m (0s eval)
+ETA:5d12h30m
+Total train time:8d5h13m
+I1130 14:40:34.813552 137274321021824 utils.py:1231] [37000] l2_params = 327.7167874478726
+I1130 14:40:34.814707 137274321021824 utils.py:1231] [37000] train/loss = 5.120352208614349
+I1130 14:40:34.814867 137274321021824 utils.py:1231] [37000] l2_grads = 1.2441948652267456
+I1130 14:40:34.814955 137274321021824 utils.py:1231] [37000] lr = 0.0008386609411028795
+I1130 14:40:34.815018 137274321021824 utils.py:1231] [37000] uptime = 233424.17738047696
+I1130 14:40:34.815093 137274321021824 utils.py:1231] [37000] examples_seen = 37888000.0
+I1130 14:40:34.815146 137274321021824 utils.py:1231] [37000] progress = 0.32858804827580085
+I1130 14:40:34.815204 137274321021824 utils.py:1231] [37000] epoch = 29.573037707028046
+I1130 14:40:34.815257 137274321021824 utils.py:1231] [37000] img/sec/core = 157.12514692590784
+I1130 14:40:35.055513 137274321021824 utils.py:1231] [37000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.80573166999804
+I1130 14:40:35.055721 137274321021824 utils.py:1231] [37000] core_hours = 64.80573166999804
+I1130 14:40:35.055810 137274321021824 train.py:125] NOTE: Steps:37000/112603 [32.9%]
+Walltime:2d16h50m (0s eval)
+ETA:5d12h25m
+Total train time:8d5h14m
+I1130 14:46:03.364309 137274321021824 utils.py:1231] [37050] l2_params = 327.6509887621271
+I1130 14:46:03.364614 137274321021824 utils.py:1231] [37050] train/loss = 2.381784200668335
+I1130 14:46:03.364778 137274321021824 utils.py:1231] [37050] l2_grads = 1.381256103515625
+I1130 14:46:03.364854 137274321021824 utils.py:1231] [37050] lr = 0.0008380973961359007
+I1130 14:46:03.364921 137274321021824 utils.py:1231] [37050] uptime = 233752.72728120798
+I1130 14:46:03.364977 137274321021824 utils.py:1231] [37050] examples_seen = 37939200.0
+I1130 14:46:03.365029 137274321021824 utils.py:1231] [37050] progress = 0.32903208617887625
+I1130 14:46:03.365080 137274321021824 utils.py:1231] [37050] epoch = 29.613001271497
+I1130 14:46:03.365135 137274321021824 utils.py:1231] [37050] img/sec/core = 155.8362972750305
+I1130 14:46:03.365193 137274321021824 utils.py:1231] [37050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.89699553131221
+I1130 14:46:03.365249 137274321021824 utils.py:1231] [37050] core_hours = 64.89699553131221
+I1130 14:46:03.365314 137274321021824 train.py:125] NOTE: Steps:37050/112603 [32.9%]
+Walltime:2d16h55m (0s eval)
+ETA:5d12h20m
+Total train time:8d5h14m
+I1130 14:51:40.268899 137274321021824 utils.py:1231] [37100] l2_params = 327.5810535466492
+I1130 14:51:40.269214 137274321021824 utils.py:1231] [37100] train/loss = 2.501563608646393
+I1130 14:51:40.269371 137274321021824 utils.py:1231] [37100] l2_grads = 1.3407092094421387
+I1130 14:51:40.269445 137274321021824 utils.py:1231] [37100] lr = 0.0008375330587380681
+I1130 14:51:40.269507 137274321021824 utils.py:1231] [37100] uptime = 234089.631868165
+I1130 14:51:40.269573 137274321021824 utils.py:1231] [37100] examples_seen = 37990400.0
+I1130 14:51:40.269638 137274321021824 utils.py:1231] [37100] progress = 0.32947612408195165
+I1130 14:51:40.269697 137274321021824 utils.py:1231] [37100] epoch = 29.65296483596596
+I1130 14:51:40.269757 137274321021824 utils.py:1231] [37100] img/sec/core = 151.97181036461558
+I1130 14:51:40.269823 137274321021824 utils.py:1231] [37100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 64.99058013880027
+I1130 14:51:40.269890 137274321021824 utils.py:1231] [37100] core_hours = 64.99058013880027
+I1130 14:51:40.269965 137274321021824 train.py:125] NOTE: Steps:37100/112603 [32.9%]
+Walltime:2d17h1m (0s eval)
+ETA:5d12h16m
+Total train time:8d5h15m
+I1130 14:56:55.395696 137274321021824 utils.py:1231] [37150] l2_params = 327.5353446986021
+I1130 14:56:55.395984 137274321021824 utils.py:1231] [37150] train/loss = 2.553354859352112
+I1130 14:56:55.396101 137274321021824 utils.py:1231] [37150] l2_grads = 1.3469032049179077
+I1130 14:56:55.396188 137274321021824 utils.py:1231] [37150] lr = 0.0008369679302320718
+I1130 14:56:55.396241 137274321021824 utils.py:1231] [37150] uptime = 234404.75860392
+I1130 14:56:55.396308 137274321021824 utils.py:1231] [37150] examples_seen = 38041600.0
+I1130 14:56:55.396376 137274321021824 utils.py:1231] [37150] progress = 0.32992016198502705
+I1130 14:56:55.396423 137274321021824 utils.py:1231] [37150] epoch = 29.692928400434916
+I1130 14:56:55.396475 137274321021824 utils.py:1231] [37150] img/sec/core = 162.4743133181967
+I1130 14:56:55.396529 137274321021824 utils.py:1231] [37150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.07811534317666
+I1130 14:56:55.396577 137274321021824 utils.py:1231] [37150] core_hours = 65.07811534317666
+I1130 14:56:55.396658 137274321021824 train.py:125] NOTE: Steps:37150/112603 [33.0%]
+Walltime:2d17h6m (0s eval)
+ETA:5d12h10m
+Total train time:8d5h15m
+I1130 15:02:07.264485 137274321021824 utils.py:1231] [37200] l2_params = 327.4921071855019
+I1130 15:02:07.264716 137274321021824 utils.py:1231] [37200] train/loss = 2.7080104649066925
+I1130 15:02:07.264827 137274321021824 utils.py:1231] [37200] l2_grads = 1.4595595598220825
+I1130 15:02:07.264912 137274321021824 utils.py:1231] [37200] lr = 0.000836402011942458
+I1130 15:02:07.264975 137274321021824 utils.py:1231] [37200] uptime = 234716.627336049
+I1130 15:02:07.265037 137274321021824 utils.py:1231] [37200] examples_seen = 38092800.0
+I1130 15:02:07.265095 137274321021824 utils.py:1231] [37200] progress = 0.33036419988810245
+I1130 15:02:07.265156 137274321021824 utils.py:1231] [37200] epoch = 29.732891964903875
+I1130 15:02:07.265215 137274321021824 utils.py:1231] [37200] img/sec/core = 164.17163609342975
+I1130 15:02:07.265279 137274321021824 utils.py:1231] [37200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.16474554654583
+I1130 15:02:07.265336 137274321021824 utils.py:1231] [37200] core_hours = 65.16474554654583
+I1130 15:02:07.265421 137274321021824 train.py:125] NOTE: Steps:37200/112603 [33.0%]
+Walltime:2d17h11m (0s eval)
+ETA:5d12h5m
+Total train time:8d5h15m
+I1130 15:07:19.099010 137274321021824 utils.py:1231] [37250] l2_params = 327.4512922492626
+I1130 15:07:19.099329 137274321021824 utils.py:1231] [37250] train/loss = 2.6512807607650757
+I1130 15:07:19.099524 137274321021824 utils.py:1231] [37250] l2_grads = 1.4124057292938232
+I1130 15:07:19.099602 137274321021824 utils.py:1231] [37250] lr = 0.0008358353051956227
+I1130 15:07:19.099680 137274321021824 utils.py:1231] [37250] uptime = 235028.46203732898
+I1130 15:07:19.099745 137274321021824 utils.py:1231] [37250] examples_seen = 38144000.0
+I1130 15:07:19.099802 137274321021824 utils.py:1231] [37250] progress = 0.33080823779117785
+I1130 15:07:19.099858 137274321021824 utils.py:1231] [37250] epoch = 29.77285552937283
+I1130 15:07:19.099922 137274321021824 utils.py:1231] [37250] img/sec/core = 164.18955231678353
+I1130 15:07:19.099986 137274321021824 utils.py:1231] [37250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.25136629690138
+I1130 15:07:19.100039 137274321021824 utils.py:1231] [37250] core_hours = 65.25136629690138
+I1130 15:07:19.100105 137274321021824 train.py:125] NOTE: Steps:37250/112603 [33.1%]
+Walltime:2d17h17m (0s eval)
+ETA:5d12h0m
+Total train time:8d5h15m
+I1130 15:12:30.900612 137274321021824 utils.py:1231] [37300] l2_params = 327.41118139518386
+I1130 15:12:30.900869 137274321021824 utils.py:1231] [37300] train/loss = 2.7083842158317566
+I1130 15:12:30.900993 137274321021824 utils.py:1231] [37300] l2_grads = 1.3761348724365234
+I1130 15:12:30.901068 137274321021824 utils.py:1231] [37300] lr = 0.0008352678113198105
+I1130 15:12:30.901134 137274321021824 utils.py:1231] [37300] uptime = 235340.26349405
+I1130 15:12:30.901211 137274321021824 utils.py:1231] [37300] examples_seen = 38195200.0
+I1130 15:12:30.901343 137274321021824 utils.py:1231] [37300] progress = 0.33125227569425325
+I1130 15:12:30.901442 137274321021824 utils.py:1231] [37300] epoch = 29.812819093841785
+I1130 15:12:30.901529 137274321021824 utils.py:1231] [37300] img/sec/core = 164.20705835831998
+I1130 15:12:30.901613 137274321021824 utils.py:1231] [37300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.33797781265723
+I1130 15:12:30.901696 137274321021824 utils.py:1231] [37300] core_hours = 65.33797781265723
+I1130 15:12:30.901784 137274321021824 train.py:125] NOTE: Steps:37300/112603 [33.1%]
+Walltime:2d17h22m (0s eval)
+ETA:5d11h54m
+Total train time:8d5h15m
+I1130 15:17:42.654920 137274321021824 utils.py:1231] [37350] l2_params = 327.3823022124512
+I1130 15:17:42.655124 137274321021824 utils.py:1231] [37350] train/loss = 4.600254774093628
+I1130 15:17:42.655236 137274321021824 utils.py:1231] [37350] l2_grads = 1.066323161125183
+I1130 15:17:42.655292 137274321021824 utils.py:1231] [37350] lr = 0.0008346995316451096
+I1130 15:17:42.655359 137274321021824 utils.py:1231] [37350] uptime = 235652.01772084698
+I1130 15:17:42.655410 137274321021824 utils.py:1231] [37350] examples_seen = 38246400.0
+I1130 15:17:42.655458 137274321021824 utils.py:1231] [37350] progress = 0.33169631359732865
+I1130 15:17:42.655506 137274321021824 utils.py:1231] [37350] epoch = 29.852782658310744
+I1130 15:17:42.655554 137274321021824 utils.py:1231] [37350] img/sec/core = 164.23193528453464
+I1130 15:17:42.655609 137274321021824 utils.py:1231] [37350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.4245762089897
+I1130 15:17:42.655656 137274321021824 utils.py:1231] [37350] core_hours = 65.4245762089897
+I1130 15:17:42.655714 137274321021824 train.py:125] NOTE: Steps:37350/112603 [33.2%]
+Walltime:2d17h27m (0s eval)
+ETA:5d11h49m
+Total train time:8d5h15m
+I1130 15:22:54.507544 137274321021824 utils.py:1231] [37400] l2_params = 327.3164189011936
+I1130 15:22:54.507814 137274321021824 utils.py:1231] [37400] train/loss = 2.6585673093795776
+I1130 15:22:54.507941 137274321021824 utils.py:1231] [37400] l2_grads = 1.344779372215271
+I1130 15:22:54.508016 137274321021824 utils.py:1231] [37400] lr = 0.0008341304675034508
+I1130 15:22:54.508077 137274321021824 utils.py:1231] [37400] uptime = 235963.870439469
+I1130 15:22:54.508141 137274321021824 utils.py:1231] [37400] examples_seen = 38297600.0
+I1130 15:22:54.508192 137274321021824 utils.py:1231] [37400] progress = 0.33214035150040405
+I1130 15:22:54.508251 137274321021824 utils.py:1231] [37400] epoch = 29.8927462227797
+I1130 15:22:54.508304 137274321021824 utils.py:1231] [37400] img/sec/core = 164.18006623843928
+I1130 15:22:54.508363 137274321021824 utils.py:1231] [37400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.5112019641625
+I1130 15:22:54.508419 137274321021824 utils.py:1231] [37400] core_hours = 65.5112019641625
+I1130 15:22:54.508480 137274321021824 train.py:125] NOTE: Steps:37400/112603 [33.2%]
+Walltime:2d17h32m (0s eval)
+ETA:5d11h44m
+Total train time:8d5h15m
+I1130 15:28:06.270264 137274321021824 utils.py:1231] [37450] l2_params = 327.2639300004926
+I1130 15:28:06.270532 137274321021824 utils.py:1231] [37450] train/loss = 3.5031222701072693
+I1130 15:28:06.270649 137274321021824 utils.py:1231] [37450] l2_grads = 1.1259174346923828
+I1130 15:28:06.270735 137274321021824 utils.py:1231] [37450] lr = 0.0008335606202286049
+I1130 15:28:06.270796 137274321021824 utils.py:1231] [37450] uptime = 236275.633157266
+I1130 15:28:06.270857 137274321021824 utils.py:1231] [37450] examples_seen = 38348800.0
+I1130 15:28:06.270922 137274321021824 utils.py:1231] [37450] progress = 0.3325843894034795
+I1130 15:28:06.270980 137274321021824 utils.py:1231] [37450] epoch = 29.93270978724866
+I1130 15:28:06.271038 137274321021824 utils.py:1231] [37450] img/sec/core = 164.22746235274982
+I1130 15:28:06.271095 137274321021824 utils.py:1231] [37450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.59780271910611
+I1130 15:28:06.271147 137274321021824 utils.py:1231] [37450] core_hours = 65.59780271910611
+I1130 15:28:06.271217 137274321021824 train.py:125] NOTE: Steps:37450/112603 [33.3%]
+Walltime:2d17h37m (0s eval)
+ETA:5d11h38m
+Total train time:8d5h14m
+I1130 15:33:18.036178 137274321021824 utils.py:1231] [37500] l2_params = 327.21191067087875
+I1130 15:33:18.036424 137274321021824 utils.py:1231] [37500] train/loss = 3.7601886689662933
+I1130 15:33:18.036524 137274321021824 utils.py:1231] [37500] l2_grads = 1.1668223142623901
+I1130 15:33:18.036608 137274321021824 utils.py:1231] [37500] lr = 0.0008329899911561761
+I1130 15:33:18.036683 137274321021824 utils.py:1231] [37500] uptime = 236587.399042513
+I1130 15:33:18.036742 137274321021824 utils.py:1231] [37500] examples_seen = 38400000.0
+I1130 15:33:18.036800 137274321021824 utils.py:1231] [37500] progress = 0.3330284273065549
+I1130 15:33:18.036854 137274321021824 utils.py:1231] [37500] epoch = 29.972673351717614
+I1130 15:33:18.036920 137274321021824 utils.py:1231] [37500] img/sec/core = 164.2257938498954
+I1130 15:33:18.036983 137274321021824 utils.py:1231] [37500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.68440435389694
+I1130 15:33:18.037038 137274321021824 utils.py:1231] [37500] core_hours = 65.68440435389694
+I1130 15:33:18.037122 137274321021824 train.py:125] NOTE: Steps:37500/112603 [33.3%]
+Walltime:2d17h43m (0s eval)
+ETA:5d11h33m
+Total train time:8d5h14m
+I1130 15:33:18.037226 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:37500/112603 [33.3%]
+Walltime:2d17h43m (0s eval)
+ETA:5d11h33m
+Total train time:8d5h14m
+I1130 15:34:55.197600 137274321021824 utils.py:1231] [37500] val/acc@1 = 0.6103515625
+I1130 15:34:55.197844 137274321021824 utils.py:1231] [37500] val/loss = 1.652819820508665
+I1130 15:34:55.198021 137274321021824 utils.py:1231] [37500] z/secs/eval/val = 97.16072078899015
+I1130 15:34:55.198084 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.16072078899015
+I1130 15:40:06.966504 137274321021824 utils.py:1231] [37550] l2_params = 327.1487197736291
+I1130 15:40:06.966733 137274321021824 utils.py:1231] [37550] train/loss = 3.3450690507888794
+I1130 15:40:06.966844 137274321021824 utils.py:1231] [37550] l2_grads = 1.2190089225769043
+I1130 15:40:06.966924 137274321021824 utils.py:1231] [37550] lr = 0.0008324185816236015
+I1130 15:40:06.966986 137274321021824 utils.py:1231] [37550] uptime = 236996.32934731396
+I1130 15:40:06.967047 137274321021824 utils.py:1231] [37550] examples_seen = 38451200.0
+I1130 15:40:06.967106 137274321021824 utils.py:1231] [37550] progress = 0.3334724652096303
+I1130 15:40:06.967164 137274321021824 utils.py:1231] [37550] epoch = 30.01263691618657
+I1130 15:40:06.967225 137274321021824 utils.py:1231] [37550] img/sec/core = 125.2047094551206
+I1130 15:40:06.967287 137274321021824 utils.py:1231] [37550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.79799610523054
+I1130 15:40:06.967349 137274321021824 utils.py:1231] [37550] core_hours = 65.79799610523054
+I1130 15:40:06.967415 137274321021824 train.py:125] NOTE: Steps:37550/112603 [33.3%]
+Walltime:2d17h49m (0s eval)
+ETA:5d11h31m
+Total train time:8d5h19m
+I1130 15:45:18.737482 137274321021824 utils.py:1231] [37600] l2_params = 327.1142139556128
+I1130 15:45:18.737751 137274321021824 utils.py:1231] [37600] train/loss = 2.7636231780052185
+I1130 15:45:18.737887 137274321021824 utils.py:1231] [37600] l2_grads = 1.3650875091552734
+I1130 15:45:18.737962 137274321021824 utils.py:1231] [37600] lr = 0.0008318463929701478
+I1130 15:45:18.738020 137274321021824 utils.py:1231] [37600] uptime = 237308.10038122901
+I1130 15:45:18.738087 137274321021824 utils.py:1231] [37600] examples_seen = 38502400.0
+I1130 15:45:18.738147 137274321021824 utils.py:1231] [37600] progress = 0.3339165031127057
+I1130 15:45:18.738202 137274321021824 utils.py:1231] [37600] epoch = 30.052600480655528
+I1130 15:45:18.738256 137274321021824 utils.py:1231] [37600] img/sec/core = 164.22308178235255
+I1130 15:45:18.738317 137274321021824 utils.py:1231] [37600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.88459917020695
+I1130 15:45:18.738373 137274321021824 utils.py:1231] [37600] core_hours = 65.88459917020695
+I1130 15:45:18.738438 137274321021824 train.py:125] NOTE: Steps:37600/112603 [33.4%]
+Walltime:2d17h55m (0s eval)
+ETA:5d11h25m
+Total train time:8d5h19m
+I1130 15:50:30.518337 137274321021824 utils.py:1231] [37650] l2_params = 327.09568312948306
+I1130 15:50:30.518578 137274321021824 utils.py:1231] [37650] train/loss = 2.7476960122585297
+I1130 15:50:30.518711 137274321021824 utils.py:1231] [37650] l2_grads = 1.3883793354034424
+I1130 15:50:30.518784 137274321021824 utils.py:1231] [37650] lr = 0.0008312734265369077
+I1130 15:50:30.518852 137274321021824 utils.py:1231] [37650] uptime = 237619.88121197198
+I1130 15:50:30.518926 137274321021824 utils.py:1231] [37650] examples_seen = 38553600.0
+I1130 15:50:30.518981 137274321021824 utils.py:1231] [37650] progress = 0.3343605410157811
+I1130 15:50:30.519035 137274321021824 utils.py:1231] [37650] epoch = 30.092564045124483
+I1130 15:50:30.519089 137274321021824 utils.py:1231] [37650] img/sec/core = 164.2179215379973
+I1130 15:50:30.519150 137274321021824 utils.py:1231] [37650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 65.97120495652443
+I1130 15:50:30.519202 137274321021824 utils.py:1231] [37650] core_hours = 65.97120495652443
+I1130 15:50:30.519265 137274321021824 train.py:125] NOTE: Steps:37650/112603 [33.4%]
+Walltime:2d18h0m (0s eval)
+ETA:5d11h20m
+Total train time:8d5h18m
+I1130 15:55:42.294503 137274321021824 utils.py:1231] [37700] l2_params = 327.0773235616103
+I1130 15:55:42.294747 137274321021824 utils.py:1231] [37700] train/loss = 2.5711256563663483
+I1130 15:55:42.294859 137274321021824 utils.py:1231] [37700] l2_grads = 1.5395079851150513
+I1130 15:55:42.294941 137274321021824 utils.py:1231] [37700] lr = 0.000830699683666797
+I1130 15:55:42.295014 137274321021824 utils.py:1231] [37700] uptime = 237931.65737476997
+I1130 15:55:42.295074 137274321021824 utils.py:1231] [37700] examples_seen = 38604800.0
+I1130 15:55:42.295125 137274321021824 utils.py:1231] [37700] progress = 0.3348045789188565
+I1130 15:55:42.295173 137274321021824 utils.py:1231] [37700] epoch = 30.132527609593442
+I1130 15:55:42.295222 137274321021824 utils.py:1231] [37700] img/sec/core = 164.22038022571255
+I1130 15:55:42.295278 137274321021824 utils.py:1231] [37700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.05780944619055
+I1130 15:55:42.295327 137274321021824 utils.py:1231] [37700] core_hours = 66.05780944619055
+I1130 15:55:42.295386 137274321021824 train.py:125] NOTE: Steps:37700/112603 [33.5%]
+Walltime:2d18h5m (0s eval)
+ETA:5d11h15m
+Total train time:8d5h18m
+I1130 16:00:54.069691 137274321021824 utils.py:1231] [37750] l2_params = 326.99971094240817
+I1130 16:00:54.069914 137274321021824 utils.py:1231] [37750] train/loss = 2.566720724105835
+I1130 16:00:54.070011 137274321021824 utils.py:1231] [37750] l2_grads = 1.3582402467727661
+I1130 16:00:54.070072 137274321021824 utils.py:1231] [37750] lr = 0.0008301251657045516
+I1130 16:00:54.070123 137274321021824 utils.py:1231] [37750] uptime = 238243.432485339
+I1130 16:00:54.070179 137274321021824 utils.py:1231] [37750] examples_seen = 38656000.0
+I1130 16:00:54.070227 137274321021824 utils.py:1231] [37750] progress = 0.3352486168219319
+I1130 16:00:54.070275 137274321021824 utils.py:1231] [37750] epoch = 30.172491174062397
+I1130 16:00:54.070325 137274321021824 utils.py:1231] [37750] img/sec/core = 164.22093446315853
+I1130 16:00:54.070380 137274321021824 utils.py:1231] [37750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.14441364357083
+I1130 16:00:54.070432 137274321021824 utils.py:1231] [37750] core_hours = 66.14441364357083
+I1130 16:00:54.070490 137274321021824 train.py:125] NOTE: Steps:37750/112603 [33.5%]
+Walltime:2d18h10m (0s eval)
+ETA:5d11h9m
+Total train time:8d5h18m
+I1130 16:06:05.850581 137274321021824 utils.py:1231] [37800] l2_params = 326.9495942141851
+I1130 16:06:05.850785 137274321021824 utils.py:1231] [37800] train/loss = 2.7348397970199585
+I1130 16:06:05.850901 137274321021824 utils.py:1231] [37800] l2_grads = 1.4042688608169556
+I1130 16:06:05.851042 137274321021824 utils.py:1231] [37800] lr = 0.0008295498739967222
+I1130 16:06:05.851105 137274321021824 utils.py:1231] [37800] uptime = 238555.21346661
+I1130 16:06:05.851189 137274321021824 utils.py:1231] [37800] examples_seen = 38707200.0
+I1130 16:06:05.851248 137274321021824 utils.py:1231] [37800] progress = 0.3356926547250073
+I1130 16:06:05.851300 137274321021824 utils.py:1231] [37800] epoch = 30.212454738531356
+I1130 16:06:05.851357 137274321021824 utils.py:1231] [37800] img/sec/core = 164.21784225348992
+I1130 16:06:05.851423 137274321021824 utils.py:1231] [37800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.23101947170167
+I1130 16:06:05.851482 137274321021824 utils.py:1231] [37800] core_hours = 66.23101947170167
+I1130 16:06:05.851542 137274321021824 train.py:125] NOTE: Steps:37800/112603 [33.6%]
+Walltime:2d18h15m (0s eval)
+ETA:5d11h4m
+Total train time:8d5h18m
+I1130 16:11:17.623067 137274321021824 utils.py:1231] [37850] l2_params = 326.9089604489556
+I1130 16:11:17.623276 137274321021824 utils.py:1231] [37850] train/loss = 3.3766738176345825
+I1130 16:11:17.623378 137274321021824 utils.py:1231] [37850] l2_grads = 1.2029002904891968
+I1130 16:11:17.623450 137274321021824 utils.py:1231] [37850] lr = 0.000828973809891676
+I1130 16:11:17.623511 137274321021824 utils.py:1231] [37850] uptime = 238866.98587214598
+I1130 16:11:17.623571 137274321021824 utils.py:1231] [37850] examples_seen = 38758400.0
+I1130 16:11:17.623630 137274321021824 utils.py:1231] [37850] progress = 0.3361366926280827
+I1130 16:11:17.623686 137274321021824 utils.py:1231] [37850] epoch = 30.25241830300031
+I1130 16:11:17.623742 137274321021824 utils.py:1231] [37850] img/sec/core = 164.22235929437494
+I1130 16:11:17.623804 137274321021824 utils.py:1231] [37850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.31762291768388
+I1130 16:11:17.623856 137274321021824 utils.py:1231] [37850] core_hours = 66.31762291768388
+I1130 16:11:17.623931 137274321021824 train.py:125] NOTE: Steps:37850/112603 [33.6%]
+Walltime:2d18h21m (0s eval)
+ETA:5d10h58m
+Total train time:8d5h18m
+I1130 16:16:27.512692 137274321021824 utils.py:1231] [37900] l2_params = 326.87344646905245
+I1130 16:16:27.512908 137274321021824 utils.py:1231] [37900] train/loss = 4.417548358440399
+I1130 16:16:27.513011 137274321021824 utils.py:1231] [37900] l2_grads = 1.2780194282531738
+I1130 16:16:27.513079 137274321021824 utils.py:1231] [37900] lr = 0.0008283969747395876
+I1130 16:16:27.513154 137274321021824 utils.py:1231] [37900] uptime = 239176.87551648502
+I1130 16:16:27.513227 137274321021824 utils.py:1231] [37900] examples_seen = 38809600.0
+I1130 16:16:27.513300 137274321021824 utils.py:1231] [37900] progress = 0.3365807305311581
+I1130 16:16:27.513359 137274321021824 utils.py:1231] [37900] epoch = 30.292381867469267
+I1130 16:16:27.513414 137274321021824 utils.py:1231] [37900] img/sec/core = 165.22010636787758
+I1130 16:16:27.513476 137274321021824 utils.py:1231] [37900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.40370337444473
+I1130 16:16:27.513526 137274321021824 utils.py:1231] [37900] core_hours = 66.40370337444473
+I1130 16:16:27.513590 137274321021824 train.py:125] NOTE: Steps:37900/112603 [33.7%]
+Walltime:2d18h26m (0s eval)
+ETA:5d10h53m
+Total train time:8d5h17m
+I1130 16:21:39.300877 137274321021824 utils.py:1231] [37950] l2_params = 326.8551092894771
+I1130 16:21:39.301095 137274321021824 utils.py:1231] [37950] train/loss = 2.847097873687744
+I1130 16:21:39.301203 137274321021824 utils.py:1231] [37950] l2_grads = 1.3948701620101929
+I1130 16:21:39.301270 137274321021824 utils.py:1231] [37950] lr = 0.0008278193698924404
+I1130 16:21:39.301345 137274321021824 utils.py:1231] [37950] uptime = 239488.66370672
+I1130 16:21:39.301403 137274321021824 utils.py:1231] [37950] examples_seen = 38860800.0
+I1130 16:21:39.301457 137274321021824 utils.py:1231] [37950] progress = 0.33702476843423357
+I1130 16:21:39.301521 137274321021824 utils.py:1231] [37950] epoch = 30.332345431938226
+I1130 16:21:39.301574 137274321021824 utils.py:1231] [37950] img/sec/core = 164.21404531523012
+I1130 16:21:39.301633 137274321021824 utils.py:1231] [37950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.49031120506555
+I1130 16:21:39.301702 137274321021824 utils.py:1231] [37950] core_hours = 66.49031120506555
+I1130 16:21:39.301770 137274321021824 train.py:125] NOTE: Steps:37950/112603 [33.7%]
+Walltime:2d18h31m (0s eval)
+ETA:5d10h48m
+Total train time:8d5h17m
+I1130 16:26:49.064445 137274321021824 utils.py:1231] [38000] l2_params = 326.8081957274102
+I1130 16:26:49.065108 137274321021824 utils.py:1231] [38000] train/loss = 3.2606326043605804
+I1130 16:26:49.065504 137274321021824 utils.py:1231] [38000] l2_grads = 1.2164674997329712
+I1130 16:26:49.065669 137274321021824 utils.py:1231] [38000] lr = 0.0008272409967040218
+I1130 16:26:49.065790 137274321021824 utils.py:1231] [38000] uptime = 239798.42814170697
+I1130 16:26:49.065909 137274321021824 utils.py:1231] [38000] examples_seen = 38912000.0
+I1130 16:26:49.066011 137274321021824 utils.py:1231] [38000] progress = 0.33746880633730897
+I1130 16:26:49.066112 137274321021824 utils.py:1231] [38000] epoch = 30.37230899640718
+I1130 16:26:49.066212 137274321021824 utils.py:1231] [38000] img/sec/core = 165.28688970427459
+I1130 16:26:49.066316 137274321021824 utils.py:1231] [38000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.57635688145082
+I1130 16:26:49.066412 137274321021824 utils.py:1231] [38000] core_hours = 66.57635688145082
+I1130 16:26:49.066516 137274321021824 train.py:125] NOTE: Steps:38000/112603 [33.7%]
+Walltime:2d18h36m (0s eval)
+ETA:5d10h42m
+Total train time:8d5h17m
+I1130 16:31:55.714713 137274321021824 utils.py:1231] [38050] l2_params = 326.7788146318476
+I1130 16:31:55.715926 137274321021824 utils.py:1231] [38050] train/loss = 4.040828615427017
+I1130 16:31:55.716104 137274321021824 utils.py:1231] [38050] l2_grads = 1.2156054973602295
+I1130 16:31:55.716159 137274321021824 utils.py:1231] [38050] lr = 0.00082666185652992
+I1130 16:31:55.716208 137274321021824 utils.py:1231] [38050] uptime = 240105.07857128402
+I1130 16:31:55.716255 137274321021824 utils.py:1231] [38050] examples_seen = 38963200.0
+I1130 16:31:55.716299 137274321021824 utils.py:1231] [38050] progress = 0.33791284424038437
+I1130 16:31:55.716341 137274321021824 utils.py:1231] [38050] epoch = 30.41227256087614
+I1130 16:31:55.716386 137274321021824 utils.py:1231] [38050] img/sec/core = 166.96536205939512
+I1130 16:31:55.716437 137274321021824 utils.py:1231] [38050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.66153755633334
+I1130 16:31:55.716480 137274321021824 utils.py:1231] [38050] core_hours = 66.66153755633334
+I1130 16:31:55.716539 137274321021824 train.py:125] NOTE: Steps:38050/112603 [33.8%]
+Walltime:2d18h41m (0s eval)
+ETA:5d10h37m
+Total train time:8d5h17m
+I1130 16:36:05.367983 137274321021824 utils.py:1231] [38100] l2_params = 326.73609371015107
+I1130 16:36:05.368281 137274321021824 utils.py:1231] [38100] train/loss = 2.695635288953781
+I1130 16:36:05.368448 137274321021824 utils.py:1231] [38100] l2_grads = 1.4687094688415527
+I1130 16:36:05.368525 137274321021824 utils.py:1231] [38100] lr = 0.0008260819507275206
+I1130 16:36:05.368615 137274321021824 utils.py:1231] [38100] uptime = 240354.73097147897
+I1130 16:36:05.368703 137274321021824 utils.py:1231] [38100] examples_seen = 39014400.0
+I1130 16:36:05.368767 137274321021824 utils.py:1231] [38100] progress = 0.33835688214345977
+I1130 16:36:05.368827 137274321021824 utils.py:1231] [38100] epoch = 30.452236125345095
+I1130 16:36:05.368898 137274321021824 utils.py:1231] [38100] img/sec/core = 205.08515023295166
+I1130 16:36:05.368957 137274321021824 utils.py:1231] [38100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.73088544527637
+I1130 16:36:05.369010 137274321021824 utils.py:1231] [38100] core_hours = 66.73088544527637
+I1130 16:36:05.369076 137274321021824 train.py:125] NOTE: Steps:38100/112603 [33.8%]
+Walltime:2d18h45m (0s eval)
+ETA:5d10h29m
+Total train time:8d5h13m
+I1130 16:40:15.426249 137274321021824 utils.py:1231] [38150] l2_params = 326.6709626858765
+I1130 16:40:15.426537 137274321021824 utils.py:1231] [38150] train/loss = 4.806692957878113
+I1130 16:40:15.426696 137274321021824 utils.py:1231] [38150] l2_grads = 1.323094129562378
+I1130 16:40:15.426778 137274321021824 utils.py:1231] [38150] lr = 0.0008255012806560031
+I1130 16:40:15.426839 137274321021824 utils.py:1231] [38150] uptime = 240604.789200928
+I1130 16:40:15.426901 137274321021824 utils.py:1231] [38150] examples_seen = 39065600.0
+I1130 16:40:15.426956 137274321021824 utils.py:1231] [38150] progress = 0.33880092004653517
+I1130 16:40:15.427007 137274321021824 utils.py:1231] [38150] epoch = 30.49219968981405
+I1130 16:40:15.427062 137274321021824 utils.py:1231] [38150] img/sec/core = 204.7523095433103
+I1130 16:40:15.427122 137274321021824 utils.py:1231] [38150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.80034606456778
+I1130 16:40:15.427174 137274321021824 utils.py:1231] [38150] core_hours = 66.80034606456778
+I1130 16:40:15.427239 137274321021824 train.py:125] NOTE: Steps:38150/112603 [33.9%]
+Walltime:2d18h50m (0s eval)
+ETA:5d10h22m
+Total train time:8d5h10m
+I1130 16:44:29.106518 137274321021824 utils.py:1231] [38200] l2_params = 326.6565944562523
+I1130 16:44:29.107365 137274321021824 utils.py:1231] [38200] train/loss = 2.601173847913742
+I1130 16:44:29.107492 137274321021824 utils.py:1231] [38200] l2_grads = 1.3786550760269165
+I1130 16:44:29.107553 137274321021824 utils.py:1231] [38200] lr = 0.0008249198476763398
+I1130 16:44:29.107604 137274321021824 utils.py:1231] [38200] uptime = 240858.469967274
+I1130 16:44:29.107687 137274321021824 utils.py:1231] [38200] examples_seen = 39116800.0
+I1130 16:44:29.316418 137274321021824 utils.py:1231] [38200] progress = 0.33924495794961057
+I1130 16:44:29.316666 137274321021824 utils.py:1231] [38200] epoch = 30.53216325428301
+I1130 16:44:29.316809 137274321021824 utils.py:1231] [38200] img/sec/core = 201.82846629440212
+I1130 16:44:29.317129 137274321021824 utils.py:1231] [38200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.87081294410832
+I1130 16:44:29.317231 137274321021824 utils.py:1231] [38200] core_hours = 66.87081294410832
+I1130 16:44:29.317309 137274321021824 train.py:125] NOTE: Steps:38200/112603 [33.9%]
+Walltime:2d18h54m (0s eval)
+ETA:5d10h15m
+Total train time:8d5h7m
+I1130 16:48:44.993437 137274321021824 utils.py:1231] [38250] l2_params = 326.6105301531845
+I1130 16:48:44.993750 137274321021824 utils.py:1231] [38250] train/loss = 2.6780634820461273
+I1130 16:48:44.993911 137274321021824 utils.py:1231] [38250] l2_grads = 1.419950246810913
+I1130 16:48:44.993977 137274321021824 utils.py:1231] [38250] lr = 0.0008243376531512897
+I1130 16:48:44.994028 137274321021824 utils.py:1231] [38250] uptime = 241114.35639052297
+I1130 16:48:44.994081 137274321021824 utils.py:1231] [38250] examples_seen = 39168000.0
+I1130 16:48:44.994132 137274321021824 utils.py:1231] [38250] progress = 0.33968899585268597
+I1130 16:48:44.994180 137274321021824 utils.py:1231] [38250] epoch = 30.572126818751965
+I1130 16:48:44.994231 137274321021824 utils.py:1231] [38250] img/sec/core = 200.0887712209078
+I1130 16:48:44.994288 137274321021824 utils.py:1231] [38250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 66.94189250612193
+I1130 16:48:44.994343 137274321021824 utils.py:1231] [38250] core_hours = 66.94189250612193
+I1130 16:48:44.994403 137274321021824 train.py:125] NOTE: Steps:38250/112603 [34.0%]
+Walltime:2d18h58m (0s eval)
+ETA:5d10h7m
+Total train time:8d5h4m
+I1130 16:53:00.739709 137274321021824 utils.py:1231] [38300] l2_params = 326.5618537342485
+I1130 16:53:00.739972 137274321021824 utils.py:1231] [38300] train/loss = 2.9147843420505524
+I1130 16:53:00.740077 137274321021824 utils.py:1231] [38300] l2_grads = 1.2819173336029053
+I1130 16:53:00.740162 137274321021824 utils.py:1231] [38300] lr = 0.0008237546984453961
+I1130 16:53:00.740222 137274321021824 utils.py:1231] [38300] uptime = 241370.102583943
+I1130 16:53:00.740294 137274321021824 utils.py:1231] [38300] examples_seen = 39219200.0
+I1130 16:53:00.740349 137274321021824 utils.py:1231] [38300] progress = 0.34013303375576137
+I1130 16:53:00.740407 137274321021824 utils.py:1231] [38300] epoch = 30.612090383220924
+I1130 16:53:00.740467 137274321021824 utils.py:1231] [38300] img/sec/core = 200.19848317316507
+I1130 16:53:00.740530 137274321021824 utils.py:1231] [38300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.01293311540527
+I1130 16:53:00.740584 137274321021824 utils.py:1231] [38300] core_hours = 67.01293311540527
+I1130 16:53:00.740648 137274321021824 train.py:125] NOTE: Steps:38300/112603 [34.0%]
+Walltime:2d19h2m (0s eval)
+ETA:5d10h0m
+Total train time:8d5h1m
+I1130 16:57:13.098855 137274321021824 utils.py:1231] [38350] l2_params = 326.54071632884774
+I1130 16:57:13.099110 137274321021824 utils.py:1231] [38350] train/loss = 4.097568511962891
+I1130 16:57:13.099215 137274321021824 utils.py:1231] [38350] l2_grads = 1.1098573207855225
+I1130 16:57:13.099295 137274321021824 utils.py:1231] [38350] lr = 0.000823170984924987
+I1130 16:57:13.099375 137274321021824 utils.py:1231] [38350] uptime = 241622.46172051097
+I1130 16:57:13.099430 137274321021824 utils.py:1231] [38350] examples_seen = 39270400.0
+I1130 16:57:13.099478 137274321021824 utils.py:1231] [38350] progress = 0.34057707165883677
+I1130 16:57:13.099525 137274321021824 utils.py:1231] [38350] epoch = 30.65205394768988
+I1130 16:57:13.099575 137274321021824 utils.py:1231] [38350] img/sec/core = 202.88546195040922
+I1130 16:57:13.099630 137274321021824 utils.py:1231] [38350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.08303287556303
+I1130 16:57:13.099681 137274321021824 utils.py:1231] [38350] core_hours = 67.08303287556303
+I1130 16:57:13.099742 137274321021824 train.py:125] NOTE: Steps:38350/112603 [34.1%]
+Walltime:2d19h7m (0s eval)
+ETA:5d9h53m
+Total train time:8d4h58m
+I1130 17:01:24.209494 137274321021824 utils.py:1231] [38400] l2_params = 326.48122329149135
+I1130 17:01:24.210429 137274321021824 utils.py:1231] [38400] train/loss = 5.005904912948608
+I1130 17:01:24.210568 137274321021824 utils.py:1231] [38400] l2_grads = 1.1850115060806274
+I1130 17:01:24.210624 137274321021824 utils.py:1231] [38400] lr = 0.000822586513958166
+I1130 17:01:24.210679 137274321021824 utils.py:1231] [38400] uptime = 241873.57304188499
+I1130 17:01:24.210723 137274321021824 utils.py:1231] [38400] examples_seen = 39321600.0
+I1130 17:01:24.210763 137274321021824 utils.py:1231] [38400] progress = 0.34102110956191223
+I1130 17:01:24.210805 137274321021824 utils.py:1231] [38400] epoch = 30.692017512158838
+I1130 17:01:24.210847 137274321021824 utils.py:1231] [38400] img/sec/core = 203.8936345834454
+I1130 17:01:24.210901 137274321021824 utils.py:1231] [38400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.15278602038916
+I1130 17:01:24.210951 137274321021824 utils.py:1231] [38400] core_hours = 67.15278602038916
+I1130 17:01:24.211006 137274321021824 train.py:125] NOTE: Steps:38400/112603 [34.1%]
+Walltime:2d19h11m (0s eval)
+ETA:5d9h46m
+Total train time:8d4h55m
+I1130 17:05:45.078717 137274321021824 utils.py:1231] [38450] l2_params = 326.4207557743761
+I1130 17:05:45.079688 137274321021824 utils.py:1231] [38450] train/loss = 4.39696592092514
+I1130 17:05:45.079844 137274321021824 utils.py:1231] [38450] l2_grads = 1.1294493675231934
+I1130 17:05:45.079914 137274321021824 utils.py:1231] [38450] lr = 0.0008220012869148134
+I1130 17:05:45.079994 137274321021824 utils.py:1231] [38450] uptime = 242134.44235633698
+I1130 17:05:45.080049 137274321021824 utils.py:1231] [38450] examples_seen = 39372800.0
+I1130 17:05:45.080133 137274321021824 utils.py:1231] [38450] progress = 0.34146514746498763
+I1130 17:05:45.080203 137274321021824 utils.py:1231] [38450] epoch = 30.731981076627793
+I1130 17:05:45.080275 137274321021824 utils.py:1231] [38450] img/sec/core = 196.2668553315886
+I1130 17:05:45.080340 137274321021824 utils.py:1231] [38450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.22524971884805
+I1130 17:05:45.080394 137274321021824 utils.py:1231] [38450] core_hours = 67.22524971884805
+I1130 17:05:45.080489 137274321021824 train.py:125] NOTE: Steps:38450/112603 [34.1%]
+Walltime:2d19h15m (0s eval)
+ETA:5d9h39m
+Total train time:8d4h52m
+I1130 17:09:55.700743 137274321021824 utils.py:1231] [38500] l2_params = 326.3623442537369
+I1130 17:09:55.701670 137274321021824 utils.py:1231] [38500] train/loss = 2.5179561972618103
+I1130 17:09:55.701813 137274321021824 utils.py:1231] [38500] l2_grads = 1.3047055006027222
+I1130 17:09:55.701872 137274321021824 utils.py:1231] [38500] lr = 0.0008214153051665811
+I1130 17:09:55.701923 137274321021824 utils.py:1231] [38500] uptime = 242385.06428579497
+I1130 17:09:55.701970 137274321021824 utils.py:1231] [38500] examples_seen = 39424000.0
+I1130 17:09:55.702014 137274321021824 utils.py:1231] [38500] progress = 0.34190918536806303
+I1130 17:09:55.702057 137274321021824 utils.py:1231] [38500] epoch = 30.77194464109675
+I1130 17:09:55.702101 137274321021824 utils.py:1231] [38500] img/sec/core = 204.29177969672295
+I1130 17:09:55.702249 137274321021824 utils.py:1231] [38500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.29486692147526
+I1130 17:09:55.702300 137274321021824 utils.py:1231] [38500] core_hours = 67.29486692147526
+I1130 17:09:55.702358 137274321021824 train.py:125] NOTE: Steps:38500/112603 [34.2%]
+Walltime:2d19h19m (0s eval)
+ETA:5d9h31m
+Total train time:8d4h49m
+I1130 17:14:12.480184 137274321021824 utils.py:1231] [38550] l2_params = 326.29773612599035
+I1130 17:14:12.480503 137274321021824 utils.py:1231] [38550] train/loss = 2.6795098781585693
+I1130 17:14:12.480675 137274321021824 utils.py:1231] [38550] l2_grads = 1.4282885789871216
+I1130 17:14:12.480776 137274321021824 utils.py:1231] [38550] lr = 0.0008208285700868891
+I1130 17:14:12.480832 137274321021824 utils.py:1231] [38550] uptime = 242641.843194545
+I1130 17:14:12.480905 137274321021824 utils.py:1231] [38550] examples_seen = 39475200.0
+I1130 17:14:12.480963 137274321021824 utils.py:1231] [38550] progress = 0.34235322327113843
+I1130 17:14:12.481010 137274321021824 utils.py:1231] [38550] epoch = 30.811908205565707
+I1130 17:14:12.481060 137274321021824 utils.py:1231] [38550] img/sec/core = 199.3933234206548
+I1130 17:14:12.481118 137274321021824 utils.py:1231] [38550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.36619439612805
+I1130 17:14:12.481168 137274321021824 utils.py:1231] [38550] core_hours = 67.36619439612805
+I1130 17:14:12.481227 137274321021824 train.py:125] NOTE: Steps:38550/112603 [34.2%]
+Walltime:2d19h24m (0s eval)
+ETA:5d9h24m
+Total train time:8d4h47m
+I1130 17:18:27.138770 137274321021824 utils.py:1231] [38600] l2_params = 326.2241590289701
+I1130 17:18:27.139662 137274321021824 utils.py:1231] [38600] train/loss = 3.079804927110672
+I1130 17:18:27.139856 137274321021824 utils.py:1231] [38600] l2_grads = 1.2782952785491943
+I1130 17:18:27.139943 137274321021824 utils.py:1231] [38600] lr = 0.0008202410830509247
+I1130 17:18:27.139997 137274321021824 utils.py:1231] [38600] uptime = 242896.50235921599
+I1130 17:18:27.140056 137274321021824 utils.py:1231] [38600] examples_seen = 39526400.0
+I1130 17:18:27.140125 137274321021824 utils.py:1231] [38600] progress = 0.34279726117421383
+I1130 17:18:27.140185 137274321021824 utils.py:1231] [38600] epoch = 30.851871770034663
+I1130 17:18:27.140243 137274321021824 utils.py:1231] [38600] img/sec/core = 201.0530430591502
+I1130 17:18:27.140387 137274321021824 utils.py:1231] [38600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.4369330529811
+I1130 17:18:27.140445 137274321021824 utils.py:1231] [38600] core_hours = 67.4369330529811
+I1130 17:18:27.140514 137274321021824 train.py:125] NOTE: Steps:38600/112603 [34.3%]
+Walltime:2d19h28m (0s eval)
+ETA:5d9h17m
+Total train time:8d4h44m
+I1130 17:22:40.599002 137274321021824 utils.py:1231] [38650] l2_params = 326.1624294594557
+I1130 17:22:40.600492 137274321021824 utils.py:1231] [38650] train/loss = 2.6248519718647003
+I1130 17:22:40.600672 137274321021824 utils.py:1231] [38650] l2_grads = 1.4644490480422974
+I1130 17:22:40.600734 137274321021824 utils.py:1231] [38650] lr = 0.0008196528454356368
+I1130 17:22:40.600787 137274321021824 utils.py:1231] [38650] uptime = 243149.963149501
+I1130 17:22:40.600838 137274321021824 utils.py:1231] [38650] examples_seen = 39577600.0
+I1130 17:22:40.600893 137274321021824 utils.py:1231] [38650] progress = 0.34324129907728923
+I1130 17:22:40.600944 137274321021824 utils.py:1231] [38650] epoch = 30.89183533450362
+I1130 17:22:40.600994 137274321021824 utils.py:1231] [38650] img/sec/core = 202.00363118265264
+I1130 17:22:40.601154 137274321021824 utils.py:1231] [38650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.50733882806028
+I1130 17:22:40.601230 137274321021824 utils.py:1231] [38650] core_hours = 67.50733882806028
+I1130 17:22:40.601294 137274321021824 train.py:125] NOTE: Steps:38650/112603 [34.3%]
+Walltime:2d19h32m (0s eval)
+ETA:5d9h10m
+Total train time:8d4h41m
+I1130 17:26:55.579876 137274321021824 utils.py:1231] [38700] l2_params = 326.1229844704534
+I1130 17:26:55.580950 137274321021824 utils.py:1231] [38700] train/loss = 3.3533547818660736
+I1130 17:26:55.581107 137274321021824 utils.py:1231] [38700] l2_grads = 1.3086591958999634
+I1130 17:26:55.581161 137274321021824 utils.py:1231] [38700] lr = 0.0008190638586197336
+I1130 17:26:55.581212 137274321021824 utils.py:1231] [38700] uptime = 243404.943574308
+I1130 17:26:55.581256 137274321021824 utils.py:1231] [38700] examples_seen = 39628800.0
+I1130 17:26:55.581303 137274321021824 utils.py:1231] [38700] progress = 0.34368533698036463
+I1130 17:26:55.581344 137274321021824 utils.py:1231] [38700] epoch = 30.931798898972577
+I1130 17:26:55.581386 137274321021824 utils.py:1231] [38700] img/sec/core = 200.7997282095457
+I1130 17:26:55.581536 137274321021824 utils.py:1231] [38700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.57816672384
+I1130 17:26:55.581585 137274321021824 utils.py:1231] [38700] core_hours = 67.57816672384
+I1130 17:26:55.581657 137274321021824 train.py:125] NOTE: Steps:38700/112603 [34.4%]
+Walltime:2d19h36m (0s eval)
+ETA:5d9h3m
+Total train time:8d4h38m
+I1130 17:31:16.763493 137274321021824 utils.py:1231] [38750] l2_params = 326.0607481654048
+I1130 17:31:16.763697 137274321021824 utils.py:1231] [38750] train/loss = 4.957574367523193
+I1130 17:31:16.763793 137274321021824 utils.py:1231] [38750] l2_grads = 1.1743714809417725
+I1130 17:31:16.763869 137274321021824 utils.py:1231] [38750] lr = 0.0008184741239836789
+I1130 17:31:16.763956 137274321021824 utils.py:1231] [38750] uptime = 243666.12631410698
+I1130 17:31:16.764013 137274321021824 utils.py:1231] [38750] examples_seen = 39680000.0
+I1130 17:31:16.764069 137274321021824 utils.py:1231] [38750] progress = 0.34412937488344003
+I1130 17:31:16.764121 137274321021824 utils.py:1231] [38750] epoch = 30.971762463441536
+I1130 17:31:16.764189 137274321021824 utils.py:1231] [38750] img/sec/core = 196.03133055196662
+I1130 17:31:16.764261 137274321021824 utils.py:1231] [38750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.65071748489527
+I1130 17:31:16.764313 137274321021824 utils.py:1231] [38750] core_hours = 67.65071748489527
+I1130 17:31:16.764376 137274321021824 train.py:125] NOTE: Steps:38750/112603 [34.4%]
+Walltime:2d19h41m (0s eval)
+ETA:5d8h56m
+Total train time:8d4h35m
+I1130 17:35:31.460044 137274321021824 utils.py:1231] [38800] l2_params = 325.9951666052407
+I1130 17:35:31.460288 137274321021824 utils.py:1231] [38800] train/loss = 2.523371934890747
+I1130 17:35:31.460401 137274321021824 utils.py:1231] [38800] l2_grads = 1.5123885869979858
+I1130 17:35:31.460487 137274321021824 utils.py:1231] [38800] lr = 0.0008178836429096895
+I1130 17:35:31.460561 137274321021824 utils.py:1231] [38800] uptime = 243920.82292298198
+I1130 17:35:31.460630 137274321021824 utils.py:1231] [38800] examples_seen = 39731200.0
+I1130 17:35:31.460682 137274321021824 utils.py:1231] [38800] progress = 0.34457341278651543
+I1130 17:35:31.460732 137274321021824 utils.py:1231] [38800] epoch = 31.01172602791049
+I1130 17:35:31.460807 137274321021824 utils.py:1231] [38800] img/sec/core = 201.02348526017056
+I1130 17:35:31.460865 137274321021824 utils.py:1231] [38800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.7214665429161
+I1130 17:35:31.460933 137274321021824 utils.py:1231] [38800] core_hours = 67.7214665429161
+I1130 17:35:31.461007 137274321021824 train.py:125] NOTE: Steps:38800/112603 [34.5%]
+Walltime:2d19h45m (0s eval)
+ETA:5d8h49m
+Total train time:8d4h32m
+I1130 17:39:46.096864 137274321021824 utils.py:1231] [38850] l2_params = 325.94246777491134
+I1130 17:39:46.097911 137274321021824 utils.py:1231] [38850] train/loss = 2.5181738436222076
+I1130 17:39:46.098067 137274321021824 utils.py:1231] [38850] l2_grads = 1.429438829421997
+I1130 17:39:46.098130 137274321021824 utils.py:1231] [38850] lr = 0.0008172924167817329
+I1130 17:39:46.098177 137274321021824 utils.py:1231] [38850] uptime = 244175.46054059902
+I1130 17:39:46.098230 137274321021824 utils.py:1231] [38850] examples_seen = 39782400.0
+I1130 17:39:46.098278 137274321021824 utils.py:1231] [38850] progress = 0.3450174506895909
+I1130 17:39:46.098329 137274321021824 utils.py:1231] [38850] epoch = 31.051689592379446
+I1130 17:39:46.098386 137274321021824 utils.py:1231] [38850] img/sec/core = 201.07005586662248
+I1130 17:39:46.098441 137274321021824 utils.py:1231] [38850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.7921992144764
+I1130 17:39:46.098490 137274321021824 utils.py:1231] [38850] core_hours = 67.7921992144764
+I1130 17:39:46.098552 137274321021824 train.py:125] NOTE: Steps:38850/112603 [34.5%]
+Walltime:2d19h49m (0s eval)
+ETA:5d8h42m
+Total train time:8d4h29m
+I1130 17:44:02.659425 137274321021824 utils.py:1231] [38900] l2_params = 325.87433384893814
+I1130 17:44:02.660451 137274321021824 utils.py:1231] [38900] train/loss = 5.0724775195121765
+I1130 17:44:02.660628 137274321021824 utils.py:1231] [38900] l2_grads = 1.1555731296539307
+I1130 17:44:02.660698 137274321021824 utils.py:1231] [38900] lr = 0.0008167004469855203
+I1130 17:44:02.660751 137274321021824 utils.py:1231] [38900] uptime = 244432.023115077
+I1130 17:44:02.660796 137274321021824 utils.py:1231] [38900] examples_seen = 39833600.0
+I1130 17:44:02.660842 137274321021824 utils.py:1231] [38900] progress = 0.3454614885926663
+I1130 17:44:02.660893 137274321021824 utils.py:1231] [38900] epoch = 31.091653156848405
+I1130 17:44:02.660937 137274321021824 utils.py:1231] [38900] img/sec/core = 199.56145242218017
+I1130 17:44:02.661094 137274321021824 utils.py:1231] [38900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.86346659627583
+I1130 17:44:02.661149 137274321021824 utils.py:1231] [38900] core_hours = 67.86346659627583
+I1130 17:44:02.661224 137274321021824 train.py:125] NOTE: Steps:38900/112603 [34.5%]
+Walltime:2d19h53m (0s eval)
+ETA:5d8h35m
+Total train time:8d4h27m
+I1130 17:48:19.759004 137274321021824 utils.py:1231] [38950] l2_params = 325.8255402834705
+I1130 17:48:19.759467 137274321021824 utils.py:1231] [38950] train/loss = 5.2159610986709595
+I1130 17:48:19.759658 137274321021824 utils.py:1231] [38950] l2_grads = 1.3198883533477783
+I1130 17:48:19.759763 137274321021824 utils.py:1231] [38950] lr = 0.0008161077349085086
+I1130 17:48:19.759834 137274321021824 utils.py:1231] [38950] uptime = 244689.12219186098
+I1130 17:48:19.759912 137274321021824 utils.py:1231] [38950] examples_seen = 39884800.0
+I1130 17:48:19.759973 137274321021824 utils.py:1231] [38950] progress = 0.3459055264957417
+I1130 17:48:19.760035 137274321021824 utils.py:1231] [38950] epoch = 31.13161672131736
+I1130 17:48:19.760107 137274321021824 utils.py:1231] [38950] img/sec/core = 199.1450169345365
+I1130 17:48:19.760170 137274321021824 utils.py:1231] [38950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 67.9348830064936
+I1130 17:48:19.760221 137274321021824 utils.py:1231] [38950] core_hours = 67.9348830064936
+I1130 17:48:19.760289 137274321021824 train.py:125] NOTE: Steps:38950/112603 [34.6%]
+Walltime:2d19h58m (0s eval)
+ETA:5d8h28m
+Total train time:8d4h24m
+I1130 17:52:56.147008 137274321021824 utils.py:1231] [39000] l2_params = 325.77446944043743
+I1130 17:52:56.148251 137274321021824 utils.py:1231] [39000] train/loss = 2.828322619199753
+I1130 17:52:56.148462 137274321021824 utils.py:1231] [39000] l2_grads = 1.4294220209121704
+I1130 17:52:56.148537 137274321021824 utils.py:1231] [39000] lr = 0.0008155142819398924
+I1130 17:52:56.148590 137274321021824 utils.py:1231] [39000] uptime = 244965.51095267
+I1130 17:52:56.148674 137274321021824 utils.py:1231] [39000] examples_seen = 39936000.0
+I1130 17:52:56.148720 137274321021824 utils.py:1231] [39000] progress = 0.3463495643988171
+I1130 17:52:56.148765 137274321021824 utils.py:1231] [39000] epoch = 31.17158028578632
+I1130 17:52:56.148809 137274321021824 utils.py:1231] [39000] img/sec/core = 185.24631699975856
+I1130 17:52:56.148900 137274321021824 utils.py:1231] [39000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.01165766227389
+I1130 17:52:56.148975 137274321021824 utils.py:1231] [39000] core_hours = 68.01165766227389
+I1130 17:52:56.149038 137274321021824 train.py:125] NOTE: Steps:39000/112603 [34.6%]
+Walltime:2d20h2m (0s eval)
+ETA:5d8h21m
+Total train time:8d4h22m
+I1130 17:57:46.874418 137274321021824 utils.py:1231] [39050] l2_params = 325.738187428077
+I1130 17:57:46.874728 137274321021824 utils.py:1231] [39050] train/loss = 4.017236769199371
+I1130 17:57:46.874890 137274321021824 utils.py:1231] [39050] l2_grads = 1.1223042011260986
+I1130 17:57:46.874992 137274321021824 utils.py:1231] [39050] lr = 0.0008149200894706029
+I1130 17:57:46.875065 137274321021824 utils.py:1231] [39050] uptime = 245256.237427173
+I1130 17:57:46.875140 137274321021824 utils.py:1231] [39050] examples_seen = 39987200.0
+I1130 17:57:46.875198 137274321021824 utils.py:1231] [39050] progress = 0.3467936023018925
+I1130 17:57:46.875250 137274321021824 utils.py:1231] [39050] epoch = 31.211543850255275
+I1130 17:57:46.875342 137274321021824 utils.py:1231] [39050] img/sec/core = 176.11055232423354
+I1130 17:57:46.875448 137274321021824 utils.py:1231] [39050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.0924150163025
+I1130 17:57:46.875523 137274321021824 utils.py:1231] [39050] core_hours = 68.0924150163025
+I1130 17:57:46.875614 137274321021824 train.py:125] NOTE: Steps:39050/112603 [34.7%]
+Walltime:2d20h7m (0s eval)
+ETA:5d8h15m
+Total train time:8d4h21m
+I1130 18:02:30.196484 137274321021824 utils.py:1231] [39100] l2_params = 325.6783165420255
+I1130 18:02:30.196774 137274321021824 utils.py:1231] [39100] train/loss = 3.038384258747101
+I1130 18:02:30.196943 137274321021824 utils.py:1231] [39100] l2_grads = 1.3248380422592163
+I1130 18:02:30.197055 137274321021824 utils.py:1231] [39100] lr = 0.0008143251588933058
+I1130 18:02:30.197172 137274321021824 utils.py:1231] [39100] uptime = 245539.55953096302
+I1130 18:02:30.197257 137274321021824 utils.py:1231] [39100] examples_seen = 40038400.0
+I1130 18:02:30.197361 137274321021824 utils.py:1231] [39100] progress = 0.3472376402049679
+I1130 18:02:30.197435 137274321021824 utils.py:1231] [39100] epoch = 31.25150741472423
+I1130 18:02:30.197504 137274321021824 utils.py:1231] [39100] img/sec/core = 180.7130446763461
+I1130 18:02:30.197596 137274321021824 utils.py:1231] [39100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.17111560068862
+I1130 18:02:30.197672 137274321021824 utils.py:1231] [39100] core_hours = 68.17111560068862
+I1130 18:02:30.197762 137274321021824 train.py:125] NOTE: Steps:39100/112603 [34.7%]
+Walltime:2d20h12m (0s eval)
+ETA:5d8h9m
+Total train time:8d4h20m
+I1130 18:07:12.050948 137274321021824 utils.py:1231] [39150] l2_params = 325.61272301639707
+I1130 18:07:12.051192 137274321021824 utils.py:1231] [39150] train/loss = 3.755597621202469
+I1130 18:07:12.051323 137274321021824 utils.py:1231] [39150] l2_grads = 1.2187167406082153
+I1130 18:07:12.051397 137274321021824 utils.py:1231] [39150] lr = 0.0008137294916023965
+I1130 18:07:12.051457 137274321021824 utils.py:1231] [39150] uptime = 245821.41381890298
+I1130 18:07:12.051520 137274321021824 utils.py:1231] [39150] examples_seen = 40089600.0
+I1130 18:07:12.051683 137274321021824 utils.py:1231] [39150] progress = 0.3476816781080433
+I1130 18:07:12.051746 137274321021824 utils.py:1231] [39150] epoch = 31.29147097919319
+I1130 18:07:12.051805 137274321021824 utils.py:1231] [39150] img/sec/core = 181.65414609873284
+I1130 18:07:12.051859 137274321021824 utils.py:1231] [39150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.24940845844972
+I1130 18:07:12.051914 137274321021824 utils.py:1231] [39150] core_hours = 68.24940845844972
+I1130 18:07:12.051992 137274321021824 train.py:125] NOTE: Steps:39150/112603 [34.8%]
+Walltime:2d20h17m (0s eval)
+ETA:5d8h3m
+Total train time:8d4h18m
+I1130 18:11:54.689302 137274321021824 utils.py:1231] [39200] l2_params = 325.5498970113013
+I1130 18:11:54.689508 137274321021824 utils.py:1231] [39200] train/loss = 5.117755651473999
+I1130 18:11:54.689601 137274321021824 utils.py:1231] [39200] l2_grads = 1.3383833169937134
+I1130 18:11:54.689663 137274321021824 utils.py:1231] [39200] lr = 0.0008131330889939963
+I1130 18:11:54.689714 137274321021824 utils.py:1231] [39200] uptime = 246104.05207604
+I1130 18:11:54.689770 137274321021824 utils.py:1231] [39200] examples_seen = 40140800.0
+I1130 18:11:54.689819 137274321021824 utils.py:1231] [39200] progress = 0.3481257160111187
+I1130 18:11:54.689867 137274321021824 utils.py:1231] [39200] epoch = 31.331434543662144
+I1130 18:11:54.689921 137274321021824 utils.py:1231] [39200] img/sec/core = 181.15028205532784
+I1130 18:11:54.689977 137274321021824 utils.py:1231] [39200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.32791908543223
+I1130 18:11:54.690024 137274321021824 utils.py:1231] [39200] core_hours = 68.32791908543223
+I1130 18:11:54.690082 137274321021824 train.py:125] NOTE: Steps:39200/112603 [34.8%]
+Walltime:2d20h21m (0s eval)
+ETA:5d7h57m
+Total train time:8d4h17m
+I1130 18:16:47.546844 137274321021824 utils.py:1231] [39250] l2_params = 325.51466858845595
+I1130 18:16:47.547081 137274321021824 utils.py:1231] [39250] train/loss = 2.6173174679279327
+I1130 18:16:47.547179 137274321021824 utils.py:1231] [39250] l2_grads = 1.4606984853744507
+I1130 18:16:47.547259 137274321021824 utils.py:1231] [39250] lr = 0.0008125359524659509
+I1130 18:16:47.547319 137274321021824 utils.py:1231] [39250] uptime = 246396.90968101
+I1130 18:16:47.547378 137274321021824 utils.py:1231] [39250] examples_seen = 40192000.0
+I1130 18:16:47.547448 137274321021824 utils.py:1231] [39250] progress = 0.3485697539141941
+I1130 18:16:47.547515 137274321021824 utils.py:1231] [39250] epoch = 31.371398108131103
+I1130 18:16:47.547572 137274321021824 utils.py:1231] [39250] img/sec/core = 174.8289924219191
+I1130 18:16:47.547642 137274321021824 utils.py:1231] [39250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.40926842014612
+I1130 18:16:47.547715 137274321021824 utils.py:1231] [39250] core_hours = 68.40926842014612
+I1130 18:16:47.547783 137274321021824 train.py:125] NOTE: Steps:39250/112603 [34.9%]
+Walltime:2d20h26m (0s eval)
+ETA:5d7h51m
+Total train time:8d4h16m
+I1130 18:21:40.190871 137274321021824 utils.py:1231] [39300] l2_params = 325.45432290641475
+I1130 18:21:40.191128 137274321021824 utils.py:1231] [39300] train/loss = 4.013837993144989
+I1130 18:21:40.191253 137274321021824 utils.py:1231] [39300] l2_grads = 1.1411042213439941
+I1130 18:21:40.191349 137274321021824 utils.py:1231] [39300] lr = 0.0008119380834178253
+I1130 18:21:40.191421 137274321021824 utils.py:1231] [39300] uptime = 246689.55378262
+I1130 18:21:40.191491 137274321021824 utils.py:1231] [39300] examples_seen = 40243200.0
+I1130 18:21:40.191568 137274321021824 utils.py:1231] [39300] progress = 0.34901379181726955
+I1130 18:21:40.191637 137274321021824 utils.py:1231] [39300] epoch = 31.41136167260006
+I1130 18:21:40.191707 137274321021824 utils.py:1231] [39300] img/sec/core = 174.95654181416953
+I1130 18:21:40.191771 137274321021824 utils.py:1231] [39300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.4905584483711
+I1130 18:21:40.191838 137274321021824 utils.py:1231] [39300] core_hours = 68.4905584483711
+I1130 18:21:40.191926 137274321021824 train.py:125] NOTE: Steps:39300/112603 [34.9%]
+Walltime:2d20h31m (0s eval)
+ETA:5d7h45m
+Total train time:8d4h15m
+I1130 18:26:35.416488 137274321021824 utils.py:1231] [39350] l2_params = 325.4323997066761
+I1130 18:26:35.416706 137274321021824 utils.py:1231] [39350] train/loss = 2.7595322132110596
+I1130 18:26:35.416821 137274321021824 utils.py:1231] [39350] l2_grads = 1.4204293489456177
+I1130 18:26:35.416896 137274321021824 utils.py:1231] [39350] lr = 0.0008113394832509018
+I1130 18:26:35.416959 137274321021824 utils.py:1231] [39350] uptime = 246984.779319547
+I1130 18:26:35.417019 137274321021824 utils.py:1231] [39350] examples_seen = 40294400.0
+I1130 18:26:35.417076 137274321021824 utils.py:1231] [39350] progress = 0.34945782972034495
+I1130 18:26:35.417132 137274321021824 utils.py:1231] [39350] epoch = 31.451325237069018
+I1130 18:26:35.417193 137274321021824 utils.py:1231] [39350] img/sec/core = 173.42673175545912
+I1130 18:26:35.417253 137274321021824 utils.py:1231] [39350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.57256554196194
+I1130 18:26:35.417304 137274321021824 utils.py:1231] [39350] core_hours = 68.57256554196194
+I1130 18:26:35.417372 137274321021824 train.py:125] NOTE: Steps:39350/112603 [34.9%]
+Walltime:2d20h36m (0s eval)
+ETA:5d7h39m
+Total train time:8d4h14m
+I1130 18:31:32.619752 137274321021824 utils.py:1231] [39400] l2_params = 325.3819787929015
+I1130 18:31:32.619992 137274321021824 utils.py:1231] [39400] train/loss = 3.103977292776108
+I1130 18:31:32.620130 137274321021824 utils.py:1231] [39400] l2_grads = 1.3992451429367065
+I1130 18:31:32.620198 137274321021824 utils.py:1231] [39400] lr = 0.0008107401533681771
+I1130 18:31:32.620255 137274321021824 utils.py:1231] [39400] uptime = 247281.98261784698
+I1130 18:31:32.620315 137274321021824 utils.py:1231] [39400] examples_seen = 40345600.0
+I1130 18:31:32.620365 137274321021824 utils.py:1231] [39400] progress = 0.34990186762342035
+I1130 18:31:32.620413 137274321021824 utils.py:1231] [39400] epoch = 31.491288801537973
+I1130 18:31:32.620464 137274321021824 utils.py:1231] [39400] img/sec/core = 172.272650717091
+I1130 18:31:32.620533 137274321021824 utils.py:1231] [39400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.65512201371193
+I1130 18:31:32.620585 137274321021824 utils.py:1231] [39400] core_hours = 68.65512201371193
+I1130 18:31:32.620665 137274321021824 train.py:125] NOTE: Steps:39400/112603 [35.0%]
+Walltime:2d20h41m (0s eval)
+ETA:5d7h33m
+Total train time:8d4h13m
+I1130 18:36:26.377838 137274321021824 utils.py:1231] [39450] l2_params = 325.3156756829622
+I1130 18:36:26.378042 137274321021824 utils.py:1231] [39450] train/loss = 2.6163199841976166
+I1130 18:36:26.378141 137274321021824 utils.py:1231] [39450] l2_grads = 1.37030827999115
+I1130 18:36:26.378204 137274321021824 utils.py:1231] [39450] lr = 0.000810140095174356
+I1130 18:36:26.378263 137274321021824 utils.py:1231] [39450] uptime = 247575.74062488897
+I1130 18:36:26.378327 137274321021824 utils.py:1231] [39450] examples_seen = 40396800.0
+I1130 18:36:26.378386 137274321021824 utils.py:1231] [39450] progress = 0.35034590552649575
+I1130 18:36:26.378448 137274321021824 utils.py:1231] [39450] epoch = 31.531252366006928
+I1130 18:36:26.378501 137274321021824 utils.py:1231] [39450] img/sec/core = 174.2931214558547
+I1130 18:36:26.378559 137274321021824 utils.py:1231] [39450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.73672146011249
+I1130 18:36:26.378611 137274321021824 utils.py:1231] [39450] core_hours = 68.73672146011249
+I1130 18:36:26.378674 137274321021824 train.py:125] NOTE: Steps:39450/112603 [35.0%]
+Walltime:2d20h46m (0s eval)
+ETA:5d7h27m
+Total train time:8d4h12m
+I1130 18:41:25.940865 137274321021824 utils.py:1231] [39500] l2_params = 325.2628298754489
+I1130 18:41:25.941153 137274321021824 utils.py:1231] [39500] train/loss = 2.9417694211006165
+I1130 18:41:25.941269 137274321021824 utils.py:1231] [39500] l2_grads = 1.2797503471374512
+I1130 18:41:25.941348 137274321021824 utils.py:1231] [39500] lr = 0.0008095393100758517
+I1130 18:41:25.941406 137274321021824 utils.py:1231] [39500] uptime = 247875.303768214
+I1130 18:41:25.941462 137274321021824 utils.py:1231] [39500] examples_seen = 40448000.0
+I1130 18:41:25.941510 137274321021824 utils.py:1231] [39500] progress = 0.35078994342957115
+I1130 18:41:25.941558 137274321021824 utils.py:1231] [39500] epoch = 31.571215930475887
+I1130 18:41:25.941606 137274321021824 utils.py:1231] [39500] img/sec/core = 170.915551999152
+I1130 18:41:25.941670 137274321021824 utils.py:1231] [39500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.81993344436945
+I1130 18:41:25.941720 137274321021824 utils.py:1231] [39500] core_hours = 68.81993344436945
+I1130 18:41:25.941779 137274321021824 train.py:125] NOTE: Steps:39500/112603 [35.1%]
+Walltime:2d20h51m (0s eval)
+ETA:5d7h22m
+Total train time:8d4h11m
+I1130 18:46:24.187607 137274321021824 utils.py:1231] [39550] l2_params = 325.21498729676756
+I1130 18:46:24.187809 137274321021824 utils.py:1231] [39550] train/loss = 2.4269077479839325
+I1130 18:46:24.442162 137274321021824 utils.py:1231] [39550] l2_grads = 1.4938966035842896
+I1130 18:46:24.442468 137274321021824 utils.py:1231] [39550] lr = 0.0008089377994807825
+I1130 18:46:24.442562 137274321021824 utils.py:1231] [39550] uptime = 248173.804920926
+I1130 18:46:24.442638 137274321021824 utils.py:1231] [39550] examples_seen = 40499200.0
+I1130 18:46:24.442713 137274321021824 utils.py:1231] [39550] progress = 0.35123398133264655
+I1130 18:46:24.442781 137274321021824 utils.py:1231] [39550] epoch = 31.611179494944842
+I1130 18:46:24.442864 137274321021824 utils.py:1231] [39550] img/sec/core = 171.5236257375611
+I1130 18:46:24.442969 137274321021824 utils.py:1231] [39550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.90285043123389
+I1130 18:46:24.443035 137274321021824 utils.py:1231] [39550] core_hours = 68.90285043123389
+I1130 18:46:24.443168 137274321021824 train.py:125] NOTE: Steps:39550/112603 [35.1%]
+Walltime:2d20h56m (0s eval)
+ETA:5d7h16m
+Total train time:8d4h11m
+I1130 18:51:27.330615 137274321021824 utils.py:1231] [39600] l2_params = 325.1700768837373
+I1130 18:51:27.330855 137274321021824 utils.py:1231] [39600] train/loss = 3.154658854007721
+I1130 18:51:27.330987 137274321021824 utils.py:1231] [39600] l2_grads = 1.2598942518234253
+I1130 18:51:27.331056 137274321021824 utils.py:1231] [39600] lr = 0.0008083355647989651
+I1130 18:51:27.331114 137274321021824 utils.py:1231] [39600] uptime = 248476.69347613497
+I1130 18:51:27.331171 137274321021824 utils.py:1231] [39600] examples_seen = 40550400.0
+I1130 18:51:27.331227 137274321021824 utils.py:1231] [39600] progress = 0.35167801923572195
+I1130 18:51:27.331294 137274321021824 utils.py:1231] [39600] epoch = 31.6511430594138
+I1130 18:51:27.331372 137274321021824 utils.py:1231] [39600] img/sec/core = 169.0390710361261
+I1130 18:51:27.331451 137274321021824 utils.py:1231] [39600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 68.98698614101416
+I1130 18:51:27.331505 137274321021824 utils.py:1231] [39600] core_hours = 68.98698614101416
+I1130 18:51:27.331592 137274321021824 train.py:125] NOTE: Steps:39600/112603 [35.2%]
+Walltime:2d21h1m (0s eval)
+ETA:5d7h11m
+Total train time:8d4h10m
+I1130 18:56:32.489422 137274321021824 utils.py:1231] [39650] l2_params = 325.1142057784196
+I1130 18:56:32.489621 137274321021824 utils.py:1231] [39650] train/loss = 2.719775378704071
+I1130 18:56:32.489726 137274321021824 utils.py:1231] [39650] l2_grads = 1.2938737869262695
+I1130 18:56:32.489788 137274321021824 utils.py:1231] [39650] lr = 0.0008077326074419132
+I1130 18:56:32.489861 137274321021824 utils.py:1231] [39650] uptime = 248781.85221826
+I1130 18:56:32.489936 137274321021824 utils.py:1231] [39650] examples_seen = 40601600.0
+I1130 18:56:32.489986 137274321021824 utils.py:1231] [39650] progress = 0.35212205713879735
+I1130 18:56:32.490035 137274321021824 utils.py:1231] [39650] epoch = 31.691106623882757
+I1130 18:56:32.490084 137274321021824 utils.py:1231] [39650] img/sec/core = 167.78152788106044
+I1130 18:56:32.490142 137274321021824 utils.py:1231] [39650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.0717524582711
+I1130 18:56:32.490191 137274321021824 utils.py:1231] [39650] core_hours = 69.0717524582711
+I1130 18:56:32.490251 137274321021824 train.py:125] NOTE: Steps:39650/112603 [35.2%]
+Walltime:2d21h6m (0s eval)
+ETA:5d7h5m
+Total train time:8d4h10m
+I1130 19:01:44.259097 137274321021824 utils.py:1231] [39700] l2_params = 325.06858522186883
+I1130 19:01:44.259307 137274321021824 utils.py:1231] [39700] train/loss = 2.6399873793125153
+I1130 19:01:44.259409 137274321021824 utils.py:1231] [39700] l2_grads = 1.4635792970657349
+I1130 19:01:44.259466 137274321021824 utils.py:1231] [39700] lr = 0.0008071289288228353
+I1130 19:01:44.259521 137274321021824 utils.py:1231] [39700] uptime = 249093.621883781
+I1130 19:01:44.259572 137274321021824 utils.py:1231] [39700] examples_seen = 40652800.0
+I1130 19:01:44.259622 137274321021824 utils.py:1231] [39700] progress = 0.35256609504187275
+I1130 19:01:44.259670 137274321021824 utils.py:1231] [39700] epoch = 31.731070188351715
+I1130 19:01:44.259719 137274321021824 utils.py:1231] [39700] img/sec/core = 164.22380257694877
+I1130 19:01:44.259773 137274321021824 utils.py:1231] [39700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.15835514313805
+I1130 19:01:44.259821 137274321021824 utils.py:1231] [39700] core_hours = 69.15835514313805
+I1130 19:01:44.259879 137274321021824 train.py:125] NOTE: Steps:39700/112603 [35.3%]
+Walltime:2d21h11m (0s eval)
+ETA:5d7h0m
+Total train time:8d4h10m
+I1130 19:06:53.671554 137274321021824 utils.py:1231] [39750] l2_params = 325.0273925144139
+I1130 19:06:53.671767 137274321021824 utils.py:1231] [39750] train/loss = 4.306567072868347
+I1130 19:06:53.671862 137274321021824 utils.py:1231] [39750] l2_grads = 1.2089980840682983
+I1130 19:06:53.671935 137274321021824 utils.py:1231] [39750] lr = 0.0008065245303566301
+I1130 19:06:53.671991 137274321021824 utils.py:1231] [39750] uptime = 249403.03435280698
+I1130 19:06:53.672048 137274321021824 utils.py:1231] [39750] examples_seen = 40704000.0
+I1130 19:06:53.672101 137274321021824 utils.py:1231] [39750] progress = 0.3530101329449482
+I1130 19:06:53.672153 137274321021824 utils.py:1231] [39750] epoch = 31.77103375282067
+I1130 19:06:53.672209 137274321021824 utils.py:1231] [39750] img/sec/core = 165.47490849730082
+I1130 19:06:53.672274 137274321021824 utils.py:1231] [39750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.24430305120083
+I1130 19:06:53.672326 137274321021824 utils.py:1231] [39750] core_hours = 69.24430305120083
+I1130 19:06:53.672387 137274321021824 train.py:125] NOTE: Steps:39750/112603 [35.3%]
+Walltime:2d21h16m (0s eval)
+ETA:5d6h54m
+Total train time:8d4h9m
+I1130 19:12:02.728837 137274321021824 utils.py:1231] [39800] l2_params = 324.97605434014565
+I1130 19:12:02.729047 137274321021824 utils.py:1231] [39800] train/loss = 2.56012424826622
+I1130 19:12:02.729152 137274321021824 utils.py:1231] [39800] l2_grads = 1.358958125114441
+I1130 19:12:02.729212 137274321021824 utils.py:1231] [39800] lr = 0.0008059194134598839
+I1130 19:12:02.729272 137274321021824 utils.py:1231] [39800] uptime = 249712.091630955
+I1130 19:12:02.729346 137274321021824 utils.py:1231] [39800] examples_seen = 40755200.0
+I1130 19:12:02.729401 137274321021824 utils.py:1231] [39800] progress = 0.3534541708480236
+I1130 19:12:02.729450 137274321021824 utils.py:1231] [39800] epoch = 31.810997317289626
+I1130 19:12:02.729501 137274321021824 utils.py:1231] [39800] img/sec/core = 165.66508417730606
+I1130 19:12:02.729556 137274321021824 utils.py:1231] [39800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.33015229513083
+I1130 19:12:02.729605 137274321021824 utils.py:1231] [39800] core_hours = 69.33015229513083
+I1130 19:12:02.729665 137274321021824 train.py:125] NOTE: Steps:39800/112603 [35.3%]
+Walltime:2d21h21m (0s eval)
+ETA:5d6h49m
+Total train time:8d4h9m
+I1130 19:17:10.899613 137274321021824 utils.py:1231] [39850] l2_params = 324.9065845295522
+I1130 19:17:10.899852 137274321021824 utils.py:1231] [39850] train/loss = 4.581334352493286
+I1130 19:17:10.899981 137274321021824 utils.py:1231] [39850] l2_grads = 1.1170130968093872
+I1130 19:17:10.900077 137274321021824 utils.py:1231] [39850] lr = 0.0008053135795508661
+I1130 19:17:10.900191 137274321021824 utils.py:1231] [39850] uptime = 250020.262532258
+I1130 19:17:10.900267 137274321021824 utils.py:1231] [39850] examples_seen = 40806400.0
+I1130 19:17:10.900323 137274321021824 utils.py:1231] [39850] progress = 0.353898208751099
+I1130 19:17:10.900377 137274321021824 utils.py:1231] [39850] epoch = 31.850960881758585
+I1130 19:17:10.900431 137274321021824 utils.py:1231] [39850] img/sec/core = 166.1415785316417
+I1130 19:17:10.900491 137274321021824 utils.py:1231] [39850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.41575532327056
+I1130 19:17:10.900545 137274321021824 utils.py:1231] [39850] core_hours = 69.41575532327056
+I1130 19:17:10.900609 137274321021824 train.py:125] NOTE: Steps:39850/112603 [35.4%]
+Walltime:2d21h27m (0s eval)
+ETA:5d6h44m
+Total train time:8d4h9m
+I1130 19:22:22.678084 137274321021824 utils.py:1231] [39900] l2_params = 324.8215469097245
+I1130 19:22:22.678281 137274321021824 utils.py:1231] [39900] train/loss = 3.2657524943351746
+I1130 19:22:22.678375 137274321021824 utils.py:1231] [39900] l2_grads = 1.2912099361419678
+I1130 19:22:22.678434 137274321021824 utils.py:1231] [39900] lr = 0.0008047070300495267
+I1130 19:22:22.678490 137274321021824 utils.py:1231] [39900] uptime = 250332.04085231497
+I1130 19:22:22.678541 137274321021824 utils.py:1231] [39900] examples_seen = 40857600.0
+I1130 19:22:22.678590 137274321021824 utils.py:1231] [39900] progress = 0.3543422466541744
+I1130 19:22:22.678637 137274321021824 utils.py:1231] [39900] epoch = 31.89092444622754
+I1130 19:22:22.678688 137274321021824 utils.py:1231] [39900] img/sec/core = 164.2192439507811
+I1130 19:22:22.678746 137274321021824 utils.py:1231] [39900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.50236041217526
+I1130 19:22:22.678795 137274321021824 utils.py:1231] [39900] core_hours = 69.50236041217526
+I1130 19:22:22.678854 137274321021824 train.py:125] NOTE: Steps:39900/112603 [35.4%]
+Walltime:2d21h32m (0s eval)
+ETA:5d6h38m
+Total train time:8d4h9m
+I1130 19:27:32.376900 137274321021824 utils.py:1231] [39950] l2_params = 324.7730207377017
+I1130 19:27:32.377150 137274321021824 utils.py:1231] [39950] train/loss = 3.2741629481315613
+I1130 19:27:32.377266 137274321021824 utils.py:1231] [39950] l2_grads = 1.2273238897323608
+I1130 19:27:32.377347 137274321021824 utils.py:1231] [39950] lr = 0.0008040997663774937
+I1130 19:27:32.377409 137274321021824 utils.py:1231] [39950] uptime = 250641.73976613797
+I1130 19:27:32.377465 137274321021824 utils.py:1231] [39950] examples_seen = 40908800.0
+I1130 19:27:32.626905 137274321021824 utils.py:1231] [39950] progress = 0.3547862845572498
+I1130 19:27:32.627118 137274321021824 utils.py:1231] [39950] epoch = 31.9308880106965
+I1130 19:27:32.627182 137274321021824 utils.py:1231] [39950] img/sec/core = 165.32185847207154
+I1130 19:27:32.627251 137274321021824 utils.py:1231] [39950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.58838788823722
+I1130 19:27:32.627319 137274321021824 utils.py:1231] [39950] core_hours = 69.58838788823722
+I1130 19:27:32.627388 137274321021824 train.py:125] NOTE: Steps:39950/112603 [35.5%]
+Walltime:2d21h37m (0s eval)
+ETA:5d6h33m
+Total train time:8d4h9m
+I1130 19:32:42.604837 137274321021824 utils.py:1231] [40000] l2_params = 324.7096378191278
+I1130 19:32:42.605042 137274321021824 utils.py:1231] [40000] train/loss = 3.7231034636497498
+I1130 19:32:42.605144 137274321021824 utils.py:1231] [40000] l2_grads = 1.2232614755630493
+I1130 19:32:42.605203 137274321021824 utils.py:1231] [40000] lr = 0.0008034917899580681
+I1130 19:32:42.605254 137274321021824 utils.py:1231] [40000] uptime = 250951.96761462797
+I1130 19:32:42.605313 137274321021824 utils.py:1231] [40000] examples_seen = 40960000.0
+I1130 19:32:42.605359 137274321021824 utils.py:1231] [40000] progress = 0.3552303224603252
+I1130 19:32:42.605406 137274321021824 utils.py:1231] [40000] epoch = 31.970851575165455
+I1130 19:32:42.605454 137274321021824 utils.py:1231] [40000] img/sec/core = 165.0399867362331
+I1130 19:32:42.605506 137274321021824 utils.py:1231] [40000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.67456229059555
+I1130 19:32:42.605553 137274321021824 utils.py:1231] [40000] core_hours = 69.67456229059555
+I1130 19:32:42.605612 137274321021824 train.py:125] NOTE: Steps:40000/112603 [35.5%]
+Walltime:2d21h42m (0s eval)
+ETA:5d6h28m
+Total train time:8d4h8m
+I1130 19:32:42.981023 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:40000/112603 [35.5%]
+Walltime:2d21h42m (0s eval)
+ETA:5d6h28m
+Total train time:8d4h8m
+I1130 19:34:19.355133 137274321021824 utils.py:1231] [40000] val/acc@1 = 0.6157326211734694
+I1130 19:34:19.355366 137274321021824 utils.py:1231] [40000] val/loss = 1.609428631407874
+I1130 19:34:19.355509 137274321021824 utils.py:1231] [40000] z/secs/eval/val = 96.37423648597905
+I1130 19:34:19.355575 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 96.37423648597905
+I1130 19:39:31.120789 137274321021824 utils.py:1231] [40050] l2_params = 324.6522395543257
+I1130 19:39:31.121045 137274321021824 utils.py:1231] [40050] train/loss = 4.899214029312134
+I1130 19:39:31.121153 137274321021824 utils.py:1231] [40050] l2_grads = 1.3068238496780396
+I1130 19:39:31.121225 137274321021824 utils.py:1231] [40050] lr = 0.0008028831022162218
+I1130 19:39:31.121287 137274321021824 utils.py:1231] [40050] uptime = 251360.48364503199
+I1130 19:39:31.121355 137274321021824 utils.py:1231] [40050] examples_seen = 41011200.0
+I1130 19:39:31.121411 137274321021824 utils.py:1231] [40050] progress = 0.3556743603634006
+I1130 19:39:31.121474 137274321021824 utils.py:1231] [40050] epoch = 32.01081513963441
+I1130 19:39:31.121530 137274321021824 utils.py:1231] [40050] img/sec/core = 125.33167902704751
+I1130 19:39:31.121592 137274321021824 utils.py:1231] [40050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.78803896570777
+I1130 19:39:31.121646 137274321021824 utils.py:1231] [40050] core_hours = 69.78803896570777
+I1130 19:39:31.121707 137274321021824 train.py:125] NOTE: Steps:40050/112603 [35.6%]
+Walltime:2d21h49m (0s eval)
+ETA:5d6h25m
+Total train time:8d4h13m
+I1130 19:44:42.887392 137274321021824 utils.py:1231] [40100] l2_params = 324.59289409283684
+I1130 19:44:42.887640 137274321021824 utils.py:1231] [40100] train/loss = 2.6780166923999786
+I1130 19:44:42.887762 137274321021824 utils.py:1231] [40100] l2_grads = 1.462078332901001
+I1130 19:44:42.887838 137274321021824 utils.py:1231] [40100] lr = 0.0008022737045785937
+I1130 19:44:42.887902 137274321021824 utils.py:1231] [40100] uptime = 251672.250264715
+I1130 19:44:42.887975 137274321021824 utils.py:1231] [40100] examples_seen = 41062400.0
+I1130 19:44:42.888021 137274321021824 utils.py:1231] [40100] progress = 0.356118398266476
+I1130 19:44:42.888067 137274321021824 utils.py:1231] [40100] epoch = 32.050778704103365
+I1130 19:44:42.888114 137274321021824 utils.py:1231] [40100] img/sec/core = 164.22540697929026
+I1130 19:44:42.888167 137274321021824 utils.py:1231] [40100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.8746408045086
+I1130 19:44:42.888215 137274321021824 utils.py:1231] [40100] core_hours = 69.8746408045086
+I1130 19:44:42.888272 137274321021824 train.py:125] NOTE: Steps:40100/112603 [35.6%]
+Walltime:2d21h54m (0s eval)
+ETA:5d6h20m
+Total train time:8d4h13m
+I1130 19:49:54.652447 137274321021824 utils.py:1231] [40150] l2_params = 324.5507445474542
+I1130 19:49:54.652726 137274321021824 utils.py:1231] [40150] train/loss = 2.566150575876236
+I1130 19:49:54.652855 137274321021824 utils.py:1231] [40150] l2_grads = 1.4473140239715576
+I1130 19:49:54.652940 137274321021824 utils.py:1231] [40150] lr = 0.0008016635984734867
+I1130 19:49:54.653012 137274321021824 utils.py:1231] [40150] uptime = 251984.01537052897
+I1130 19:49:54.653103 137274321021824 utils.py:1231] [40150] examples_seen = 41113600.0
+I1130 19:49:54.653180 137274321021824 utils.py:1231] [40150] progress = 0.3565624361695514
+I1130 19:49:54.653245 137274321021824 utils.py:1231] [40150] epoch = 32.09074226857233
+I1130 19:49:54.653312 137274321021824 utils.py:1231] [40150] img/sec/core = 164.22620442503998
+I1130 19:49:54.653376 137274321021824 utils.py:1231] [40150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 69.96124222279028
+I1130 19:49:54.653435 137274321021824 utils.py:1231] [40150] core_hours = 69.96124222279028
+I1130 19:49:54.653503 137274321021824 train.py:125] NOTE: Steps:40150/112603 [35.7%]
+Walltime:2d21h59m (0s eval)
+ETA:5d6h15m
+Total train time:8d4h13m
+I1130 19:55:06.431608 137274321021824 utils.py:1231] [40200] l2_params = 324.50862387793984
+I1130 19:55:06.431877 137274321021824 utils.py:1231] [40200] train/loss = 4.641149401664734
+I1130 19:55:06.432013 137274321021824 utils.py:1231] [40200] l2_grads = 1.1759753227233887
+I1130 19:55:06.432103 137274321021824 utils.py:1231] [40200] lr = 0.0008010527853308648
+I1130 19:55:06.432200 137274321021824 utils.py:1231] [40200] uptime = 252295.79455593298
+I1130 19:55:06.432296 137274321021824 utils.py:1231] [40200] examples_seen = 41164800.0
+I1130 19:55:06.432357 137274321021824 utils.py:1231] [40200] progress = 0.35700647407262687
+I1130 19:55:06.432410 137274321021824 utils.py:1231] [40200] epoch = 32.13070583304128
+I1130 19:55:06.432465 137274321021824 utils.py:1231] [40200] img/sec/core = 164.21878815820793
+I1130 19:55:06.432525 137274321021824 utils.py:1231] [40200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.04784755206916
+I1130 19:55:06.432574 137274321021824 utils.py:1231] [40200] core_hours = 70.04784755206916
+I1130 19:55:06.432635 137274321021824 train.py:125] NOTE: Steps:40200/112603 [35.7%]
+Walltime:2d22h4m (0s eval)
+ETA:5d6h10m
+Total train time:8d4h13m
+I1130 20:00:18.206740 137274321021824 utils.py:1231] [40250] l2_params = 324.4450567798782
+I1130 20:00:18.206964 137274321021824 utils.py:1231] [40250] train/loss = 2.648200750350952
+I1130 20:00:18.207064 137274321021824 utils.py:1231] [40250] l2_grads = 1.3971726894378662
+I1130 20:00:18.207142 137274321021824 utils.py:1231] [40250] lr = 0.0008004412665823475
+I1130 20:00:18.207201 137274321021824 utils.py:1231] [40250] uptime = 252607.569562589
+I1130 20:00:18.207260 137274321021824 utils.py:1231] [40250] examples_seen = 41216000.0
+I1130 20:00:18.207315 137274321021824 utils.py:1231] [40250] progress = 0.35745051197570227
+I1130 20:00:18.207369 137274321021824 utils.py:1231] [40250] epoch = 32.17066939751024
+I1130 20:00:18.207430 137274321021824 utils.py:1231] [40250] img/sec/core = 164.2209891971559
+I1130 20:00:18.207491 137274321021824 utils.py:1231] [40250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.13445172058472
+I1130 20:00:18.207547 137274321021824 utils.py:1231] [40250] core_hours = 70.13445172058472
+I1130 20:00:18.207612 137274321021824 train.py:125] NOTE: Steps:40250/112603 [35.7%]
+Walltime:2d22h10m (0s eval)
+ETA:5d6h4m
+Total train time:8d4h13m
+I1130 20:05:29.977159 137274321021824 utils.py:1231] [40300] l2_params = 324.38045287262264
+I1130 20:05:29.977400 137274321021824 utils.py:1231] [40300] train/loss = 2.582225978374481
+I1130 20:05:29.977502 137274321021824 utils.py:1231] [40300] l2_grads = 1.3900896310806274
+I1130 20:05:29.977597 137274321021824 utils.py:1231] [40300] lr = 0.0007998290436612102
+I1130 20:05:29.977688 137274321021824 utils.py:1231] [40300] uptime = 252919.34003753797
+I1130 20:05:29.977771 137274321021824 utils.py:1231] [40300] examples_seen = 41267200.0
+I1130 20:05:29.977832 137274321021824 utils.py:1231] [40300] progress = 0.35789454987877767
+I1130 20:05:29.977896 137274321021824 utils.py:1231] [40300] epoch = 32.210632961979194
+I1130 20:05:29.977957 137274321021824 utils.py:1231] [40300] img/sec/core = 164.22337621411688
+I1130 20:05:29.978019 137274321021824 utils.py:1231] [40300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.22105463029276
+I1130 20:05:29.978076 137274321021824 utils.py:1231] [40300] core_hours = 70.22105463029276
+I1130 20:05:29.978143 137274321021824 train.py:125] NOTE: Steps:40300/112603 [35.8%]
+Walltime:2d22h15m (0s eval)
+ETA:5d5h59m
+Total train time:8d4h12m
+I1130 20:10:41.790035 137274321021824 utils.py:1231] [40350] l2_params = 324.3203700066963
+I1130 20:10:41.790220 137274321021824 utils.py:1231] [40350] train/loss = 2.460991531610489
+I1130 20:10:41.790320 137274321021824 utils.py:1231] [40350] l2_grads = 1.4908581972122192
+I1130 20:10:41.790385 137274321021824 utils.py:1231] [40350] lr = 0.0007992161180023776
+I1130 20:10:41.790446 137274321021824 utils.py:1231] [40350] uptime = 253231.152807371
+I1130 20:10:41.790501 137274321021824 utils.py:1231] [40350] examples_seen = 41318400.0
+I1130 20:10:41.790554 137274321021824 utils.py:1231] [40350] progress = 0.35833858778185307
+I1130 20:10:41.790606 137274321021824 utils.py:1231] [40350] epoch = 32.25059652644815
+I1130 20:10:41.790661 137274321021824 utils.py:1231] [40350] img/sec/core = 164.20110063939381
+I1130 20:10:41.790720 137274321021824 utils.py:1231] [40350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.30766928857972
+I1130 20:10:41.790774 137274321021824 utils.py:1231] [40350] core_hours = 70.30766928857972
+I1130 20:10:41.790837 137274321021824 train.py:125] NOTE: Steps:40350/112603 [35.8%]
+Walltime:2d22h20m (0s eval)
+ETA:5d5h54m
+Total train time:8d4h12m
+I1130 20:15:53.424226 137274321021824 utils.py:1231] [40400] l2_params = 324.24544903734994
+I1130 20:15:53.424436 137274321021824 utils.py:1231] [40400] train/loss = 2.9209649562835693
+I1130 20:15:53.424532 137274321021824 utils.py:1231] [40400] l2_grads = 1.2473280429840088
+I1130 20:15:53.424595 137274321021824 utils.py:1231] [40400] lr = 0.0007986024910424207
+I1130 20:15:53.424648 137274321021824 utils.py:1231] [40400] uptime = 253542.78700997698
+I1130 20:15:53.424702 137274321021824 utils.py:1231] [40400] examples_seen = 41369600.0
+I1130 20:15:53.424753 137274321021824 utils.py:1231] [40400] progress = 0.35878262568492847
+I1130 20:15:53.424804 137274321021824 utils.py:1231] [40400] epoch = 32.29056009091711
+I1130 20:15:53.424858 137274321021824 utils.py:1231] [40400] img/sec/core = 164.29518830683205
+I1130 20:15:53.424921 137274321021824 utils.py:1231] [40400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.39423434485916
+I1130 20:15:53.424972 137274321021824 utils.py:1231] [40400] core_hours = 70.39423434485916
+I1130 20:15:53.425038 137274321021824 train.py:125] NOTE: Steps:40400/112603 [35.9%]
+Walltime:2d22h25m (0s eval)
+ETA:5d5h48m
+Total train time:8d4h12m
+I1130 20:21:05.211218 137274321021824 utils.py:1231] [40450] l2_params = 324.18478120846356
+I1130 20:21:05.211619 137274321021824 utils.py:1231] [40450] train/loss = 2.5996840596199036
+I1130 20:21:05.211901 137274321021824 utils.py:1231] [40450] l2_grads = 1.4222445487976074
+I1130 20:21:05.212008 137274321021824 utils.py:1231] [40450] lr = 0.0007979881642195551
+I1130 20:21:05.212084 137274321021824 utils.py:1231] [40450] uptime = 253854.574443408
+I1130 20:21:05.212155 137274321021824 utils.py:1231] [40450] examples_seen = 41420800.0
+I1130 20:21:05.212217 137274321021824 utils.py:1231] [40450] progress = 0.35922666358800387
+I1130 20:21:05.212282 137274321021824 utils.py:1231] [40450] epoch = 32.33052365538607
+I1130 20:21:05.212359 137274321021824 utils.py:1231] [40450] img/sec/core = 164.21444391320478
+I1130 20:21:05.212442 137274321021824 utils.py:1231] [40450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.48084196525666
+I1130 20:21:05.212510 137274321021824 utils.py:1231] [40450] core_hours = 70.48084196525666
+I1130 20:21:05.212614 137274321021824 train.py:125] NOTE: Steps:40450/112603 [35.9%]
+Walltime:2d22h30m (0s eval)
+ETA:5d5h43m
+Total train time:8d4h12m
+I1130 20:26:16.893571 137274321021824 utils.py:1231] [40500] l2_params = 324.11232092588443
+I1130 20:26:16.893765 137274321021824 utils.py:1231] [40500] train/loss = 4.987960338592529
+I1130 20:26:16.893867 137274321021824 utils.py:1231] [40500] l2_grads = 1.1314576864242554
+I1130 20:26:16.893942 137274321021824 utils.py:1231] [40500] lr = 0.0007973731389736368
+I1130 20:26:16.894003 137274321021824 utils.py:1231] [40500] uptime = 254166.25636349298
+I1130 20:26:16.894063 137274321021824 utils.py:1231] [40500] examples_seen = 41472000.0
+I1130 20:26:16.894120 137274321021824 utils.py:1231] [40500] progress = 0.35967070149107927
+I1130 20:26:16.894175 137274321021824 utils.py:1231] [40500] epoch = 32.37048721985502
+I1130 20:26:16.894231 137274321021824 utils.py:1231] [40500] img/sec/core = 164.27003525273253
+I1130 20:26:16.894297 137274321021824 utils.py:1231] [40500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.56742027639139
+I1130 20:26:16.894354 137274321021824 utils.py:1231] [40500] core_hours = 70.56742027639139
+I1130 20:26:16.894418 137274321021824 train.py:125] NOTE: Steps:40500/112603 [36.0%]
+Walltime:2d22h36m (0s eval)
+ETA:5d5h38m
+Total train time:8d4h12m
+I1130 20:31:28.675672 137274321021824 utils.py:1231] [40550] l2_params = 324.04691989300017
+I1130 20:31:28.675929 137274321021824 utils.py:1231] [40550] train/loss = 2.510503500699997
+I1130 20:31:28.676051 137274321021824 utils.py:1231] [40550] l2_grads = 1.3535457849502563
+I1130 20:31:28.676148 137274321021824 utils.py:1231] [40550] lr = 0.0007967574167461598
+I1130 20:31:28.676225 137274321021824 utils.py:1231] [40550] uptime = 254478.03858633398
+I1130 20:31:28.676286 137274321021824 utils.py:1231] [40550] examples_seen = 41523200.0
+I1130 20:31:28.676387 137274321021824 utils.py:1231] [40550] progress = 0.36011473939415467
+I1130 20:31:28.676454 137274321021824 utils.py:1231] [40550] epoch = 32.41045078432398
+I1130 20:31:28.676511 137274321021824 utils.py:1231] [40550] img/sec/core = 164.21718831003284
+I1130 20:31:28.676591 137274321021824 utils.py:1231] [40550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.65402644940278
+I1130 20:31:28.676678 137274321021824 utils.py:1231] [40550] core_hours = 70.65402644940278
+I1130 20:31:28.676805 137274321021824 train.py:125] NOTE: Steps:40550/112603 [36.0%]
+Walltime:2d22h41m (0s eval)
+ETA:5d5h33m
+Total train time:8d4h12m
+I1130 20:36:40.457191 137274321021824 utils.py:1231] [40600] l2_params = 323.9894297371744
+I1130 20:36:40.457464 137274321021824 utils.py:1231] [40600] train/loss = 2.5671925246715546
+I1130 20:36:40.457589 137274321021824 utils.py:1231] [40600] l2_grads = 1.4150192737579346
+I1130 20:36:40.457667 137274321021824 utils.py:1231] [40600] lr = 0.0007961409989802487
+I1130 20:36:40.457737 137274321021824 utils.py:1231] [40600] uptime = 254789.820095677
+I1130 20:36:40.457802 137274321021824 utils.py:1231] [40600] examples_seen = 41574400.0
+I1130 20:36:40.457861 137274321021824 utils.py:1231] [40600] progress = 0.36055877729723007
+I1130 20:36:40.457928 137274321021824 utils.py:1231] [40600] epoch = 32.45041434879294
+I1130 20:36:40.457988 137274321021824 utils.py:1231] [40600] img/sec/core = 164.21756411369347
+I1130 20:36:40.458052 137274321021824 utils.py:1231] [40600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.74063242422027
+I1130 20:36:40.458110 137274321021824 utils.py:1231] [40600] core_hours = 70.74063242422027
+I1130 20:36:40.458178 137274321021824 train.py:125] NOTE: Steps:40600/112603 [36.1%]
+Walltime:2d22h46m (0s eval)
+ETA:5d5h27m
+Total train time:8d4h12m
+I1130 20:41:52.246278 137274321021824 utils.py:1231] [40650] l2_params = 323.9399500958815
+I1130 20:41:52.246510 137274321021824 utils.py:1231] [40650] train/loss = 2.9291187822818756
+I1130 20:41:52.246639 137274321021824 utils.py:1231] [40650] l2_grads = 1.4269442558288574
+I1130 20:41:52.246703 137274321021824 utils.py:1231] [40650] lr = 0.0007955238871206619
+I1130 20:41:52.246756 137274321021824 utils.py:1231] [40650] uptime = 255101.60911838798
+I1130 20:41:52.246809 137274321021824 utils.py:1231] [40650] examples_seen = 41625600.0
+I1130 20:41:52.246857 137274321021824 utils.py:1231] [40650] progress = 0.36100281520030547
+I1130 20:41:52.246915 137274321021824 utils.py:1231] [40650] epoch = 32.490377913261895
+I1130 20:41:52.246967 137274321021824 utils.py:1231] [40650] img/sec/core = 164.2136068640861
+I1130 20:41:52.247022 137274321021824 utils.py:1231] [40650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.82724048608443
+I1130 20:41:52.247073 137274321021824 utils.py:1231] [40650] core_hours = 70.82724048608443
+I1130 20:41:52.247134 137274321021824 train.py:125] NOTE: Steps:40650/112603 [36.1%]
+Walltime:2d22h51m (0s eval)
+ETA:5d5h22m
+Total train time:8d4h12m
+I1130 20:47:04.041034 137274321021824 utils.py:1231] [40700] l2_params = 323.89249560240927
+I1130 20:47:04.041237 137274321021824 utils.py:1231] [40700] train/loss = 2.4251584708690643
+I1130 20:47:04.041329 137274321021824 utils.py:1231] [40700] l2_grads = 1.356008529663086
+I1130 20:47:04.041389 137274321021824 utils.py:1231] [40700] lr = 0.0007949060826137824
+I1130 20:47:04.041442 137274321021824 utils.py:1231] [40700] uptime = 255413.403803977
+I1130 20:47:04.041494 137274321021824 utils.py:1231] [40700] examples_seen = 41676800.0
+I1130 20:47:04.041542 137274321021824 utils.py:1231] [40700] progress = 0.3614468531033809
+I1130 20:47:04.041591 137274321021824 utils.py:1231] [40700] epoch = 32.53034147773085
+I1130 20:47:04.041640 137274321021824 utils.py:1231] [40700] img/sec/core = 164.2106243833967
+I1130 20:47:04.041695 137274321021824 utils.py:1231] [40700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 70.91385012097027
+I1130 20:47:04.041749 137274321021824 utils.py:1231] [40700] core_hours = 70.91385012097027
+I1130 20:47:04.041809 137274321021824 train.py:125] NOTE: Steps:40700/112603 [36.1%]
+Walltime:2d22h56m (0s eval)
+ETA:5d5h17m
+Total train time:8d4h12m
+I1130 20:52:15.824030 137274321021824 utils.py:1231] [40750] l2_params = 323.82207329675896
+I1130 20:52:15.824289 137274321021824 utils.py:1231] [40750] train/loss = 4.644877076148987
+I1130 20:52:15.824408 137274321021824 utils.py:1231] [40750] l2_grads = 1.2579479217529297
+I1130 20:52:15.824487 137274321021824 utils.py:1231] [40750] lr = 0.0007942875869076165
+I1130 20:52:15.824565 137274321021824 utils.py:1231] [40750] uptime = 255725.18691838498
+I1130 20:52:15.824620 137274321021824 utils.py:1231] [40750] examples_seen = 41728000.0
+I1130 20:52:15.824675 137274321021824 utils.py:1231] [40750] progress = 0.36189089100645633
+I1130 20:52:15.824728 137274321021824 utils.py:1231] [40750] epoch = 32.570305042199806
+I1130 20:52:15.824783 137274321021824 utils.py:1231] [40750] img/sec/core = 164.21671871878473
+I1130 20:52:15.824845 137274321021824 utils.py:1231] [40750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.00045654163915
+I1130 20:52:15.824918 137274321021824 utils.py:1231] [40750] core_hours = 71.00045654163915
+I1130 20:52:15.824991 137274321021824 train.py:125] NOTE: Steps:40750/112603 [36.2%]
+Walltime:2d23h2m (0s eval)
+ETA:5d5h11m
+Total train time:8d4h12m
+I1130 20:57:27.609531 137274321021824 utils.py:1231] [40800] l2_params = 323.7656135311468
+I1130 20:57:27.609774 137274321021824 utils.py:1231] [40800] train/loss = 3.733905792236328
+I1130 20:57:27.609903 137274321021824 utils.py:1231] [40800] l2_grads = 1.2624868154525757
+I1130 20:57:27.609979 137274321021824 utils.py:1231] [40800] lr = 0.0007936684014517912
+I1130 20:57:27.610040 137274321021824 utils.py:1231] [40800] uptime = 256036.97240186797
+I1130 20:57:27.610100 137274321021824 utils.py:1231] [40800] examples_seen = 41779200.0
+I1130 20:57:27.610157 137274321021824 utils.py:1231] [40800] progress = 0.36233492890953173
+I1130 20:57:27.610212 137274321021824 utils.py:1231] [40800] epoch = 32.61026860666876
+I1130 20:57:27.610275 137274321021824 utils.py:1231] [40800] img/sec/core = 164.21547093225374
+I1130 20:57:27.610333 137274321021824 utils.py:1231] [40800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.08706362038444
+I1130 20:57:27.610386 137274321021824 utils.py:1231] [40800] core_hours = 71.08706362038444
+I1130 20:57:27.610451 137274321021824 train.py:125] NOTE: Steps:40800/112603 [36.2%]
+Walltime:2d23h7m (0s eval)
+ETA:5d5h6m
+Total train time:8d4h12m
+I1130 21:02:39.387709 137274321021824 utils.py:1231] [40850] l2_params = 323.71216011594424
+I1130 21:02:39.387929 137274321021824 utils.py:1231] [40850] train/loss = 2.55643492937088
+I1130 21:02:39.388036 137274321021824 utils.py:1231] [40850] l2_grads = 1.4671399593353271
+I1130 21:02:39.388107 137274321021824 utils.py:1231] [40850] lr = 0.0007930485276975499
+I1130 21:02:39.388167 137274321021824 utils.py:1231] [40850] uptime = 256348.75052849296
+I1130 21:02:39.388228 137274321021824 utils.py:1231] [40850] examples_seen = 41830400.0
+I1130 21:02:39.388288 137274321021824 utils.py:1231] [40850] progress = 0.36277896681260713
+I1130 21:02:39.388344 137274321021824 utils.py:1231] [40850] epoch = 32.65023217113772
+I1130 21:02:39.388402 137274321021824 utils.py:1231] [40850] img/sec/core = 164.21934583494078
+I1130 21:02:39.388465 137274321021824 utils.py:1231] [40850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.17366865555805
+I1130 21:02:39.388526 137274321021824 utils.py:1231] [40850] core_hours = 71.17366865555805
+I1130 21:02:39.388592 137274321021824 train.py:125] NOTE: Steps:40850/112603 [36.3%]
+Walltime:2d23h12m (0s eval)
+ETA:5d5h1m
+Total train time:8d4h11m
+I1130 21:07:51.168822 137274321021824 utils.py:1231] [40900] l2_params = 323.65368504744976
+I1130 21:07:51.169042 137274321021824 utils.py:1231] [40900] train/loss = 4.094592273235321
+I1130 21:07:51.169156 137274321021824 utils.py:1231] [40900] l2_grads = 1.1927131414413452
+I1130 21:07:51.169245 137274321021824 utils.py:1231] [40900] lr = 0.00079242796709775
+I1130 21:07:51.169334 137274321021824 utils.py:1231] [40900] uptime = 256660.531690705
+I1130 21:07:51.169417 137274321021824 utils.py:1231] [40900] examples_seen = 41881600.0
+I1130 21:07:51.169487 137274321021824 utils.py:1231] [40900] progress = 0.36322300471568253
+I1130 21:07:51.169558 137274321021824 utils.py:1231] [40900] epoch = 32.69019573560668
+I1130 21:07:51.169641 137274321021824 utils.py:1231] [40900] img/sec/core = 164.217746950287
+I1130 21:07:51.169758 137274321021824 utils.py:1231] [40900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.26027453395027
+I1130 21:07:51.169828 137274321021824 utils.py:1231] [40900] core_hours = 71.26027453395027
+I1130 21:07:51.169926 137274321021824 train.py:125] NOTE: Steps:40900/112603 [36.3%]
+Walltime:2d23h17m (0s eval)
+ETA:5d4h56m
+Total train time:8d4h11m
+I1130 21:13:02.961525 137274321021824 utils.py:1231] [40950] l2_params = 323.62879153187697
+I1130 21:13:02.961765 137274321021824 utils.py:1231] [40950] train/loss = 2.4924383759498596
+I1130 21:13:02.961874 137274321021824 utils.py:1231] [40950] l2_grads = 1.381649374961853
+I1130 21:13:02.961968 137274321021824 utils.py:1231] [40950] lr = 0.0007918067211068578
+I1130 21:13:02.962022 137274321021824 utils.py:1231] [40950] uptime = 256972.32438434596
+I1130 21:13:02.962071 137274321021824 utils.py:1231] [40950] examples_seen = 41932800.0
+I1130 21:13:02.962117 137274321021824 utils.py:1231] [40950] progress = 0.36366704261875793
+I1130 21:13:02.962163 137274321021824 utils.py:1231] [40950] epoch = 32.730159300075634
+I1130 21:13:02.962210 137274321021824 utils.py:1231] [40950] img/sec/core = 164.2116734747977
+I1130 21:13:02.962263 137274321021824 utils.py:1231] [40950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.34688361551721
+I1130 21:13:02.962310 137274321021824 utils.py:1231] [40950] core_hours = 71.34688361551721
+I1130 21:13:02.962365 137274321021824 train.py:125] NOTE: Steps:40950/112603 [36.4%]
+Walltime:2d23h22m (0s eval)
+ETA:5d4h50m
+Total train time:8d4h11m
+I1130 21:18:14.738527 137274321021824 utils.py:1231] [41000] l2_params = 323.5673794559007
+I1130 21:18:14.738773 137274321021824 utils.py:1231] [41000] train/loss = 2.555657923221588
+I1130 21:18:14.738868 137274321021824 utils.py:1231] [41000] l2_grads = 1.3436036109924316
+I1130 21:18:14.738933 137274321021824 utils.py:1231] [41000] lr = 0.0007911847911809455
+I1130 21:18:14.738985 137274321021824 utils.py:1231] [41000] uptime = 257284.10134648898
+I1130 21:18:14.739037 137274321021824 utils.py:1231] [41000] examples_seen = 41984000.0
+I1130 21:18:14.739086 137274321021824 utils.py:1231] [41000] progress = 0.36411108052183333
+I1130 21:18:14.739135 137274321021824 utils.py:1231] [41000] epoch = 32.77012286454459
+I1130 21:18:14.739187 137274321021824 utils.py:1231] [41000] img/sec/core = 164.2199591915752
+I1130 21:18:14.739247 137274321021824 utils.py:1231] [41000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.4334883272236
+I1130 21:18:14.739297 137274321021824 utils.py:1231] [41000] core_hours = 71.4334883272236
+I1130 21:18:14.739358 137274321021824 train.py:125] NOTE: Steps:41000/112603 [36.4%]
+Walltime:2d23h28m (0s eval)
+ETA:5d4h45m
+Total train time:8d4h11m
+I1130 21:23:26.890311 137274321021824 utils.py:1231] [41050] l2_params = 323.49542682184887
+I1130 21:23:26.890548 137274321021824 utils.py:1231] [41050] train/loss = 2.8699378669261932
+I1130 21:23:26.890678 137274321021824 utils.py:1231] [41050] l2_grads = 1.4381476640701294
+I1130 21:23:26.890766 137274321021824 utils.py:1231] [41050] lr = 0.00079056217877769
+I1130 21:23:26.890841 137274321021824 utils.py:1231] [41050] uptime = 257596.253202788
+I1130 21:23:26.890919 137274321021824 utils.py:1231] [41050] examples_seen = 42035200.0
+I1130 21:23:26.890978 137274321021824 utils.py:1231] [41050] progress = 0.36455511842490873
+I1130 21:23:26.891033 137274321021824 utils.py:1231] [41050] epoch = 32.810086429013545
+I1130 21:23:26.891089 137274321021824 utils.py:1231] [41050] img/sec/core = 164.02273113812032
+I1130 21:23:26.891151 137274321021824 utils.py:1231] [41050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.52019717619555
+I1130 21:23:26.891203 137274321021824 utils.py:1231] [41050] core_hours = 71.52019717619555
+I1130 21:23:26.891263 137274321021824 train.py:125] NOTE: Steps:41050/112603 [36.5%]
+Walltime:2d23h33m (0s eval)
+ETA:5d4h40m
+Total train time:8d4h11m
+I1130 21:28:39.319848 137274321021824 utils.py:1231] [41100] l2_params = 323.39922098534174
+I1130 21:28:39.320238 137274321021824 utils.py:1231] [41100] train/loss = 3.0608028769493103
+I1130 21:28:39.320461 137274321021824 utils.py:1231] [41100] l2_grads = 1.2329655885696411
+I1130 21:28:39.320542 137274321021824 utils.py:1231] [41100] lr = 0.0007899388853563677
+I1130 21:28:39.320614 137274321021824 utils.py:1231] [41100] uptime = 257908.682975028
+I1130 21:28:39.320728 137274321021824 utils.py:1231] [41100] examples_seen = 42086400.0
+I1130 21:28:39.320826 137274321021824 utils.py:1231] [41100] progress = 0.36499915632798413
+I1130 21:28:39.320936 137274321021824 utils.py:1231] [41100] epoch = 32.85004999348251
+I1130 21:28:39.321012 137274321021824 utils.py:1231] [41100] img/sec/core = 163.87682784811767
+I1130 21:28:39.321094 137274321021824 utils.py:1231] [41100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.60698322404001
+I1130 21:28:39.321187 137274321021824 utils.py:1231] [41100] core_hours = 71.60698322404001
+I1130 21:28:39.321260 137274321021824 train.py:125] NOTE: Steps:41100/112603 [36.5%]
+Walltime:2d23h38m (0s eval)
+ETA:5d4h34m
+Total train time:8d4h11m
+I1130 21:34:21.672501 137274321021824 utils.py:1231] [41150] l2_params = 323.3539473370039
+I1130 21:34:21.673442 137274321021824 utils.py:1231] [41150] train/loss = 5.0127280950546265
+I1130 21:34:21.673613 137274321021824 utils.py:1231] [41150] l2_grads = 1.1808161735534668
+I1130 21:34:21.673697 137274321021824 utils.py:1231] [41150] lr = 0.000789314912377849
+I1130 21:34:21.673750 137274321021824 utils.py:1231] [41150] uptime = 258251.03611255198
+I1130 21:34:21.673802 137274321021824 utils.py:1231] [41150] examples_seen = 42137600.0
+I1130 21:34:21.673851 137274321021824 utils.py:1231] [41150] progress = 0.3654431942310596
+I1130 21:34:21.673914 137274321021824 utils.py:1231] [41150] epoch = 32.89001355795146
+I1130 21:34:21.673971 137274321021824 utils.py:1231] [41150] img/sec/core = 149.553178832524
+I1130 21:34:21.674132 137274321021824 utils.py:1231] [41150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.70208131779665
+I1130 21:34:21.674193 137274321021824 utils.py:1231] [41150] core_hours = 71.70208131779665
+I1130 21:34:21.674259 137274321021824 train.py:125] NOTE: Steps:41150/112603 [36.5%]
+Walltime:2d23h44m (0s eval)
+ETA:5d4h30m
+Total train time:8d4h12m
+I1130 21:39:58.212131 137274321021824 utils.py:1231] [41200] l2_params = 323.2987908057732
+I1130 21:39:58.212515 137274321021824 utils.py:1231] [41200] train/loss = 2.5929504930973053
+I1130 21:39:58.212710 137274321021824 utils.py:1231] [41200] l2_grads = 1.3527270555496216
+I1130 21:39:58.212801 137274321021824 utils.py:1231] [41200] lr = 0.0007886902613045985
+I1130 21:39:58.212871 137274321021824 utils.py:1231] [41200] uptime = 258587.57523146598
+I1130 21:39:58.212941 137274321021824 utils.py:1231] [41200] examples_seen = 42188800.0
+I1130 21:39:58.213001 137274321021824 utils.py:1231] [41200] progress = 0.365887232134135
+I1130 21:39:58.213059 137274321021824 utils.py:1231] [41200] epoch = 32.92997712242042
+I1130 21:39:58.213120 137274321021824 utils.py:1231] [41200] img/sec/core = 152.1368456814823
+I1130 21:39:58.213186 137274321021824 utils.py:1231] [41200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.79556440638387
+I1130 21:39:58.213244 137274321021824 utils.py:1231] [41200] core_hours = 71.79556440638387
+I1130 21:39:58.213316 137274321021824 train.py:125] NOTE: Steps:41200/112603 [36.6%]
+Walltime:2d23h49m (0s eval)
+ETA:5d4h26m
+Total train time:8d4h13m
+I1130 21:45:35.328769 137274321021824 utils.py:1231] [41250] l2_params = 323.2550191840868
+I1130 21:45:35.329057 137274321021824 utils.py:1231] [41250] train/loss = 2.878406673669815
+I1130 21:45:35.329222 137274321021824 utils.py:1231] [41250] l2_grads = 1.2917447090148926
+I1130 21:45:35.329298 137274321021824 utils.py:1231] [41250] lr = 0.00078806493360067
+I1130 21:45:35.329362 137274321021824 utils.py:1231] [41250] uptime = 258924.69172230497
+I1130 21:45:35.329427 137274321021824 utils.py:1231] [41250] examples_seen = 42240000.0
+I1130 21:45:35.329486 137274321021824 utils.py:1231] [41250] progress = 0.3663312700372104
+I1130 21:45:35.329545 137274321021824 utils.py:1231] [41250] epoch = 32.96994068688937
+I1130 21:45:35.329605 137274321021824 utils.py:1231] [41250] img/sec/core = 151.87628428552148
+I1130 21:45:35.329671 137274321021824 utils.py:1231] [41250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.88920787606138
+I1130 21:45:35.329729 137274321021824 utils.py:1231] [41250] core_hours = 71.88920787606138
+I1130 21:45:35.329799 137274321021824 train.py:125] NOTE: Steps:41250/112603 [36.6%]
+Walltime:2d23h55m (0s eval)
+ETA:5d4h21m
+Total train time:8d4h15m
+I1130 21:51:29.238440 137274321021824 utils.py:1231] [41300] l2_params = 323.19202234425313
+I1130 21:51:29.238758 137274321021824 utils.py:1231] [41300] train/loss = 2.5000751614570618
+I1130 21:51:29.238940 137274321021824 utils.py:1231] [41300] l2_grads = 1.4890730381011963
+I1130 21:51:29.239045 137274321021824 utils.py:1231] [41300] lr = 0.0007874389307317041
+I1130 21:51:29.239109 137274321021824 utils.py:1231] [41300] uptime = 259278.60146942397
+I1130 21:51:29.239170 137274321021824 utils.py:1231] [41300] examples_seen = 42291200.0
+I1130 21:51:29.239228 137274321021824 utils.py:1231] [41300] progress = 0.3667753079402858
+I1130 21:51:29.239285 137274321021824 utils.py:1231] [41300] epoch = 33.00990425135833
+I1130 21:51:29.239346 137274321021824 utils.py:1231] [41300] img/sec/core = 144.66965212682828
+I1130 21:51:29.239411 137274321021824 utils.py:1231] [41300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 71.98751613915
+I1130 21:51:29.239470 137274321021824 utils.py:1231] [41300] core_hours = 71.98751613915
+I1130 21:51:29.239539 137274321021824 train.py:125] NOTE: Steps:41300/112603 [36.7%]
+Walltime:3d0h1m (0s eval)
+ETA:5d4h17m
+Total train time:8d4h16m
+I1130 21:57:13.160166 137274321021824 utils.py:1231] [41350] l2_params = 323.1355785527158
+I1130 21:57:13.161604 137274321021824 utils.py:1231] [41350] train/loss = 4.040723264217377
+I1130 21:57:13.161898 137274321021824 utils.py:1231] [41350] l2_grads = 1.1901354789733887
+I1130 21:57:13.161979 137274321021824 utils.py:1231] [41350] lr = 0.0007868122541649217
+I1130 21:57:13.162040 137274321021824 utils.py:1231] [41350] uptime = 259622.52440104098
+I1130 21:57:13.162096 137274321021824 utils.py:1231] [41350] examples_seen = 42342400.0
+I1130 21:57:13.162147 137274321021824 utils.py:1231] [41350] progress = 0.3672193458433612
+I1130 21:57:13.162197 137274321021824 utils.py:1231] [41350] epoch = 33.04986781582729
+I1130 21:57:13.162250 137274321021824 utils.py:1231] [41350] img/sec/core = 148.87056166704235
+I1130 21:57:13.162433 137274321021824 utils.py:1231] [41350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.08305028682138
+I1130 21:57:13.162504 137274321021824 utils.py:1231] [41350] core_hours = 72.08305028682138
+I1130 21:57:13.162576 137274321021824 train.py:125] NOTE: Steps:41350/112603 [36.7%]
+Walltime:3d0h7m (0s eval)
+ETA:5d4h13m
+Total train time:8d4h18m
+I1130 22:02:51.249326 137274321021824 utils.py:1231] [41400] l2_params = 323.0688027638653
+I1130 22:02:51.250695 137274321021824 utils.py:1231] [41400] train/loss = 2.522010251879692
+I1130 22:02:51.251114 137274321021824 utils.py:1231] [41400] l2_grads = 1.3242720365524292
+I1130 22:02:51.251258 137274321021824 utils.py:1231] [41400] lr = 0.000786184905369125
+I1130 22:02:51.251352 137274321021824 utils.py:1231] [41400] uptime = 259960.61370975798
+I1130 22:02:51.251432 137274321021824 utils.py:1231] [41400] examples_seen = 42393600.0
+I1130 22:02:51.251516 137274321021824 utils.py:1231] [41400] progress = 0.3676633837464366
+I1130 22:02:51.251595 137274321021824 utils.py:1231] [41400] epoch = 33.089831380296246
+I1130 22:02:51.251678 137274321021824 utils.py:1231] [41400] img/sec/core = 151.43927559938834
+I1130 22:02:51.251769 137274321021824 utils.py:1231] [41400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.17696398368722
+I1130 22:02:51.251837 137274321021824 utils.py:1231] [41400] core_hours = 72.17696398368722
+I1130 22:02:51.251933 137274321021824 train.py:125] NOTE: Steps:41400/112603 [36.8%]
+Walltime:3d0h12m (0s eval)
+ETA:5d4h8m
+Total train time:8d4h19m
+I1130 22:09:36.419321 137274321021824 utils.py:1231] [41450] l2_params = 323.0073029877979
+I1130 22:09:36.420495 137274321021824 utils.py:1231] [41450] train/loss = 2.5188015401363373
+I1130 22:09:36.420709 137274321021824 utils.py:1231] [41450] l2_grads = 1.4783998727798462
+I1130 22:09:36.420785 137274321021824 utils.py:1231] [41450] lr = 0.0007855568858146908
+I1130 22:09:36.420846 137274321021824 utils.py:1231] [41450] uptime = 260365.78320783097
+I1130 22:09:36.420927 137274321021824 utils.py:1231] [41450] examples_seen = 42444800.0
+I1130 22:09:36.420997 137274321021824 utils.py:1231] [41450] progress = 0.368107421649512
+I1130 22:09:36.421063 137274321021824 utils.py:1231] [41450] epoch = 33.1297949447652
+I1130 22:09:36.421141 137274321021824 utils.py:1231] [41450] img/sec/core = 126.36686681378914
+I1130 22:09:36.421311 137274321021824 utils.py:1231] [41450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.28951106648528
+I1130 22:09:36.421394 137274321021824 utils.py:1231] [41450] core_hours = 72.28951106648528
+I1130 22:09:36.421467 137274321021824 train.py:125] NOTE: Steps:41450/112603 [36.8%]
+Walltime:3d0h19m (0s eval)
+ETA:5d4h5m
+Total train time:8d4h23m
+I1130 22:15:32.970607 137274321021824 utils.py:1231] [41500] l2_params = 322.9415258505861
+I1130 22:15:32.971741 137274321021824 utils.py:1231] [41500] train/loss = 2.485159397125244
+I1130 22:15:32.971944 137274321021824 utils.py:1231] [41500] l2_grads = 1.4529696702957153
+I1130 22:15:32.972000 137274321021824 utils.py:1231] [41500] lr = 0.0007849281969735675
+I1130 22:15:32.972047 137274321021824 utils.py:1231] [41500] uptime = 260722.33440983196
+I1130 22:15:32.972101 137274321021824 utils.py:1231] [41500] examples_seen = 42496000.0
+I1130 22:15:32.972149 137274321021824 utils.py:1231] [41500] progress = 0.3685514595525874
+I1130 22:15:32.972196 137274321021824 utils.py:1231] [41500] epoch = 33.16975850923416
+I1130 22:15:32.972245 137274321021824 utils.py:1231] [41500] img/sec/core = 143.59788920262164
+I1130 22:15:32.972388 137274321021824 utils.py:1231] [41500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.3885530670411
+I1130 22:15:32.972440 137274321021824 utils.py:1231] [41500] core_hours = 72.3885530670411
+I1130 22:15:32.972505 137274321021824 train.py:125] NOTE: Steps:41500/112603 [36.9%]
+Walltime:3d0h25m (0s eval)
+ETA:5d4h1m
+Total train time:8d4h25m
+I1130 22:21:19.335829 137274321021824 utils.py:1231] [41550] l2_params = 322.90360169786186
+I1130 22:21:19.336074 137274321021824 utils.py:1231] [41550] train/loss = 2.4978725910186768
+I1130 22:21:19.336194 137274321021824 utils.py:1231] [41550] l2_grads = 1.3759267330169678
+I1130 22:21:19.336291 137274321021824 utils.py:1231] [41550] lr = 0.0007842988403192721
+I1130 22:21:19.336357 137274321021824 utils.py:1231] [41550] uptime = 261068.698717842
+I1130 22:21:19.336435 137274321021824 utils.py:1231] [41550] examples_seen = 42547200.0
+I1130 22:21:19.336498 137274321021824 utils.py:1231] [41550] progress = 0.3689954974556628
+I1130 22:21:19.336555 137274321021824 utils.py:1231] [41550] epoch = 33.20972207370312
+I1130 22:21:19.336616 137274321021824 utils.py:1231] [41550] img/sec/core = 147.82123566414742
+I1130 22:21:19.336699 137274321021824 utils.py:1231] [41550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.48476537482166
+I1130 22:21:19.336756 137274321021824 utils.py:1231] [41550] core_hours = 72.48476537482166
+I1130 22:21:19.336843 137274321021824 train.py:125] NOTE: Steps:41550/112603 [36.9%]
+Walltime:3d0h31m (0s eval)
+ETA:5d3h57m
+Total train time:8d4h26m
+I1130 22:27:28.348886 137274321021824 utils.py:1231] [41600] l2_params = 322.88517604174086
+I1130 22:27:28.349143 137274321021824 utils.py:1231] [41600] train/loss = 2.8273876905441284
+I1130 22:27:28.349292 137274321021824 utils.py:1231] [41600] l2_grads = 1.3894721269607544
+I1130 22:27:28.349367 137274321021824 utils.py:1231] [41600] lr = 0.0007836688173268882
+I1130 22:27:28.349433 137274321021824 utils.py:1231] [41600] uptime = 261437.71179116197
+I1130 22:27:28.349500 137274321021824 utils.py:1231] [41600] examples_seen = 42598400.0
+I1130 22:27:28.349557 137274321021824 utils.py:1231] [41600] progress = 0.36943953535873825
+I1130 22:27:28.349615 137274321021824 utils.py:1231] [41600] epoch = 33.249685638172075
+I1130 22:27:28.349674 137274321021824 utils.py:1231] [41600] img/sec/core = 138.74847180713223
+I1130 22:27:28.349744 137274321021824 utils.py:1231] [41600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.58726900629944
+I1130 22:27:28.349802 137274321021824 utils.py:1231] [41600] core_hours = 72.58726900629944
+I1130 22:27:28.349869 137274321021824 train.py:125] NOTE: Steps:41600/112603 [36.9%]
+Walltime:3d0h37m (0s eval)
+ETA:5d3h53m
+Total train time:8d4h29m
+I1130 22:33:17.065892 137274321021824 utils.py:1231] [41650] l2_params = 322.82175560937
+I1130 22:33:17.066219 137274321021824 utils.py:1231] [41650] train/loss = 2.506780982017517
+I1130 22:33:17.066396 137274321021824 utils.py:1231] [41650] l2_grads = 1.4682403802871704
+I1130 22:33:17.066453 137274321021824 utils.py:1231] [41650] lr = 0.0007830381294730596
+I1130 22:33:17.066500 137274321021824 utils.py:1231] [41650] uptime = 261786.42886354297
+I1130 22:33:17.066548 137274321021824 utils.py:1231] [41650] examples_seen = 42649600.0
+I1130 22:33:17.066592 137274321021824 utils.py:1231] [41650] progress = 0.36988357326181365
+I1130 22:33:17.066635 137274321021824 utils.py:1231] [41650] epoch = 33.28964920264103
+I1130 22:33:17.066682 137274321021824 utils.py:1231] [41650] img/sec/core = 146.8238983839037
+I1130 22:33:17.066735 137274321021824 utils.py:1231] [41650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.6841348597386
+I1130 22:33:17.066780 137274321021824 utils.py:1231] [41650] core_hours = 72.6841348597386
+I1130 22:33:17.066836 137274321021824 train.py:125] NOTE: Steps:41650/112603 [37.0%]
+Walltime:3d0h43m (0s eval)
+ETA:5d3h49m
+Total train time:8d4h30m
+I1130 22:38:49.369647 137274321021824 utils.py:1231] [41700] l2_params = 322.7670031347907
+I1130 22:38:49.369867 137274321021824 utils.py:1231] [41700] train/loss = 2.479904919862747
+I1130 22:38:49.369974 137274321021824 utils.py:1231] [41700] l2_grads = 1.3626452684402466
+I1130 22:38:49.370059 137274321021824 utils.py:1231] [41700] lr = 0.0007824067782359899
+I1130 22:38:49.370118 137274321021824 utils.py:1231] [41700] uptime = 262118.73247989797
+I1130 22:38:49.370177 137274321021824 utils.py:1231] [41700] examples_seen = 42700800.0
+I1130 22:38:49.370234 137274321021824 utils.py:1231] [41700] progress = 0.37032761116488905
+I1130 22:38:49.370293 137274321021824 utils.py:1231] [41700] epoch = 33.329612767109985
+I1130 22:38:49.370351 137274321021824 utils.py:1231] [41700] img/sec/core = 154.07596390796795
+I1130 22:38:49.370412 137274321021824 utils.py:1231] [41700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.77644141983721
+I1130 22:38:49.370467 137274321021824 utils.py:1231] [41700] core_hours = 72.77644141983721
+I1130 22:38:49.370538 137274321021824 train.py:125] NOTE: Steps:41700/112603 [37.0%]
+Walltime:3d0h48m (0s eval)
+ETA:5d3h44m
+Total train time:8d4h31m
+I1130 22:44:07.503846 137274321021824 utils.py:1231] [41750] l2_params = 322.7115126560418
+I1130 22:44:07.504074 137274321021824 utils.py:1231] [41750] train/loss = 5.092734336853027
+I1130 22:44:07.504192 137274321021824 utils.py:1231] [41750] l2_grads = 1.2691729068756104
+I1130 22:44:07.504267 137274321021824 utils.py:1231] [41750] lr = 0.0007817747650954357
+I1130 22:44:07.504330 137274321021824 utils.py:1231] [41750] uptime = 262436.866690971
+I1130 22:44:07.504393 137274321021824 utils.py:1231] [41750] examples_seen = 42752000.0
+I1130 22:44:07.504461 137274321021824 utils.py:1231] [41750] progress = 0.37077164906796445
+I1130 22:44:07.504520 137274321021824 utils.py:1231] [41750] epoch = 33.36957633157894
+I1130 22:44:07.504599 137274321021824 utils.py:1231] [41750] img/sec/core = 160.9383656894538
+I1130 22:44:07.504681 137274321021824 utils.py:1231] [41750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.86481203402417
+I1130 22:44:07.504746 137274321021824 utils.py:1231] [41750] core_hours = 72.86481203402417
+I1130 22:44:07.504818 137274321021824 train.py:125] NOTE: Steps:41750/112603 [37.1%]
+Walltime:3d0h53m (0s eval)
+ETA:5d3h39m
+Total train time:8d4h31m
+I1130 22:49:19.335092 137274321021824 utils.py:1231] [41800] l2_params = 322.6666704876343
+I1130 22:49:19.335374 137274321021824 utils.py:1231] [41800] train/loss = 5.066150367259979
+I1130 22:49:19.335559 137274321021824 utils.py:1231] [41800] l2_grads = 1.2275876998901367
+I1130 22:49:19.335639 137274321021824 utils.py:1231] [41800] lr = 0.0007811420915327065
+I1130 22:49:19.335701 137274321021824 utils.py:1231] [41800] uptime = 262748.698062712
+I1130 22:49:19.335768 137274321021824 utils.py:1231] [41800] examples_seen = 42803200.0
+I1130 22:49:19.335826 137274321021824 utils.py:1231] [41800] progress = 0.37121568697103985
+I1130 22:49:19.335889 137274321021824 utils.py:1231] [41800] epoch = 33.4095398960479
+I1130 22:49:19.335949 137274321021824 utils.py:1231] [41800] img/sec/core = 164.19130542943512
+I1130 22:49:19.336015 137274321021824 utils.py:1231] [41800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 72.95143185950776
+I1130 22:49:19.336072 137274321021824 utils.py:1231] [41800] core_hours = 72.95143185950776
+I1130 22:49:19.336142 137274321021824 train.py:125] NOTE: Steps:41800/112603 [37.1%]
+Walltime:3d0h59m (0s eval)
+ETA:5d3h34m
+Total train time:8d4h31m
+I1130 22:54:31.212429 137274321021824 utils.py:1231] [41850] l2_params = 322.5643746494891
+I1130 22:54:31.212661 137274321021824 utils.py:1231] [41850] train/loss = 5.016940414905548
+I1130 22:54:31.212768 137274321021824 utils.py:1231] [41850] l2_grads = 1.2120388746261597
+I1130 22:54:31.212842 137274321021824 utils.py:1231] [41850] lr = 0.0007805087590306595
+I1130 22:54:31.212944 137274321021824 utils.py:1231] [41850] uptime = 263060.57530062797
+I1130 22:54:31.213012 137274321021824 utils.py:1231] [41850] examples_seen = 42854400.0
+I1130 22:54:31.213072 137274321021824 utils.py:1231] [41850] progress = 0.37165972487411525
+I1130 22:54:31.213129 137274321021824 utils.py:1231] [41850] epoch = 33.44950346051686
+I1130 22:54:31.213190 137274321021824 utils.py:1231] [41850] img/sec/core = 164.16715866193235
+I1130 22:54:31.213255 137274321021824 utils.py:1231] [41850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.03806442559555
+I1130 22:54:31.213313 137274321021824 utils.py:1231] [41850] core_hours = 73.03806442559555
+I1130 22:54:31.213379 137274321021824 train.py:125] NOTE: Steps:41850/112603 [37.2%]
+Walltime:3d1h4m (0s eval)
+ETA:5d3h29m
+Total train time:8d4h31m
+I1130 22:59:43.100945 137274321021824 utils.py:1231] [41900] l2_params = 322.53010597549104
+I1130 22:59:43.101189 137274321021824 utils.py:1231] [41900] train/loss = 2.5735904574394226
+I1130 22:59:43.101324 137274321021824 utils.py:1231] [41900] l2_grads = 1.502725601196289
+I1130 22:59:43.101431 137274321021824 utils.py:1231] [41900] lr = 0.0007798747690736955
+I1130 22:59:43.101505 137274321021824 utils.py:1231] [41900] uptime = 263372.463860816
+I1130 22:59:43.101582 137274321021824 utils.py:1231] [41900] examples_seen = 42905600.0
+I1130 22:59:43.101643 137274321021824 utils.py:1231] [41900] progress = 0.37210376277719065
+I1130 22:59:43.101698 137274321021824 utils.py:1231] [41900] epoch = 33.489467024985814
+I1130 22:59:43.101751 137274321021824 utils.py:1231] [41900] img/sec/core = 164.16119901652442
+I1130 22:59:43.101809 137274321021824 utils.py:1231] [41900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.12470013675889
+I1130 22:59:43.101874 137274321021824 utils.py:1231] [41900] core_hours = 73.12470013675889
+I1130 22:59:43.101958 137274321021824 train.py:125] NOTE: Steps:41900/112603 [37.2%]
+Walltime:3d1h9m (0s eval)
+ETA:5d3h23m
+Total train time:8d4h31m
+I1130 23:04:54.963669 137274321021824 utils.py:1231] [41950] l2_params = 322.4903911694044
+I1130 23:04:54.963960 137274321021824 utils.py:1231] [41950] train/loss = 2.5966721177101135
+I1130 23:04:54.964143 137274321021824 utils.py:1231] [41950] l2_grads = 1.418729543685913
+I1130 23:04:54.964231 137274321021824 utils.py:1231] [41950] lr = 0.0007792401231477576
+I1130 23:04:54.964294 137274321021824 utils.py:1231] [41950] uptime = 263684.32665640296
+I1130 23:04:54.964348 137274321021824 utils.py:1231] [41950] examples_seen = 42956800.0
+I1130 23:04:54.964398 137274321021824 utils.py:1231] [41950] progress = 0.37254780068026605
+I1130 23:04:54.964447 137274321021824 utils.py:1231] [41950] epoch = 33.52943058945477
+I1130 23:04:54.964500 137274321021824 utils.py:1231] [41950] img/sec/core = 164.17476122354145
+I1130 23:04:54.964557 137274321021824 utils.py:1231] [41950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.2113286910886
+I1130 23:04:54.964611 137274321021824 utils.py:1231] [41950] core_hours = 73.2113286910886
+I1130 23:04:54.964689 137274321021824 train.py:125] NOTE: Steps:41950/112603 [37.3%]
+Walltime:3d1h14m (0s eval)
+ETA:5d3h18m
+Total train time:8d4h31m
+I1130 23:10:06.796637 137274321021824 utils.py:1231] [42000] l2_params = 322.436632420289
+I1130 23:10:06.796915 137274321021824 utils.py:1231] [42000] train/loss = 4.498368203639984
+I1130 23:10:06.797087 137274321021824 utils.py:1231] [42000] l2_grads = 1.1299827098846436
+I1130 23:10:06.797194 137274321021824 utils.py:1231] [42000] lr = 0.0007786048227403243
+I1130 23:10:06.797261 137274321021824 utils.py:1231] [42000] uptime = 263996.159622564
+I1130 23:10:06.797322 137274321021824 utils.py:1231] [42000] examples_seen = 43008000.0
+I1130 23:10:06.797378 137274321021824 utils.py:1231] [42000] progress = 0.37299183858334145
+I1130 23:10:06.797434 137274321021824 utils.py:1231] [42000] epoch = 33.569394153923724
+I1130 23:10:06.797490 137274321021824 utils.py:1231] [42000] img/sec/core = 164.19046590975972
+I1130 23:10:06.797550 137274321021824 utils.py:1231] [42000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.29794895946665
+I1130 23:10:06.797605 137274321021824 utils.py:1231] [42000] core_hours = 73.29794895946665
+I1130 23:10:06.797671 137274321021824 train.py:125] NOTE: Steps:42000/112603 [37.3%]
+Walltime:3d1h19m (0s eval)
+ETA:5d3h13m
+Total train time:8d4h31m
+I1130 23:15:19.028589 137274321021824 utils.py:1231] [42050] l2_params = 322.36305106445155
+I1130 23:15:19.028836 137274321021824 utils.py:1231] [42050] train/loss = 2.804260343313217
+I1130 23:15:19.028959 137274321021824 utils.py:1231] [42050] l2_grads = 1.3870586156845093
+I1130 23:15:19.029034 137274321021824 utils.py:1231] [42050] lr = 0.0007779688693404092
+I1130 23:15:19.029096 137274321021824 utils.py:1231] [42050] uptime = 264308.391457345
+I1130 23:15:19.029160 137274321021824 utils.py:1231] [42050] examples_seen = 43059200.0
+I1130 23:15:19.029231 137274321021824 utils.py:1231] [42050] progress = 0.3734358764864169
+I1130 23:15:19.029298 137274321021824 utils.py:1231] [42050] epoch = 33.60935771839269
+I1130 23:15:19.029364 137274321021824 utils.py:1231] [42050] img/sec/core = 163.9807165592536
+I1130 23:15:19.029439 137274321021824 utils.py:1231] [42050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.3846800246836
+I1130 23:15:19.029504 137274321021824 utils.py:1231] [42050] core_hours = 73.3846800246836
+I1130 23:15:19.029575 137274321021824 train.py:125] NOTE: Steps:42050/112603 [37.3%]
+Walltime:3d1h25m (0s eval)
+ETA:5d3h7m
+Total train time:8d4h31m
+I1130 23:20:30.852006 137274321021824 utils.py:1231] [42100] l2_params = 322.3183701072006
+I1130 23:20:30.852264 137274321021824 utils.py:1231] [42100] train/loss = 4.190802246332169
+I1130 23:20:30.852369 137274321021824 utils.py:1231] [42100] l2_grads = 1.1799418926239014
+I1130 23:20:30.852440 137274321021824 utils.py:1231] [42100] lr = 0.0007773322644385577
+I1130 23:20:30.852502 137274321021824 utils.py:1231] [42100] uptime = 264620.21486319497
+I1130 23:20:30.852565 137274321021824 utils.py:1231] [42100] examples_seen = 43110400.0
+I1130 23:20:30.852623 137274321021824 utils.py:1231] [42100] progress = 0.3738799143894923
+I1130 23:20:30.852688 137274321021824 utils.py:1231] [42100] epoch = 33.64932128286164
+I1130 23:20:30.852750 137274321021824 utils.py:1231] [42100] img/sec/core = 164.19549988698597
+I1130 23:20:30.852814 137274321021824 utils.py:1231] [42100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.47129763741971
+I1130 23:20:30.852873 137274321021824 utils.py:1231] [42100] core_hours = 73.47129763741971
+I1130 23:20:30.852951 137274321021824 train.py:125] NOTE: Steps:42100/112603 [37.4%]
+Walltime:3d1h30m (0s eval)
+ETA:5d3h2m
+Total train time:8d4h31m
+I1130 23:25:42.702077 137274321021824 utils.py:1231] [42150] l2_params = 322.25648321116034
+I1130 23:25:42.702291 137274321021824 utils.py:1231] [42150] train/loss = 2.5993431210517883
+I1130 23:25:42.702381 137274321021824 utils.py:1231] [42150] l2_grads = 1.426995038986206
+I1130 23:25:42.702440 137274321021824 utils.py:1231] [42150] lr = 0.0007766950095268406
+I1130 23:25:42.702513 137274321021824 utils.py:1231] [42150] uptime = 264932.06486185297
+I1130 23:25:42.702570 137274321021824 utils.py:1231] [42150] examples_seen = 43161600.0
+I1130 23:25:42.702618 137274321021824 utils.py:1231] [42150] progress = 0.3743239522925677
+I1130 23:25:42.702667 137274321021824 utils.py:1231] [42150] epoch = 33.6892848473306
+I1130 23:25:42.702718 137274321021824 utils.py:1231] [42150] img/sec/core = 164.18149822136192
+I1130 23:25:42.702772 137274321021824 utils.py:1231] [42150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.55792263704693
+I1130 23:25:42.702822 137274321021824 utils.py:1231] [42150] core_hours = 73.55792263704693
+I1130 23:25:42.702885 137274321021824 train.py:125] NOTE: Steps:42150/112603 [37.4%]
+Walltime:3d1h35m (0s eval)
+ETA:5d2h57m
+Total train time:8d4h31m
+I1130 23:30:54.533370 137274321021824 utils.py:1231] [42200] l2_params = 322.18380612717266
+I1130 23:30:54.533581 137274321021824 utils.py:1231] [42200] train/loss = 2.6641100347042084
+I1130 23:30:54.533674 137274321021824 utils.py:1231] [42200] l2_grads = 1.4656028747558594
+I1130 23:30:54.533732 137274321021824 utils.py:1231] [42200] lr = 0.0007760571060988518
+I1130 23:30:54.533784 137274321021824 utils.py:1231] [42200] uptime = 265243.896145196
+I1130 23:30:54.533835 137274321021824 utils.py:1231] [42200] examples_seen = 43212800.0
+I1130 23:30:54.533887 137274321021824 utils.py:1231] [42200] progress = 0.3747679901956431
+I1130 23:30:54.533936 137274321021824 utils.py:1231] [42200] epoch = 33.72924841179955
+I1130 23:30:54.533985 137274321021824 utils.py:1231] [42200] img/sec/core = 164.1913519743891
+I1130 23:30:54.534039 137274321021824 utils.py:1231] [42200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.64454243797556
+I1130 23:30:54.534087 137274321021824 utils.py:1231] [42200] core_hours = 73.64454243797556
+I1130 23:30:54.534146 137274321021824 train.py:125] NOTE: Steps:42200/112603 [37.5%]
+Walltime:3d1h40m (0s eval)
+ETA:5d2h52m
+Total train time:8d4h30m
+I1130 23:36:06.370606 137274321021824 utils.py:1231] [42250] l2_params = 322.12829763828785
+I1130 23:36:06.370856 137274321021824 utils.py:1231] [42250] train/loss = 4.55600243806839
+I1130 23:36:06.370987 137274321021824 utils.py:1231] [42250] l2_grads = 1.2145341634750366
+I1130 23:36:06.371101 137274321021824 utils.py:1231] [42250] lr = 0.0007754185556497067
+I1130 23:36:06.371179 137274321021824 utils.py:1231] [42250] uptime = 265555.73354005
+I1130 23:36:06.371264 137274321021824 utils.py:1231] [42250] examples_seen = 43264000.0
+I1130 23:36:06.371340 137274321021824 utils.py:1231] [42250] progress = 0.3752120280987185
+I1130 23:36:06.371409 137274321021824 utils.py:1231] [42250] epoch = 33.76921197626851
+I1130 23:36:06.371469 137274321021824 utils.py:1231] [42250] img/sec/core = 164.188134088191
+I1130 23:36:06.371533 137274321021824 utils.py:1231] [42250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.73116393654611
+I1130 23:36:06.371594 137274321021824 utils.py:1231] [42250] core_hours = 73.73116393654611
+I1130 23:36:06.371665 137274321021824 train.py:125] NOTE: Steps:42250/112603 [37.5%]
+Walltime:3d1h45m (0s eval)
+ETA:5d2h46m
+Total train time:8d4h30m
+I1130 23:41:18.428019 137274321021824 utils.py:1231] [42300] l2_params = 322.048711289178
+I1130 23:41:18.428232 137274321021824 utils.py:1231] [42300] train/loss = 3.1750177145004272
+I1130 23:41:18.428336 137274321021824 utils.py:1231] [42300] l2_grads = 1.290504813194275
+I1130 23:41:18.428407 137274321021824 utils.py:1231] [42300] lr = 0.0007747793596760361
+I1130 23:41:18.428492 137274321021824 utils.py:1231] [42300] uptime = 265867.79084823997
+I1130 23:41:18.428574 137274321021824 utils.py:1231] [42300] examples_seen = 43315200.0
+I1130 23:41:18.428634 137274321021824 utils.py:1231] [42300] progress = 0.3756560660017939
+I1130 23:41:18.428704 137274321021824 utils.py:1231] [42300] epoch = 33.80917554073747
+I1130 23:41:18.428762 137274321021824 utils.py:1231] [42300] img/sec/core = 164.07242726335954
+I1130 23:41:18.428824 137274321021824 utils.py:1231] [42300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.81784652215444
+I1130 23:41:18.428888 137274321021824 utils.py:1231] [42300] core_hours = 73.81784652215444
+I1130 23:41:18.428988 137274321021824 train.py:125] NOTE: Steps:42300/112603 [37.6%]
+Walltime:3d1h51m (0s eval)
+ETA:5d2h41m
+Total train time:8d4h30m
+I1130 23:46:30.257792 137274321021824 utils.py:1231] [42350] l2_params = 321.986182459839
+I1130 23:46:30.258072 137274321021824 utils.py:1231] [42350] train/loss = 4.7137563824653625
+I1130 23:46:30.258250 137274321021824 utils.py:1231] [42350] l2_grads = 1.3207790851593018
+I1130 23:46:30.258343 137274321021824 utils.py:1231] [42350] lr = 0.0007741395196759844
+I1130 23:46:30.258415 137274321021824 utils.py:1231] [42350] uptime = 266179.620773163
+I1130 23:46:30.258488 137274321021824 utils.py:1231] [42350] examples_seen = 43366400.0
+I1130 23:46:30.258552 137274321021824 utils.py:1231] [42350] progress = 0.3761001039048693
+I1130 23:46:30.258620 137274321021824 utils.py:1231] [42350] epoch = 33.849139105206426
+I1130 23:46:30.258689 137274321021824 utils.py:1231] [42350] img/sec/core = 164.1920672387031
+I1130 23:46:30.258766 137274321021824 utils.py:1231] [42350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.90446594574415
+I1130 23:46:30.258830 137274321021824 utils.py:1231] [42350] core_hours = 73.90446594574415
+I1130 23:46:30.258913 137274321021824 train.py:125] NOTE: Steps:42350/112603 [37.6%]
+Walltime:3d1h56m (0s eval)
+ETA:5d2h36m
+Total train time:8d4h30m
+I1130 23:51:42.070195 137274321021824 utils.py:1231] [42400] l2_params = 321.95450551755977
+I1130 23:51:42.070384 137274321021824 utils.py:1231] [42400] train/loss = 3.8080238699913025
+I1130 23:51:42.070480 137274321021824 utils.py:1231] [42400] l2_grads = 1.1610909700393677
+I1130 23:51:42.070541 137274321021824 utils.py:1231] [42400] lr = 0.0007734990371492046
+I1130 23:51:42.070590 137274321021824 utils.py:1231] [42400] uptime = 266491.432953008
+I1130 23:51:42.070641 137274321021824 utils.py:1231] [42400] examples_seen = 43417600.0
+I1130 23:51:42.070688 137274321021824 utils.py:1231] [42400] progress = 0.3765441418079447
+I1130 23:51:42.070739 137274321021824 utils.py:1231] [42400] epoch = 33.88910266967538
+I1130 23:51:42.070787 137274321021824 utils.py:1231] [42400] img/sec/core = 164.20141132861147
+I1130 23:51:42.070842 137274321021824 utils.py:1231] [42400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 73.99108044014555
+I1130 23:51:42.070895 137274321021824 utils.py:1231] [42400] core_hours = 73.99108044014555
+I1130 23:51:42.070954 137274321021824 train.py:125] NOTE: Steps:42400/112603 [37.7%]
+Walltime:3d2h1m (0s eval)
+ETA:5d2h30m
+Total train time:8d4h30m
+I1130 23:56:53.899490 137274321021824 utils.py:1231] [42450] l2_params = 321.9046020195293
+I1130 23:56:53.899698 137274321021824 utils.py:1231] [42450] train/loss = 4.597927808761597
+I1130 23:56:53.899792 137274321021824 utils.py:1231] [42450] l2_grads = 1.277267575263977
+I1130 23:56:53.899850 137274321021824 utils.py:1231] [42450] lr = 0.0007728579135968564
+I1130 23:56:53.899905 137274321021824 utils.py:1231] [42450] uptime = 266803.262266903
+I1130 23:56:53.899977 137274321021824 utils.py:1231] [42450] examples_seen = 43468800.0
+I1130 23:56:53.900029 137274321021824 utils.py:1231] [42450] progress = 0.3769881797110201
+I1130 23:56:53.900076 137274321021824 utils.py:1231] [42450] epoch = 33.92906623414434
+I1130 23:56:53.900126 137274321021824 utils.py:1231] [42450] img/sec/core = 164.19238897224437
+I1130 23:56:53.900181 137274321021824 utils.py:1231] [42450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.07769969400528
+I1130 23:56:53.900228 137274321021824 utils.py:1231] [42450] core_hours = 74.07769969400528
+I1130 23:56:53.900287 137274321021824 train.py:125] NOTE: Steps:42450/112603 [37.7%]
+Walltime:3d2h6m (0s eval)
+ETA:5d2h25m
+Total train time:8d4h30m
+I1201 00:02:05.743616 137274321021824 utils.py:1231] [42500] l2_params = 321.8271967001288
+I1201 00:02:05.743911 137274321021824 utils.py:1231] [42500] train/loss = 2.5246008336544037
+I1201 00:02:05.744057 137274321021824 utils.py:1231] [42500] l2_grads = 1.4289101362228394
+I1201 00:02:05.744151 137274321021824 utils.py:1231] [42500] lr = 0.0007722161505216015
+I1201 00:02:05.744228 137274321021824 utils.py:1231] [42500] uptime = 267115.106589745
+I1201 00:02:05.744288 137274321021824 utils.py:1231] [42500] examples_seen = 43520000.0
+I1201 00:02:05.744344 137274321021824 utils.py:1231] [42500] progress = 0.37743221761409557
+I1201 00:02:05.744400 137274321021824 utils.py:1231] [42500] epoch = 33.9690297986133
+I1201 00:02:05.744460 137274321021824 utils.py:1231] [42500] img/sec/core = 164.1844864559138
+I1201 00:02:05.744522 137274321021824 utils.py:1231] [42500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.16432311701693
+I1201 00:02:05.744576 137274321021824 utils.py:1231] [42500] core_hours = 74.16432311701693
+I1201 00:02:05.744641 137274321021824 train.py:125] NOTE: Steps:42500/112603 [37.7%]
+Walltime:3d2h11m (0s eval)
+ETA:5d2h20m
+Total train time:8d4h30m
+I1201 00:02:05.744745 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:42500/112603 [37.7%]
+Walltime:3d2h11m (0s eval)
+ETA:5d2h20m
+Total train time:8d4h30m
+I1201 00:03:45.619764 137274321021824 utils.py:1231] [42500] val/acc@1 = 0.6224689094387755
+I1201 00:03:45.620033 137274321021824 utils.py:1231] [42500] val/loss = 1.5828200739865401
+I1201 00:03:45.620219 137274321021824 utils.py:1231] [42500] z/secs/eval/val = 99.87539771496085
+I1201 00:03:45.620289 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 99.87539771496085
+I1201 00:08:57.390667 137274321021824 utils.py:1231] [42550] l2_params = 321.744021611277
+I1201 00:08:57.390908 137274321021824 utils.py:1231] [42550] train/loss = 2.7438822090625763
+I1201 00:08:57.391046 137274321021824 utils.py:1231] [42550] l2_grads = 1.4352706670761108
+I1201 00:08:57.391153 137274321021824 utils.py:1231] [42550] lr = 0.0007715737494276011
+I1201 00:08:57.391236 137274321021824 utils.py:1231] [42550] uptime = 267526.753594087
+I1201 00:08:57.391311 137274321021824 utils.py:1231] [42550] examples_seen = 43571200.0
+I1201 00:08:57.391377 137274321021824 utils.py:1231] [42550] progress = 0.37787625551717097
+I1201 00:08:57.391444 137274321021824 utils.py:1231] [42550] epoch = 34.008993363082254
+I1201 00:08:57.391508 137274321021824 utils.py:1231] [42550] img/sec/core = 124.3784102883034
+I1201 00:08:57.391574 137274321021824 utils.py:1231] [42550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.27866950711194
+I1201 00:08:57.391652 137274321021824 utils.py:1231] [42550] core_hours = 74.27866950711194
+I1201 00:08:57.391721 137274321021824 train.py:125] NOTE: Steps:42550/112603 [37.8%]
+Walltime:3d2h18m (0s eval)
+ETA:5d2h17m
+Total train time:8d4h34m
+I1201 00:14:09.209384 137274321021824 utils.py:1231] [42600] l2_params = 321.683126826658
+I1201 00:14:09.209730 137274321021824 utils.py:1231] [42600] train/loss = 2.569148361682892
+I1201 00:14:09.209930 137274321021824 utils.py:1231] [42600] l2_grads = 1.4354416131973267
+I1201 00:14:09.210014 137274321021824 utils.py:1231] [42600] lr = 0.0007709307118205112
+I1201 00:14:09.210096 137274321021824 utils.py:1231] [42600] uptime = 267838.57245547
+I1201 00:14:09.210166 137274321021824 utils.py:1231] [42600] examples_seen = 43622400.0
+I1201 00:14:09.210238 137274321021824 utils.py:1231] [42600] progress = 0.37832029342024637
+I1201 00:14:09.210307 137274321021824 utils.py:1231] [42600] epoch = 34.04895692755121
+I1201 00:14:09.210375 137274321021824 utils.py:1231] [42600] img/sec/core = 164.19789288214352
+I1201 00:14:09.210444 137274321021824 utils.py:1231] [42600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.3652858574961
+I1201 00:14:09.210496 137274321021824 utils.py:1231] [42600] core_hours = 74.3652858574961
+I1201 00:14:09.210587 137274321021824 train.py:125] NOTE: Steps:42600/112603 [37.8%]
+Walltime:3d2h23m (0s eval)
+ETA:5d2h12m
+Total train time:8d4h34m
+I1201 00:19:21.028155 137274321021824 utils.py:1231] [42650] l2_params = 321.6376949983892
+I1201 00:19:21.028521 137274321021824 utils.py:1231] [42650] train/loss = 2.561094790697098
+I1201 00:19:21.028706 137274321021824 utils.py:1231] [42650] l2_grads = 1.504846453666687
+I1201 00:19:21.028801 137274321021824 utils.py:1231] [42650] lr = 0.0007702870392074796
+I1201 00:19:21.028866 137274321021824 utils.py:1231] [42650] uptime = 268150.39122682996
+I1201 00:19:21.028942 137274321021824 utils.py:1231] [42650] examples_seen = 43673600.0
+I1201 00:19:21.029002 137274321021824 utils.py:1231] [42650] progress = 0.37876433132332177
+I1201 00:19:21.029060 137274321021824 utils.py:1231] [42650] epoch = 34.088920492020165
+I1201 00:19:21.029118 137274321021824 utils.py:1231] [42650] img/sec/core = 164.19794028657955
+I1201 00:19:21.029200 137274321021824 utils.py:1231] [42650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.45190218287388
+I1201 00:19:21.029256 137274321021824 utils.py:1231] [42650] core_hours = 74.45190218287388
+I1201 00:19:21.029324 137274321021824 train.py:125] NOTE: Steps:42650/112603 [37.9%]
+Walltime:3d2h29m (0s eval)
+ETA:5d2h7m
+Total train time:8d4h34m
+I1201 00:24:32.839989 137274321021824 utils.py:1231] [42700] l2_params = 321.58764434067535
+I1201 00:24:32.840197 137274321021824 utils.py:1231] [42700] train/loss = 2.496429353952408
+I1201 00:24:32.840287 137274321021824 utils.py:1231] [42700] l2_grads = 1.5614757537841797
+I1201 00:24:32.840345 137274321021824 utils.py:1231] [42700] lr = 0.0007696427330971433
+I1201 00:24:32.840396 137274321021824 utils.py:1231] [42700] uptime = 268462.202758444
+I1201 00:24:32.840448 137274321021824 utils.py:1231] [42700] examples_seen = 43724800.0
+I1201 00:24:32.840496 137274321021824 utils.py:1231] [42700] progress = 0.37920836922639717
+I1201 00:24:32.840543 137274321021824 utils.py:1231] [42700] epoch = 34.12888405648912
+I1201 00:24:32.840594 137274321021824 utils.py:1231] [42700] img/sec/core = 164.20175269006484
+I1201 00:24:32.840652 137274321021824 utils.py:1231] [42700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.5385164972111
+I1201 00:24:32.840701 137274321021824 utils.py:1231] [42700] core_hours = 74.5385164972111
+I1201 00:24:32.840760 137274321021824 train.py:125] NOTE: Steps:42700/112603 [37.9%]
+Walltime:3d2h34m (0s eval)
+ETA:5d2h1m
+Total train time:8d4h34m
+I1201 00:29:44.669811 137274321021824 utils.py:1231] [42750] l2_params = 321.53726109530413
+I1201 00:29:44.670071 137274321021824 utils.py:1231] [42750] train/loss = 4.55317622423172
+I1201 00:29:44.670195 137274321021824 utils.py:1231] [42750] l2_grads = 1.4036002159118652
+I1201 00:29:44.670266 137274321021824 utils.py:1231] [42750] lr = 0.0007689977949996231
+I1201 00:29:44.670323 137274321021824 utils.py:1231] [42750] uptime = 268774.03268418
+I1201 00:29:44.670381 137274321021824 utils.py:1231] [42750] examples_seen = 43776000.0
+I1201 00:29:44.670437 137274321021824 utils.py:1231] [42750] progress = 0.37965240712947257
+I1201 00:29:44.670488 137274321021824 utils.py:1231] [42750] epoch = 34.16884762095808
+I1201 00:29:44.670544 137274321021824 utils.py:1231] [42750] img/sec/core = 164.19206681063028
+I1201 00:29:44.670602 137274321021824 utils.py:1231] [42750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.62513592102667
+I1201 00:29:44.670654 137274321021824 utils.py:1231] [42750] core_hours = 74.62513592102667
+I1201 00:29:44.670724 137274321021824 train.py:125] NOTE: Steps:42750/112603 [38.0%]
+Walltime:3d2h39m (0s eval)
+ETA:5d1h56m
+Total train time:8d4h34m
+I1201 00:34:56.505319 137274321021824 utils.py:1231] [42800] l2_params = 321.4672419123385
+I1201 00:34:56.505596 137274321021824 utils.py:1231] [42800] train/loss = 2.484053134918213
+I1201 00:34:56.505716 137274321021824 utils.py:1231] [42800] l2_grads = 1.4149765968322754
+I1201 00:34:56.505796 137274321021824 utils.py:1231] [42800] lr = 0.0007683522264265214
+I1201 00:34:56.505863 137274321021824 utils.py:1231] [42800] uptime = 269085.868225166
+I1201 00:34:56.505936 137274321021824 utils.py:1231] [42800] examples_seen = 43827200.0
+I1201 00:34:56.505987 137274321021824 utils.py:1231] [42800] progress = 0.38009644503254797
+I1201 00:34:56.506036 137274321021824 utils.py:1231] [42800] epoch = 34.20881118542704
+I1201 00:34:56.506088 137274321021824 utils.py:1231] [42800] img/sec/core = 164.18911018966241
+I1201 00:34:56.506146 137274321021824 utils.py:1231] [42800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.71175690463389
+I1201 00:34:56.506199 137274321021824 utils.py:1231] [42800] core_hours = 74.71175690463389
+I1201 00:34:56.506261 137274321021824 train.py:125] NOTE: Steps:42800/112603 [38.0%]
+Walltime:3d2h44m (0s eval)
+ETA:5d1h51m
+Total train time:8d4h34m
+I1201 00:40:08.340214 137274321021824 utils.py:1231] [42850] l2_params = 321.38724534683877
+I1201 00:40:08.340447 137274321021824 utils.py:1231] [42850] train/loss = 2.4377508610486984
+I1201 00:40:08.340563 137274321021824 utils.py:1231] [42850] l2_grads = 1.3533568382263184
+I1201 00:40:08.340640 137274321021824 utils.py:1231] [42850] lr = 0.000767706028890918
+I1201 00:40:08.340703 137274321021824 utils.py:1231] [42850] uptime = 269397.70306395
+I1201 00:40:08.340766 137274321021824 utils.py:1231] [42850] examples_seen = 43878400.0
+I1201 00:40:08.340829 137274321021824 utils.py:1231] [42850] progress = 0.38054048293562337
+I1201 00:40:08.340892 137274321021824 utils.py:1231] [42850] epoch = 34.24877474989599
+I1201 00:40:08.340956 137274321021824 utils.py:1231] [42850] img/sec/core = 164.18947991716695
+I1201 00:40:08.341020 137274321021824 utils.py:1231] [42850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.798377693185
+I1201 00:40:08.341078 137274321021824 utils.py:1231] [42850] core_hours = 74.798377693185
+I1201 00:40:08.341157 137274321021824 train.py:125] NOTE: Steps:42850/112603 [38.1%]
+Walltime:3d2h49m (0s eval)
+ETA:5d1h45m
+Total train time:8d4h34m
+I1201 00:45:20.176972 137274321021824 utils.py:1231] [42900] l2_params = 321.3355754451359
+I1201 00:45:20.177219 137274321021824 utils.py:1231] [42900] train/loss = 2.498391807079315
+I1201 00:45:20.177330 137274321021824 utils.py:1231] [42900] l2_grads = 1.473616361618042
+I1201 00:45:20.177408 137274321021824 utils.py:1231] [42900] lr = 0.0007670592039073683
+I1201 00:45:20.177488 137274321021824 utils.py:1231] [42900] uptime = 269709.539849084
+I1201 00:45:20.177560 137274321021824 utils.py:1231] [42900] examples_seen = 43929600.0
+I1201 00:45:20.177634 137274321021824 utils.py:1231] [42900] progress = 0.38098452083869877
+I1201 00:45:20.177704 137274321021824 utils.py:1231] [42900] epoch = 34.28873831436495
+I1201 00:45:20.177773 137274321021824 utils.py:1231] [42900] img/sec/core = 164.18845511763791
+I1201 00:45:20.177852 137274321021824 utils.py:1231] [42900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.88499902238888
+I1201 00:45:20.177925 137274321021824 utils.py:1231] [42900] core_hours = 74.88499902238888
+I1201 00:45:20.177990 137274321021824 train.py:125] NOTE: Steps:42900/112603 [38.1%]
+Walltime:3d2h55m (0s eval)
+ETA:5d1h40m
+Total train time:8d4h33m
+I1201 00:50:32.013036 137274321021824 utils.py:1231] [42950] l2_params = 321.24538786607206
+I1201 00:50:32.013272 137274321021824 utils.py:1231] [42950] train/loss = 2.782257556915283
+I1201 00:50:32.013368 137274321021824 utils.py:1231] [42950] l2_grads = 1.391147255897522
+I1201 00:50:32.013431 137274321021824 utils.py:1231] [42950] lr = 0.000766411752991897
+I1201 00:50:32.013485 137274321021824 utils.py:1231] [42950] uptime = 270021.375846542
+I1201 00:50:32.013540 137274321021824 utils.py:1231] [42950] examples_seen = 43980800.0
+I1201 00:50:32.013591 137274321021824 utils.py:1231] [42950] progress = 0.38142855874177417
+I1201 00:50:32.013640 137274321021824 utils.py:1231] [42950] epoch = 34.328701878833904
+I1201 00:50:32.013697 137274321021824 utils.py:1231] [42950] img/sec/core = 164.18886984621483
+I1201 00:50:32.013754 137274321021824 utils.py:1231] [42950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 74.9716201327939
+I1201 00:50:32.013805 137274321021824 utils.py:1231] [42950] core_hours = 74.9716201327939
+I1201 00:50:32.013866 137274321021824 train.py:125] NOTE: Steps:42950/112603 [38.1%]
+Walltime:3d3h0m (0s eval)
+ETA:5d1h35m
+Total train time:8d4h33m
+I1201 00:55:43.858241 137274321021824 utils.py:1231] [43000] l2_params = 321.1933353999639
+I1201 00:55:43.858503 137274321021824 utils.py:1231] [43000] train/loss = 2.5431202352046967
+I1201 00:55:43.858613 137274321021824 utils.py:1231] [43000] l2_grads = 1.5034258365631104
+I1201 00:55:43.858683 137274321021824 utils.py:1231] [43000] lr = 0.0007657636776619951
+I1201 00:55:43.858746 137274321021824 utils.py:1231] [43000] uptime = 270333.22110763
+I1201 00:55:43.858806 137274321021824 utils.py:1231] [43000] examples_seen = 44032000.0
+I1201 00:55:43.858858 137274321021824 utils.py:1231] [43000] progress = 0.3818725966448496
+I1201 00:55:43.858922 137274321021824 utils.py:1231] [43000] epoch = 34.368665443302866
+I1201 00:55:43.858978 137274321021824 utils.py:1231] [43000] img/sec/core = 164.18399247552904
+I1201 00:55:43.859048 137274321021824 utils.py:1231] [43000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.05824381642944
+I1201 00:55:43.859105 137274321021824 utils.py:1231] [43000] core_hours = 75.05824381642944
+I1201 00:55:43.859176 137274321021824 train.py:125] NOTE: Steps:43000/112603 [38.2%]
+Walltime:3d3h5m (0s eval)
+ETA:5d1h30m
+Total train time:8d4h33m
+I1201 01:00:56.073076 137274321021824 utils.py:1231] [43050] l2_params = 321.1360719774987
+I1201 01:00:56.073310 137274321021824 utils.py:1231] [43050] train/loss = 2.584228128194809
+I1201 01:00:56.073411 137274321021824 utils.py:1231] [43050] l2_grads = 1.4196280241012573
+I1201 01:00:56.073488 137274321021824 utils.py:1231] [43050] lr = 0.0007651149794366191
+I1201 01:00:56.073539 137274321021824 utils.py:1231] [43050] uptime = 270645.435901301
+I1201 01:00:56.073592 137274321021824 utils.py:1231] [43050] examples_seen = 44083200.0
+I1201 01:00:56.073641 137274321021824 utils.py:1231] [43050] progress = 0.382316634547925
+I1201 01:00:56.073688 137274321021824 utils.py:1231] [43050] epoch = 34.40862900777182
+I1201 01:00:56.073743 137274321021824 utils.py:1231] [43050] img/sec/core = 163.98966685081476
+I1201 01:00:56.073808 137274321021824 utils.py:1231] [43050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.1449701480047
+I1201 01:00:56.073857 137274321021824 utils.py:1231] [43050] core_hours = 75.1449701480047
+I1201 01:00:56.073925 137274321021824 train.py:125] NOTE: Steps:43050/112603 [38.2%]
+Walltime:3d3h10m (0s eval)
+ETA:5d1h24m
+Total train time:8d4h33m
+I1201 01:06:08.135878 137274321021824 utils.py:1231] [43100] l2_params = 321.09903874517727
+I1201 01:06:08.136176 137274321021824 utils.py:1231] [43100] train/loss = 4.754967451095581
+I1201 01:06:08.136405 137274321021824 utils.py:1231] [43100] l2_grads = 1.2941174507141113
+I1201 01:06:08.136550 137274321021824 utils.py:1231] [43100] lr = 0.0007644656598361838
+I1201 01:06:08.136647 137274321021824 utils.py:1231] [43100] uptime = 270957.499002996
+I1201 01:06:08.136752 137274321021824 utils.py:1231] [43100] examples_seen = 44134400.0
+I1201 01:06:08.136854 137274321021824 utils.py:1231] [43100] progress = 0.3827606724510004
+I1201 01:06:08.136951 137274321021824 utils.py:1231] [43100] epoch = 34.44859257224078
+I1201 01:06:08.137051 137274321021824 utils.py:1231] [43100] img/sec/core = 164.06938123057984
+I1201 01:06:08.137140 137274321021824 utils.py:1231] [43100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.23165434292
+I1201 01:06:08.137241 137274321021824 utils.py:1231] [43100] core_hours = 75.23165434292
+I1201 01:06:08.137329 137274321021824 train.py:125] NOTE: Steps:43100/112603 [38.3%]
+Walltime:3d3h15m (0s eval)
+ETA:5d1h19m
+Total train time:8d4h33m
+I1201 01:11:19.962552 137274321021824 utils.py:1231] [43150] l2_params = 321.0631108479614
+I1201 01:11:19.962809 137274321021824 utils.py:1231] [43150] train/loss = 4.89202094078064
+I1201 01:11:19.962961 137274321021824 utils.py:1231] [43150] l2_grads = 1.3206562995910645
+I1201 01:11:19.963053 137274321021824 utils.py:1231] [43150] lr = 0.0007638157203825614
+I1201 01:11:19.963134 137274321021824 utils.py:1231] [43150] uptime = 271269.32549500297
+I1201 01:11:19.963207 137274321021824 utils.py:1231] [43150] examples_seen = 44185600.0
+I1201 01:11:19.963276 137274321021824 utils.py:1231] [43150] progress = 0.38320471035407583
+I1201 01:11:19.963335 137274321021824 utils.py:1231] [43150] epoch = 34.48855613670973
+I1201 01:11:19.963407 137274321021824 utils.py:1231] [43150] img/sec/core = 164.1938748387647
+I1201 01:11:19.963481 137274321021824 utils.py:1231] [43150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.31827281292193
+I1201 01:11:19.963548 137274321021824 utils.py:1231] [43150] core_hours = 75.31827281292193
+I1201 01:11:19.963628 137274321021824 train.py:125] NOTE: Steps:43150/112603 [38.3%]
+Walltime:3d3h21m (0s eval)
+ETA:5d1h14m
+Total train time:8d4h33m
+I1201 01:16:31.764748 137274321021824 utils.py:1231] [43200] l2_params = 321.003243982462
+I1201 01:16:31.764960 137274321021824 utils.py:1231] [43200] train/loss = 3.02153018116951
+I1201 01:16:31.765064 137274321021824 utils.py:1231] [43200] l2_grads = 1.4631553888320923
+I1201 01:16:31.765145 137274321021824 utils.py:1231] [43200] lr = 0.000763165162599076
+I1201 01:16:31.765227 137274321021824 utils.py:1231] [43200] uptime = 271581.127587151
+I1201 01:16:31.765296 137274321021824 utils.py:1231] [43200] examples_seen = 44236800.0
+I1201 01:16:31.765355 137274321021824 utils.py:1231] [43200] progress = 0.38364874825715123
+I1201 01:16:31.765414 137274321021824 utils.py:1231] [43200] epoch = 34.52851970117869
+I1201 01:16:31.765473 137274321021824 utils.py:1231] [43200] img/sec/core = 164.20672371784252
+I1201 01:16:31.765563 137274321021824 utils.py:1231] [43200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.40488450518528
+I1201 01:16:31.765619 137274321021824 utils.py:1231] [43200] core_hours = 75.40488450518528
+I1201 01:16:31.765690 137274321021824 train.py:125] NOTE: Steps:43200/112603 [38.4%]
+Walltime:3d3h26m (0s eval)
+ETA:5d1h8m
+Total train time:8d4h33m
+I1201 01:21:43.549666 137274321021824 utils.py:1231] [43250] l2_params = 320.92967007640067
+I1201 01:21:43.549962 137274321021824 utils.py:1231] [43250] train/loss = 2.4450496435165405
+I1201 01:21:43.550201 137274321021824 utils.py:1231] [43250] l2_grads = 1.4308308362960815
+I1201 01:21:43.550316 137274321021824 utils.py:1231] [43250] lr = 0.0007625139880105012
+I1201 01:21:43.550393 137274321021824 utils.py:1231] [43250] uptime = 271892.91275129
+I1201 01:21:43.550475 137274321021824 utils.py:1231] [43250] examples_seen = 44288000.0
+I1201 01:21:43.550587 137274321021824 utils.py:1231] [43250] progress = 0.38409278616022663
+I1201 01:21:43.550666 137274321021824 utils.py:1231] [43250] epoch = 34.56848326564765
+I1201 01:21:43.550732 137274321021824 utils.py:1231] [43250] img/sec/core = 164.21563912891605
+I1201 01:21:43.550801 137274321021824 utils.py:1231] [43250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.4914914952239
+I1201 01:21:43.550864 137274321021824 utils.py:1231] [43250] core_hours = 75.4914914952239
+I1201 01:21:43.550935 137274321021824 train.py:125] NOTE: Steps:43250/112603 [38.4%]
+Walltime:3d3h31m (0s eval)
+ETA:5d1h3m
+Total train time:8d4h33m
+I1201 01:26:55.337051 137274321021824 utils.py:1231] [43300] l2_params = 320.88903646953116
+I1201 01:26:55.337287 137274321021824 utils.py:1231] [43300] train/loss = 3.2482466995716095
+I1201 01:26:55.337397 137274321021824 utils.py:1231] [43300] l2_grads = 1.303200602531433
+I1201 01:26:55.337469 137274321021824 utils.py:1231] [43300] lr = 0.0007618621981430568
+I1201 01:26:55.337531 137274321021824 utils.py:1231] [43300] uptime = 272204.699894217
+I1201 01:26:55.337588 137274321021824 utils.py:1231] [43300] examples_seen = 44339200.0
+I1201 01:26:55.337635 137274321021824 utils.py:1231] [43300] progress = 0.38453682406330203
+I1201 01:26:55.337682 137274321021824 utils.py:1231] [43300] epoch = 34.608446830116605
+I1201 01:26:55.337730 137274321021824 utils.py:1231] [43300] img/sec/core = 164.2145969180884
+I1201 01:26:55.337784 137274321021824 utils.py:1231] [43300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.57809903492583
+I1201 01:26:55.337831 137274321021824 utils.py:1231] [43300] core_hours = 75.57809903492583
+I1201 01:26:55.337892 137274321021824 train.py:125] NOTE: Steps:43300/112603 [38.5%]
+Walltime:3d3h36m (0s eval)
+ETA:5d0h58m
+Total train time:8d4h33m
+I1201 01:32:07.109058 137274321021824 utils.py:1231] [43350] l2_params = 320.81029287409893
+I1201 01:32:07.109345 137274321021824 utils.py:1231] [43350] train/loss = 4.032649040222168
+I1201 01:32:07.109585 137274321021824 utils.py:1231] [43350] l2_grads = 1.2645777463912964
+I1201 01:32:07.109674 137274321021824 utils.py:1231] [43350] lr = 0.0007612097945244047
+I1201 01:32:07.109736 137274321021824 utils.py:1231] [43350] uptime = 272516.47209689696
+I1201 01:32:07.109798 137274321021824 utils.py:1231] [43350] examples_seen = 44390400.0
+I1201 01:32:07.109861 137274321021824 utils.py:1231] [43350] progress = 0.38498086196637743
+I1201 01:32:07.109928 137274321021824 utils.py:1231] [43350] epoch = 34.64841039458556
+I1201 01:32:07.110006 137274321021824 utils.py:1231] [43350] img/sec/core = 164.22246614641307
+I1201 01:32:07.110076 137274321021824 utils.py:1231] [43350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.66470242455915
+I1201 01:32:07.110138 137274321021824 utils.py:1231] [43350] core_hours = 75.66470242455915
+I1201 01:32:07.110232 137274321021824 train.py:125] NOTE: Steps:43350/112603 [38.5%]
+Walltime:3d3h41m (0s eval)
+ETA:5d0h52m
+Total train time:8d4h33m
+I1201 01:37:18.885848 137274321021824 utils.py:1231] [43400] l2_params = 320.7525666055955
+I1201 01:37:18.886074 137274321021824 utils.py:1231] [43400] train/loss = 2.588801294565201
+I1201 01:37:18.886217 137274321021824 utils.py:1231] [43400] l2_grads = 1.4592260122299194
+I1201 01:37:18.886317 137274321021824 utils.py:1231] [43400] lr = 0.0007605567786836452
+I1201 01:37:18.886410 137274321021824 utils.py:1231] [43400] uptime = 272828.248766791
+I1201 01:37:18.886496 137274321021824 utils.py:1231] [43400] examples_seen = 44441600.0
+I1201 01:37:18.886574 137274321021824 utils.py:1231] [43400] progress = 0.38542489986945283
+I1201 01:37:18.886661 137274321021824 utils.py:1231] [43400] epoch = 34.688373959054516
+I1201 01:37:18.886733 137274321021824 utils.py:1231] [43400] img/sec/core = 164.22011312584246
+I1201 01:37:18.886808 137274321021824 utils.py:1231] [43400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.75130705508529
+I1201 01:37:18.886878 137274321021824 utils.py:1231] [43400] core_hours = 75.75130705508529
+I1201 01:37:18.886969 137274321021824 train.py:125] NOTE: Steps:43400/112603 [38.5%]
+Walltime:3d3h47m (0s eval)
+ETA:5d0h47m
+Total train time:8d4h32m
+I1201 01:42:30.681530 137274321021824 utils.py:1231] [43450] l2_params = 320.69131640579803
+I1201 01:42:30.681772 137274321021824 utils.py:1231] [43450] train/loss = 2.481417328119278
+I1201 01:42:30.681874 137274321021824 utils.py:1231] [43450] l2_grads = 1.3977973461151123
+I1201 01:42:30.681953 137274321021824 utils.py:1231] [43450] lr = 0.0007599031521513125
+I1201 01:42:30.682024 137274321021824 utils.py:1231] [43450] uptime = 273140.04438503
+I1201 01:42:30.682086 137274321021824 utils.py:1231] [43450] examples_seen = 44492800.0
+I1201 01:42:30.682140 137274321021824 utils.py:1231] [43450] progress = 0.3858689377725283
+I1201 01:42:30.682199 137274321021824 utils.py:1231] [43450] epoch = 34.72833752352348
+I1201 01:42:30.682257 137274321021824 utils.py:1231] [43450] img/sec/core = 164.21013319294897
+I1201 01:42:30.682319 137274321021824 utils.py:1231] [43450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.83791694904056
+I1201 01:42:30.906381 137274321021824 utils.py:1231] [43450] core_hours = 75.83791694904056
+I1201 01:42:30.906671 137274321021824 train.py:125] NOTE: Steps:43450/112603 [38.6%]
+Walltime:3d3h52m (0s eval)
+ETA:5d0h42m
+Total train time:8d4h32m
+I1201 01:47:42.693056 137274321021824 utils.py:1231] [43500] l2_params = 320.6270949491393
+I1201 01:47:42.693287 137274321021824 utils.py:1231] [43500] train/loss = 2.683636575937271
+I1201 01:47:42.693414 137274321021824 utils.py:1231] [43500] l2_grads = 1.2474738359451294
+I1201 01:47:42.693483 137274321021824 utils.py:1231] [43500] lr = 0.0007592489164593735
+I1201 01:47:42.693535 137274321021824 utils.py:1231] [43500] uptime = 273452.05589669297
+I1201 01:47:42.693586 137274321021824 utils.py:1231] [43500] examples_seen = 44544000.0
+I1201 01:47:42.693639 137274321021824 utils.py:1231] [43500] progress = 0.3863129756756037
+I1201 01:47:42.693700 137274321021824 utils.py:1231] [43500] epoch = 34.768301087992434
+I1201 01:47:42.693758 137274321021824 utils.py:1231] [43500] img/sec/core = 164.09650953938083
+I1201 01:47:42.693814 137274321021824 utils.py:1231] [43500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 75.92458681339139
+I1201 01:47:42.693863 137274321021824 utils.py:1231] [43500] core_hours = 75.92458681339139
+I1201 01:47:42.693943 137274321021824 train.py:125] NOTE: Steps:43500/112603 [38.6%]
+Walltime:3d3h57m (0s eval)
+ETA:5d0h37m
+Total train time:8d4h32m
+I1201 01:52:54.483749 137274321021824 utils.py:1231] [43550] l2_params = 320.5572396466134
+I1201 01:52:54.483993 137274321021824 utils.py:1231] [43550] train/loss = 2.5303618013858795
+I1201 01:52:54.484096 137274321021824 utils.py:1231] [43550] l2_grads = 1.619947075843811
+I1201 01:52:54.484159 137274321021824 utils.py:1231] [43550] lr = 0.0007585940731412223
+I1201 01:52:54.484215 137274321021824 utils.py:1231] [43550] uptime = 273763.84657420596
+I1201 01:52:54.484308 137274321021824 utils.py:1231] [43550] examples_seen = 44595200.0
+I1201 01:52:54.484373 137274321021824 utils.py:1231] [43550] progress = 0.3867570135786791
+I1201 01:52:54.484431 137274321021824 utils.py:1231] [43550] epoch = 34.80826465246139
+I1201 01:52:54.484485 137274321021824 utils.py:1231] [43550] img/sec/core = 164.2127353145955
+I1201 01:52:54.484543 137274321021824 utils.py:1231] [43550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.01119533492277
+I1201 01:52:54.484597 137274321021824 utils.py:1231] [43550] core_hours = 76.01119533492277
+I1201 01:52:54.484668 137274321021824 train.py:125] NOTE: Steps:43550/112603 [38.7%]
+Walltime:3d4h2m (0s eval)
+ETA:5d0h31m
+Total train time:8d4h32m
+I1201 01:58:06.206274 137274321021824 utils.py:1231] [43600] l2_params = 320.4846299802636
+I1201 01:58:06.206539 137274321021824 utils.py:1231] [43600] train/loss = 3.072679817676544
+I1201 01:58:06.206681 137274321021824 utils.py:1231] [43600] l2_grads = 1.3276902437210083
+I1201 01:58:06.206781 137274321021824 utils.py:1231] [43600] lr = 0.000757938623731676
+I1201 01:58:06.206848 137274321021824 utils.py:1231] [43600] uptime = 274075.56920850696
+I1201 01:58:06.206915 137274321021824 utils.py:1231] [43600] examples_seen = 44646400.0
+I1201 01:58:06.206975 137274321021824 utils.py:1231] [43600] progress = 0.3872010514817545
+I1201 01:58:06.207035 137274321021824 utils.py:1231] [43600] epoch = 34.848228216930345
+I1201 01:58:06.207094 137274321021824 utils.py:1231] [43600] img/sec/core = 164.2485798787446
+I1201 01:58:06.207154 137274321021824 utils.py:1231] [43600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.09778495556193
+I1201 01:58:06.207211 137274321021824 utils.py:1231] [43600] core_hours = 76.09778495556193
+I1201 01:58:06.207276 137274321021824 train.py:125] NOTE: Steps:43600/112603 [38.7%]
+Walltime:3d4h7m (0s eval)
+ETA:5d0h26m
+Total train time:8d4h32m
+I1201 02:03:17.984227 137274321021824 utils.py:1231] [43650] l2_params = 320.42594135042646
+I1201 02:03:17.984446 137274321021824 utils.py:1231] [43650] train/loss = 3.732458680868149
+I1201 02:03:17.984545 137274321021824 utils.py:1231] [43650] l2_grads = 1.2074196338653564
+I1201 02:03:17.984614 137274321021824 utils.py:1231] [43650] lr = 0.0007572825697669749
+I1201 02:03:17.984675 137274321021824 utils.py:1231] [43650] uptime = 274387.347036643
+I1201 02:03:17.984737 137274321021824 utils.py:1231] [43650] examples_seen = 44697600.0
+I1201 02:03:17.984795 137274321021824 utils.py:1231] [43650] progress = 0.3876450893848299
+I1201 02:03:17.984850 137274321021824 utils.py:1231] [43650] epoch = 34.8881917813993
+I1201 02:03:17.984912 137274321021824 utils.py:1231] [43650] img/sec/core = 164.21950305477682
+I1201 02:03:17.984976 137274321021824 utils.py:1231] [43650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.18438990782194
+I1201 02:03:17.985032 137274321021824 utils.py:1231] [43650] core_hours = 76.18438990782194
+I1201 02:03:17.985099 137274321021824 train.py:125] NOTE: Steps:43650/112603 [38.8%]
+Walltime:3d4h13m (0s eval)
+ETA:5d0h21m
+Total train time:8d4h32m
+I1201 02:08:29.768563 137274321021824 utils.py:1231] [43700] l2_params = 320.387729776059
+I1201 02:08:29.768914 137274321021824 utils.py:1231] [43700] train/loss = 2.51125168800354
+I1201 02:08:29.769103 137274321021824 utils.py:1231] [43700] l2_grads = 1.5345733165740967
+I1201 02:08:29.769191 137274321021824 utils.py:1231] [43700] lr = 0.000756625912784774
+I1201 02:08:29.769262 137274321021824 utils.py:1231] [43700] uptime = 274699.131623275
+I1201 02:08:29.769335 137274321021824 utils.py:1231] [43700] examples_seen = 44748800.0
+I1201 02:08:29.769392 137274321021824 utils.py:1231] [43700] progress = 0.3880891272879053
+I1201 02:08:29.769459 137274321021824 utils.py:1231] [43700] epoch = 34.92815534586826
+I1201 02:08:29.769537 137274321021824 utils.py:1231] [43700] img/sec/core = 164.21594329944892
+I1201 02:08:29.769601 137274321021824 utils.py:1231] [43700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.27099673744193
+I1201 02:08:29.769655 137274321021824 utils.py:1231] [43700] core_hours = 76.27099673744193
+I1201 02:08:29.769724 137274321021824 train.py:125] NOTE: Steps:43700/112603 [38.8%]
+Walltime:3d4h18m (0s eval)
+ETA:5d0h15m
+Total train time:8d4h32m
+I1201 02:13:41.516926 137274321021824 utils.py:1231] [43750] l2_params = 320.30351805305406
+I1201 02:13:41.517160 137274321021824 utils.py:1231] [43750] train/loss = 3.503010243177414
+I1201 02:13:41.517256 137274321021824 utils.py:1231] [43750] l2_grads = 1.2188364267349243
+I1201 02:13:41.517322 137274321021824 utils.py:1231] [43750] lr = 0.000755968654324143
+I1201 02:13:41.517380 137274321021824 utils.py:1231] [43750] uptime = 275010.879741915
+I1201 02:13:41.517436 137274321021824 utils.py:1231] [43750] examples_seen = 44800000.0
+I1201 02:13:41.517491 137274321021824 utils.py:1231] [43750] progress = 0.3885331651909807
+I1201 02:13:41.517547 137274321021824 utils.py:1231] [43750] epoch = 34.96811891033722
+I1201 02:13:41.517601 137274321021824 utils.py:1231] [43750] img/sec/core = 164.23515312093235
+I1201 02:13:41.517660 137274321021824 utils.py:1231] [43750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.35759343706417
+I1201 02:13:41.517714 137274321021824 utils.py:1231] [43750] core_hours = 76.35759343706417
+I1201 02:13:41.517776 137274321021824 train.py:125] NOTE: Steps:43750/112603 [38.9%]
+Walltime:3d4h23m (0s eval)
+ETA:5d0h10m
+Total train time:8d4h32m
+I1201 02:18:53.291346 137274321021824 utils.py:1231] [43800] l2_params = 320.24083085859377
+I1201 02:18:53.291555 137274321021824 utils.py:1231] [43800] train/loss = 3.826468586921692
+I1201 02:18:53.291657 137274321021824 utils.py:1231] [43800] l2_grads = 1.1758570671081543
+I1201 02:18:53.291716 137274321021824 utils.py:1231] [43800] lr = 0.00075531079592556
+I1201 02:18:53.291768 137274321021824 utils.py:1231] [43800] uptime = 275322.654130636
+I1201 02:18:53.291820 137274321021824 utils.py:1231] [43800] examples_seen = 44851200.0
+I1201 02:18:53.291869 137274321021824 utils.py:1231] [43800] progress = 0.3889772030940561
+I1201 02:18:53.291923 137274321021824 utils.py:1231] [43800] epoch = 35.00808247480617
+I1201 02:18:53.291974 137274321021824 utils.py:1231] [43800] img/sec/core = 164.22131468217387
+I1201 02:18:53.292030 137274321021824 utils.py:1231] [43800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.44419743393111
+I1201 02:18:53.292081 137274321021824 utils.py:1231] [43800] core_hours = 76.44419743393111
+I1201 02:18:53.292147 137274321021824 train.py:125] NOTE: Steps:43800/112603 [38.9%]
+Walltime:3d4h28m (0s eval)
+ETA:5d0h5m
+Total train time:8d4h32m
+I1201 02:24:05.070170 137274321021824 utils.py:1231] [43850] l2_params = 320.17854439087574
+I1201 02:24:05.070392 137274321021824 utils.py:1231] [43850] train/loss = 2.5552212595939636
+I1201 02:24:05.070517 137274321021824 utils.py:1231] [43850] l2_grads = 1.5379071235656738
+I1201 02:24:05.070596 137274321021824 utils.py:1231] [43850] lr = 0.0007546523391309106
+I1201 02:24:05.306499 137274321021824 utils.py:1231] [43850] uptime = 275634.668810507
+I1201 02:24:05.306790 137274321021824 utils.py:1231] [43850] examples_seen = 44902400.0
+I1201 02:24:05.306901 137274321021824 utils.py:1231] [43850] progress = 0.3894212409971315
+I1201 02:24:05.306953 137274321021824 utils.py:1231] [43850] epoch = 35.04804603927513
+I1201 02:24:05.307001 137274321021824 utils.py:1231] [43850] img/sec/core = 164.09484329766863
+I1201 02:24:05.307075 137274321021824 utils.py:1231] [43850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.53086817833972
+I1201 02:24:05.307125 137274321021824 utils.py:1231] [43850] core_hours = 76.53086817833972
+I1201 02:24:05.307181 137274321021824 train.py:125] NOTE: Steps:43850/112603 [38.9%]
+Walltime:3d4h33m (0s eval)
+ETA:4d23h59m
+Total train time:8d4h32m
+I1201 02:29:17.093396 137274321021824 utils.py:1231] [43900] l2_params = 320.12038701542156
+I1201 02:29:17.093672 137274321021824 utils.py:1231] [43900] train/loss = 4.007271647453308
+I1201 02:29:17.093825 137274321021824 utils.py:1231] [43900] l2_grads = 1.1587462425231934
+I1201 02:29:17.093894 137274321021824 utils.py:1231] [43900] lr = 0.000753993285483482
+I1201 02:29:17.093950 137274321021824 utils.py:1231] [43900] uptime = 275946.456311294
+I1201 02:29:17.094022 137274321021824 utils.py:1231] [43900] examples_seen = 44953600.0
+I1201 02:29:17.094085 137274321021824 utils.py:1231] [43900] progress = 0.38986527890020695
+I1201 02:29:17.094135 137274321021824 utils.py:1231] [43900] epoch = 35.088009603744084
+I1201 02:29:17.094187 137274321021824 utils.py:1231] [43900] img/sec/core = 164.21440843767903
+I1201 02:29:17.094245 137274321021824 utils.py:1231] [43900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.61747581744721
+I1201 02:29:17.094298 137274321021824 utils.py:1231] [43900] core_hours = 76.61747581744721
+I1201 02:29:17.094361 137274321021824 train.py:125] NOTE: Steps:43900/112603 [39.0%]
+Walltime:3d4h39m (0s eval)
+ETA:4d23h54m
+Total train time:8d4h31m
+I1201 02:34:28.881745 137274321021824 utils.py:1231] [43950] l2_params = 320.06278884418776
+I1201 02:34:28.882002 137274321021824 utils.py:1231] [43950] train/loss = 2.582271248102188
+I1201 02:34:28.882147 137274321021824 utils.py:1231] [43950] l2_grads = 1.425732970237732
+I1201 02:34:28.882241 137274321021824 utils.py:1231] [43950] lr = 0.0007533336365279608
+I1201 02:34:28.882317 137274321021824 utils.py:1231] [43950] uptime = 276258.244678796
+I1201 02:34:28.882380 137274321021824 utils.py:1231] [43950] examples_seen = 45004800.0
+I1201 02:34:28.882445 137274321021824 utils.py:1231] [43950] progress = 0.39030931680328235
+I1201 02:34:28.882508 137274321021824 utils.py:1231] [43950] epoch = 35.127973168213046
+I1201 02:34:28.882582 137274321021824 utils.py:1231] [43950] img/sec/core = 164.21395195147625
+I1201 02:34:28.882647 137274321021824 utils.py:1231] [43950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.70408369730889
+I1201 02:34:28.882709 137274321021824 utils.py:1231] [43950] core_hours = 76.70408369730889
+I1201 02:34:28.882774 137274321021824 train.py:125] NOTE: Steps:43950/112603 [39.0%]
+Walltime:3d4h44m (0s eval)
+ETA:4d23h49m
+Total train time:8d4h31m
+I1201 02:39:40.684469 137274321021824 utils.py:1231] [44000] l2_params = 320.0128141371669
+I1201 02:39:40.684663 137274321021824 utils.py:1231] [44000] train/loss = 3.1923522651195526
+I1201 02:39:40.684759 137274321021824 utils.py:1231] [44000] l2_grads = 1.2515078783035278
+I1201 02:39:40.684830 137274321021824 utils.py:1231] [44000] lr = 0.000752673393810428
+I1201 02:39:40.684893 137274321021824 utils.py:1231] [44000] uptime = 276570.047249785
+I1201 02:39:40.684956 137274321021824 utils.py:1231] [44000] examples_seen = 45056000.0
+I1201 02:39:40.685012 137274321021824 utils.py:1231] [44000] progress = 0.39075335470635775
+I1201 02:39:40.685064 137274321021824 utils.py:1231] [44000] epoch = 35.167936732682
+I1201 02:39:40.685118 137274321021824 utils.py:1231] [44000] img/sec/core = 164.20647154254945
+I1201 02:39:40.685179 137274321021824 utils.py:1231] [44000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.7906955225836
+I1201 02:39:40.685229 137274321021824 utils.py:1231] [44000] core_hours = 76.7906955225836
+I1201 02:39:40.685304 137274321021824 train.py:125] NOTE: Steps:44000/112603 [39.1%]
+Walltime:3d4h49m (0s eval)
+ETA:4d23h44m
+Total train time:8d4h31m
+I1201 02:44:52.822359 137274321021824 utils.py:1231] [44050] l2_params = 319.9371333425686
+I1201 02:44:52.822606 137274321021824 utils.py:1231] [44050] train/loss = 2.483464241027832
+I1201 02:44:52.822830 137274321021824 utils.py:1231] [44050] l2_grads = 1.4721577167510986
+I1201 02:44:52.822950 137274321021824 utils.py:1231] [44050] lr = 0.0007520125588783571
+I1201 02:44:52.823050 137274321021824 utils.py:1231] [44050] uptime = 276882.185403881
+I1201 02:44:52.823161 137274321021824 utils.py:1231] [44050] examples_seen = 45107200.0
+I1201 02:44:52.823235 137274321021824 utils.py:1231] [44050] progress = 0.39119739260943315
+I1201 02:44:52.823318 137274321021824 utils.py:1231] [44050] epoch = 35.20790029715096
+I1201 02:44:52.823385 137274321021824 utils.py:1231] [44050] img/sec/core = 164.029931388165
+I1201 02:44:52.823454 137274321021824 utils.py:1231] [44050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.87740056538806
+I1201 02:44:52.823516 137274321021824 utils.py:1231] [44050] core_hours = 76.87740056538806
+I1201 02:44:52.823589 137274321021824 train.py:125] NOTE: Steps:44050/112603 [39.1%]
+Walltime:3d4h54m (0s eval)
+ETA:4d23h38m
+Total train time:8d4h31m
+I1201 02:50:04.611616 137274321021824 utils.py:1231] [44100] l2_params = 319.89988018204986
+I1201 02:50:04.611857 137274321021824 utils.py:1231] [44100] train/loss = 2.4650488197803497
+I1201 02:50:04.611978 137274321021824 utils.py:1231] [44100] l2_grads = 1.500726580619812
+I1201 02:50:04.612051 137274321021824 utils.py:1231] [44100] lr = 0.0007513511332806101
+I1201 02:50:04.612119 137274321021824 utils.py:1231] [44100] uptime = 277193.974480232
+I1201 02:50:04.612172 137274321021824 utils.py:1231] [44100] examples_seen = 45158400.0
+I1201 02:50:04.612219 137274321021824 utils.py:1231] [44100] progress = 0.39164143051250855
+I1201 02:50:04.612265 137274321021824 utils.py:1231] [44100] epoch = 35.24786386161991
+I1201 02:50:04.612314 137274321021824 utils.py:1231] [44100] img/sec/core = 164.21357861288428
+I1201 02:50:04.612367 137274321021824 utils.py:1231] [44100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 76.96400864215222
+I1201 02:50:04.612417 137274321021824 utils.py:1231] [44100] core_hours = 76.96400864215222
+I1201 02:50:04.612477 137274321021824 train.py:125] NOTE: Steps:44100/112603 [39.2%]
+Walltime:3d4h59m (0s eval)
+ETA:4d23h33m
+Total train time:8d4h31m
+I1201 02:55:16.413864 137274321021824 utils.py:1231] [44150] l2_params = 319.83197802174396
+I1201 02:55:16.414077 137274321021824 utils.py:1231] [44150] train/loss = 3.988792300224304
+I1201 02:55:16.414171 137274321021824 utils.py:1231] [44150] l2_grads = 1.3257673978805542
+I1201 02:55:16.414232 137274321021824 utils.py:1231] [44150] lr = 0.0007506891185674322
+I1201 02:55:16.414288 137274321021824 utils.py:1231] [44150] uptime = 277505.776650432
+I1201 02:55:16.414346 137274321021824 utils.py:1231] [44150] examples_seen = 45209600.0
+I1201 02:55:16.414392 137274321021824 utils.py:1231] [44150] progress = 0.39208546841558395
+I1201 02:55:16.414438 137274321021824 utils.py:1231] [44150] epoch = 35.28782742608887
+I1201 02:55:16.414485 137274321021824 utils.py:1231] [44150] img/sec/core = 164.20668261276552
+I1201 02:55:16.414541 137274321021824 utils.py:1231] [44150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.05062035609666
+I1201 02:55:16.414592 137274321021824 utils.py:1231] [44150] core_hours = 77.05062035609666
+I1201 02:55:16.414661 137274321021824 train.py:125] NOTE: Steps:44150/112603 [39.2%]
+Walltime:3d5h5m (0s eval)
+ETA:4d23h28m
+Total train time:8d4h31m
+I1201 03:00:28.203655 137274321021824 utils.py:1231] [44200] l2_params = 319.76055042369967
+I1201 03:00:28.203924 137274321021824 utils.py:1231] [44200] train/loss = 3.577933967113495
+I1201 03:00:28.204049 137274321021824 utils.py:1231] [44200] l2_grads = 1.2467482089996338
+I1201 03:00:28.204128 137274321021824 utils.py:1231] [44200] lr = 0.0007500265162904502
+I1201 03:00:28.204194 137274321021824 utils.py:1231] [44200] uptime = 277817.566553949
+I1201 03:00:28.204259 137274321021824 utils.py:1231] [44200] examples_seen = 45260800.0
+I1201 03:00:28.204319 137274321021824 utils.py:1231] [44200] progress = 0.39252950631865935
+I1201 03:00:28.204384 137274321021824 utils.py:1231] [44200] epoch = 35.32779099055783
+I1201 03:00:28.204446 137274321021824 utils.py:1231] [44200] img/sec/core = 164.21314296089378
+I1201 03:00:28.204512 137274321021824 utils.py:1231] [44200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.13722866262916
+I1201 03:00:28.204569 137274321021824 utils.py:1231] [44200] core_hours = 77.13722866262916
+I1201 03:00:28.204640 137274321021824 train.py:125] NOTE: Steps:44200/112603 [39.3%]
+Walltime:3d5h10m (0s eval)
+ETA:4d23h22m
+Total train time:8d4h31m
+I1201 03:05:39.987336 137274321021824 utils.py:1231] [44250] l2_params = 319.6837254968885
+I1201 03:05:39.987564 137274321021824 utils.py:1231] [44250] train/loss = 3.0705546438694
+I1201 03:05:39.987668 137274321021824 utils.py:1231] [44250] l2_grads = 1.4061801433563232
+I1201 03:05:39.987739 137274321021824 utils.py:1231] [44250] lr = 0.0007493633280026674
+I1201 03:05:39.987800 137274321021824 utils.py:1231] [44250] uptime = 278129.350161095
+I1201 03:05:39.987865 137274321021824 utils.py:1231] [44250] examples_seen = 45312000.0
+I1201 03:05:39.987929 137274321021824 utils.py:1231] [44250] progress = 0.39297354422173475
+I1201 03:05:39.987986 137274321021824 utils.py:1231] [44250] epoch = 35.367754555026785
+I1201 03:05:39.988044 137274321021824 utils.py:1231] [44250] img/sec/core = 164.2164591931984
+I1201 03:05:39.988102 137274321021824 utils.py:1231] [44250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.22383522016972
+I1201 03:05:39.988157 137274321021824 utils.py:1231] [44250] core_hours = 77.22383522016972
+I1201 03:05:39.988223 137274321021824 train.py:125] NOTE: Steps:44250/112603 [39.3%]
+Walltime:3d5h15m (0s eval)
+ETA:4d23h17m
+Total train time:8d4h31m
+I1201 03:10:51.758830 137274321021824 utils.py:1231] [44300] l2_params = 319.6306973820614
+I1201 03:10:51.759042 137274321021824 utils.py:1231] [44300] train/loss = 3.9637866020202637
+I1201 03:10:51.759145 137274321021824 utils.py:1231] [44300] l2_grads = 1.3313711881637573
+I1201 03:10:51.759215 137274321021824 utils.py:1231] [44300] lr = 0.0007486995552584609
+I1201 03:10:51.759279 137274321021824 utils.py:1231] [44300] uptime = 278441.121640746
+I1201 03:10:51.759337 137274321021824 utils.py:1231] [44300] examples_seen = 45363200.0
+I1201 03:10:51.759393 137274321021824 utils.py:1231] [44300] progress = 0.39341758212481015
+I1201 03:10:51.759448 137274321021824 utils.py:1231] [44300] epoch = 35.40771811949574
+I1201 03:10:51.759505 137274321021824 utils.py:1231] [44300] img/sec/core = 164.22284699455852
+I1201 03:10:51.759571 137274321021824 utils.py:1231] [44300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.31043840896167
+I1201 03:10:51.759627 137274321021824 utils.py:1231] [44300] core_hours = 77.31043840896167
+I1201 03:10:51.759690 137274321021824 train.py:125] NOTE: Steps:44300/112603 [39.3%]
+Walltime:3d5h20m (0s eval)
+ETA:4d23h12m
+Total train time:8d4h31m
+I1201 03:16:03.548724 137274321021824 utils.py:1231] [44350] l2_params = 319.5619390084714
+I1201 03:16:03.548934 137274321021824 utils.py:1231] [44350] train/loss = 2.6380571126937866
+I1201 03:16:03.549043 137274321021824 utils.py:1231] [44350] l2_grads = 1.4661425352096558
+I1201 03:16:03.549105 137274321021824 utils.py:1231] [44350] lr = 0.0007480351996135777
+I1201 03:16:03.549158 137274321021824 utils.py:1231] [44350] uptime = 278752.911519485
+I1201 03:16:03.549212 137274321021824 utils.py:1231] [44350] examples_seen = 45414400.0
+I1201 03:16:03.549261 137274321021824 utils.py:1231] [44350] progress = 0.3938616200278856
+I1201 03:16:03.549309 137274321021824 utils.py:1231] [44350] epoch = 35.447681683964696
+I1201 03:16:03.549359 137274321021824 utils.py:1231] [44350] img/sec/core = 164.21315601094915
+I1201 03:16:03.549414 137274321021824 utils.py:1231] [44350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.3970467086114
+I1201 03:16:03.549498 137274321021824 utils.py:1231] [44350] core_hours = 77.3970467086114
+I1201 03:16:03.549564 137274321021824 train.py:125] NOTE: Steps:44350/112603 [39.4%]
+Walltime:3d5h25m (0s eval)
+ETA:4d23h6m
+Total train time:8d4h31m
+I1201 03:21:15.323264 137274321021824 utils.py:1231] [44400] l2_params = 319.53046493424506
+I1201 03:21:15.323526 137274321021824 utils.py:1231] [44400] train/loss = 4.677810430526733
+I1201 03:21:15.323651 137274321021824 utils.py:1231] [44400] l2_grads = 1.256385326385498
+I1201 03:21:15.323748 137274321021824 utils.py:1231] [44400] lr = 0.0007473702626251314
+I1201 03:21:15.323826 137274321021824 utils.py:1231] [44400] uptime = 279064.686181616
+I1201 03:21:15.323902 137274321021824 utils.py:1231] [44400] examples_seen = 45465600.0
+I1201 03:21:15.323980 137274321021824 utils.py:1231] [44400] progress = 0.394305657930961
+I1201 03:21:15.324053 137274321021824 utils.py:1231] [44400] epoch = 35.48764524843365
+I1201 03:21:15.324124 137274321021824 utils.py:1231] [44400] img/sec/core = 164.22117066873975
+I1201 03:21:15.324199 137274321021824 utils.py:1231] [44400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.48365078142554
+I1201 03:21:15.324255 137274321021824 utils.py:1231] [44400] core_hours = 77.48365078142554
+I1201 03:21:15.324329 137274321021824 train.py:125] NOTE: Steps:44400/112603 [39.4%]
+Walltime:3d5h31m (0s eval)
+ETA:4d23h1m
+Total train time:8d4h30m
+I1201 03:26:27.112580 137274321021824 utils.py:1231] [44450] l2_params = 319.47886265920687
+I1201 03:26:27.112818 137274321021824 utils.py:1231] [44450] train/loss = 2.586194261908531
+I1201 03:26:27.112922 137274321021824 utils.py:1231] [44450] l2_grads = 1.4232388734817505
+I1201 03:26:27.112992 137274321021824 utils.py:1231] [44450] lr = 0.0007467047458515975
+I1201 03:26:27.113048 137274321021824 utils.py:1231] [44450] uptime = 279376.475409811
+I1201 03:26:27.113104 137274321021824 utils.py:1231] [44450] examples_seen = 45516800.0
+I1201 03:26:27.113155 137274321021824 utils.py:1231] [44450] progress = 0.3947496958340364
+I1201 03:26:27.113207 137274321021824 utils.py:1231] [44450] epoch = 35.52760881290261
+I1201 03:26:27.113261 137274321021824 utils.py:1231] [44450] img/sec/core = 164.21349863944903
+I1201 03:26:27.113319 137274321021824 utils.py:1231] [44450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.57025890036861
+I1201 03:26:27.113376 137274321021824 utils.py:1231] [44450] core_hours = 77.57025890036861
+I1201 03:26:27.113440 137274321021824 train.py:125] NOTE: Steps:44450/112603 [39.5%]
+Walltime:3d5h36m (0s eval)
+ETA:4d22h56m
+Total train time:8d4h30m
+I1201 03:31:38.888089 137274321021824 utils.py:1231] [44500] l2_params = 319.4053730579867
+I1201 03:31:38.888314 137274321021824 utils.py:1231] [44500] train/loss = 4.368955552577972
+I1201 03:31:38.888417 137274321021824 utils.py:1231] [44500] l2_grads = 1.3869980573654175
+I1201 03:31:38.888486 137274321021824 utils.py:1231] [44500] lr = 0.0007460386508528106
+I1201 03:31:38.888547 137274321021824 utils.py:1231] [44500] uptime = 279688.250909017
+I1201 03:31:38.888617 137274321021824 utils.py:1231] [44500] examples_seen = 45568000.0
+I1201 03:31:38.888675 137274321021824 utils.py:1231] [44500] progress = 0.3951937337371118
+I1201 03:31:38.888732 137274321021824 utils.py:1231] [44500] epoch = 35.56757237737157
+I1201 03:31:38.888790 137274321021824 utils.py:1231] [44500] img/sec/core = 164.22072975713246
+I1201 03:31:38.888856 137274321021824 utils.py:1231] [44500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.6568632057036
+I1201 03:31:38.888918 137274321021824 utils.py:1231] [44500] core_hours = 77.6568632057036
+I1201 03:31:38.888985 137274321021824 train.py:125] NOTE: Steps:44500/112603 [39.5%]
+Walltime:3d5h41m (0s eval)
+ETA:4d22h51m
+Total train time:8d4h30m
+I1201 03:36:50.662091 137274321021824 utils.py:1231] [44550] l2_params = 319.3442579306992
+I1201 03:36:50.662304 137274321021824 utils.py:1231] [44550] train/loss = 2.5195007920265198
+I1201 03:36:50.662413 137274321021824 utils.py:1231] [44550] l2_grads = 1.4161549806594849
+I1201 03:36:50.662484 137274321021824 utils.py:1231] [44550] lr = 0.000745371979189961
+I1201 03:36:50.662544 137274321021824 utils.py:1231] [44550] uptime = 280000.024905749
+I1201 03:36:50.662604 137274321021824 utils.py:1231] [44550] examples_seen = 45619200.0
+I1201 03:36:50.662660 137274321021824 utils.py:1231] [44550] progress = 0.3956377716401872
+I1201 03:36:50.662717 137274321021824 utils.py:1231] [44550] epoch = 35.607535941840524
+I1201 03:36:50.662773 137274321021824 utils.py:1231] [44550] img/sec/core = 164.22152115529883
+I1201 03:36:50.662843 137274321021824 utils.py:1231] [44550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.74346709368473
+I1201 03:36:50.662903 137274321021824 utils.py:1231] [44550] core_hours = 77.74346709368473
+I1201 03:36:50.662972 137274321021824 train.py:125] NOTE: Steps:44550/112603 [39.6%]
+Walltime:3d5h46m (0s eval)
+ETA:4d22h45m
+Total train time:8d4h30m
+I1201 03:42:02.435623 137274321021824 utils.py:1231] [44600] l2_params = 319.27976038356525
+I1201 03:42:02.435822 137274321021824 utils.py:1231] [44600] train/loss = 3.0994080901145935
+I1201 03:42:02.435943 137274321021824 utils.py:1231] [44600] l2_grads = 1.3215162754058838
+I1201 03:42:02.436013 137274321021824 utils.py:1231] [44600] lr = 0.0007447047324255897
+I1201 03:42:02.436063 137274321021824 utils.py:1231] [44600] uptime = 280311.798425225
+I1201 03:42:02.436115 137274321021824 utils.py:1231] [44600] examples_seen = 45670400.0
+I1201 03:42:02.436163 137274321021824 utils.py:1231] [44600] progress = 0.3960818095432626
+I1201 03:42:02.436211 137274321021824 utils.py:1231] [44600] epoch = 35.64749950630948
+I1201 03:42:02.436261 137274321021824 utils.py:1231] [44600] img/sec/core = 164.22177254197788
+I1201 03:42:02.436316 137274321021824 utils.py:1231] [44600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.83007084909472
+I1201 03:42:02.436365 137274321021824 utils.py:1231] [44600] core_hours = 77.83007084909472
+I1201 03:42:02.436428 137274321021824 train.py:125] NOTE: Steps:44600/112603 [39.6%]
+Walltime:3d5h51m (0s eval)
+ETA:4d22h40m
+Total train time:8d4h30m
+I1201 03:47:14.218029 137274321021824 utils.py:1231] [44650] l2_params = 319.2034885390979
+I1201 03:47:14.218244 137274321021824 utils.py:1231] [44650] train/loss = 2.3814045786857605
+I1201 03:47:14.218351 137274321021824 utils.py:1231] [44650] l2_grads = 1.427248239517212
+I1201 03:47:14.218417 137274321021824 utils.py:1231] [44650] lr = 0.0007440369121235861
+I1201 03:47:14.218472 137274321021824 utils.py:1231] [44650] uptime = 280623.580832958
+I1201 03:47:14.218526 137274321021824 utils.py:1231] [44650] examples_seen = 45721600.0
+I1201 03:47:14.218579 137274321021824 utils.py:1231] [44650] progress = 0.396525847446338
+I1201 03:47:14.218629 137274321021824 utils.py:1231] [44650] epoch = 35.68746307077844
+I1201 03:47:14.218687 137274321021824 utils.py:1231] [44650] img/sec/core = 164.21709092660112
+I1201 03:47:14.218744 137274321021824 utils.py:1231] [44650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 77.916677073465
+I1201 03:47:14.218795 137274321021824 utils.py:1231] [44650] core_hours = 77.916677073465
+I1201 03:47:14.218856 137274321021824 train.py:125] NOTE: Steps:44650/112603 [39.7%]
+Walltime:3d5h57m (0s eval)
+ETA:4d22h35m
+Total train time:8d4h30m
+I1201 03:52:26.220858 137274321021824 utils.py:1231] [44700] l2_params = 319.13595358801985
+I1201 03:52:26.221118 137274321021824 utils.py:1231] [44700] train/loss = 3.41259104013443
+I1201 03:52:26.221260 137274321021824 utils.py:1231] [44700] l2_grads = 1.2735729217529297
+I1201 03:52:26.221332 137274321021824 utils.py:1231] [44700] lr = 0.0007433685198491837
+I1201 03:52:26.221384 137274321021824 utils.py:1231] [44700] uptime = 280935.583745651
+I1201 03:52:26.221441 137274321021824 utils.py:1231] [44700] examples_seen = 45772800.0
+I1201 03:52:26.221488 137274321021824 utils.py:1231] [44700] progress = 0.3969698853494134
+I1201 03:52:26.221536 137274321021824 utils.py:1231] [44700] epoch = 35.7274266352474
+I1201 03:52:26.221586 137274321021824 utils.py:1231] [44700] img/sec/core = 164.10103212842847
+I1201 03:52:26.221641 137274321021824 utils.py:1231] [44700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.00334454921305
+I1201 03:52:26.221688 137274321021824 utils.py:1231] [44700] core_hours = 78.00334454921305
+I1201 03:52:26.221747 137274321021824 train.py:125] NOTE: Steps:44700/112603 [39.7%]
+Walltime:3d6h2m (0s eval)
+ETA:4d22h29m
+Total train time:8d4h30m
+I1201 03:57:38.000694 137274321021824 utils.py:1231] [44750] l2_params = 319.0649073910659
+I1201 03:57:38.000927 137274321021824 utils.py:1231] [44750] train/loss = 5.084886729717255
+I1201 03:57:38.001091 137274321021824 utils.py:1231] [44750] l2_grads = 1.4290579557418823
+I1201 03:57:38.001200 137274321021824 utils.py:1231] [44750] lr = 0.0007426995571689567
+I1201 03:57:38.001304 137274321021824 utils.py:1231] [44750] uptime = 281247.363660117
+I1201 03:57:38.001416 137274321021824 utils.py:1231] [44750] examples_seen = 45824000.0
+I1201 03:57:38.001510 137274321021824 utils.py:1231] [44750] progress = 0.3974139232524888
+I1201 03:57:38.001605 137274321021824 utils.py:1231] [44750] epoch = 35.76739019971635
+I1201 03:57:38.001702 137274321021824 utils.py:1231] [44750] img/sec/core = 164.21840415120474
+I1201 03:57:38.001789 137274321021824 utils.py:1231] [44750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.08995008100916
+I1201 03:57:38.001868 137274321021824 utils.py:1231] [44750] core_hours = 78.08995008100916
+I1201 03:57:38.001956 137274321021824 train.py:125] NOTE: Steps:44750/112603 [39.7%]
+Walltime:3d6h7m (0s eval)
+ETA:4d22h24m
+Total train time:8d4h30m
+I1201 04:02:49.785876 137274321021824 utils.py:1231] [44800] l2_params = 318.9900423218697
+I1201 04:02:49.786130 137274321021824 utils.py:1231] [44800] train/loss = 3.3524806797504425
+I1201 04:02:49.786259 137274321021824 utils.py:1231] [44800] l2_grads = 1.2615913152694702
+I1201 04:02:49.786341 137274321021824 utils.py:1231] [44800] lr = 0.0007420300256508162
+I1201 04:02:49.786418 137274321021824 utils.py:1231] [44800] uptime = 281559.148778589
+I1201 04:02:49.786489 137274321021824 utils.py:1231] [44800] examples_seen = 45875200.0
+I1201 04:02:49.786543 137274321021824 utils.py:1231] [44800] progress = 0.39785796115556427
+I1201 04:02:49.786600 137274321021824 utils.py:1231] [44800] epoch = 35.80735376418531
+I1201 04:02:49.786657 137274321021824 utils.py:1231] [44800] img/sec/core = 164.2156631814905
+I1201 04:02:49.786731 137274321021824 utils.py:1231] [44800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.17655705836249
+I1201 04:02:49.786790 137274321021824 utils.py:1231] [44800] core_hours = 78.17655705836249
+I1201 04:02:49.786851 137274321021824 train.py:125] NOTE: Steps:44800/112603 [39.8%]
+Walltime:3d6h12m (0s eval)
+ETA:4d22h19m
+Total train time:8d4h30m
+I1201 04:08:01.564838 137274321021824 utils.py:1231] [44850] l2_params = 318.91309605940734
+I1201 04:08:01.565088 137274321021824 utils.py:1231] [44850] train/loss = 2.403921604156494
+I1201 04:08:01.565244 137274321021824 utils.py:1231] [44850] l2_grads = 1.4446041584014893
+I1201 04:08:01.565326 137274321021824 utils.py:1231] [44850] lr = 0.0007413599268640071
+I1201 04:08:01.565387 137274321021824 utils.py:1231] [44850] uptime = 281870.927749525
+I1201 04:08:01.565446 137274321021824 utils.py:1231] [44850] examples_seen = 45926400.0
+I1201 04:08:01.565502 137274321021824 utils.py:1231] [44850] progress = 0.39830199905863967
+I1201 04:08:01.565557 137274321021824 utils.py:1231] [44850] epoch = 35.84731732865426
+I1201 04:08:01.565618 137274321021824 utils.py:1231] [44850] img/sec/core = 164.2189011218026
+I1201 04:08:01.565680 137274321021824 utils.py:1231] [44850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.26316232806694
+I1201 04:08:01.565734 137274321021824 utils.py:1231] [44850] core_hours = 78.26316232806694
+I1201 04:08:01.565800 137274321021824 train.py:125] NOTE: Steps:44850/112603 [39.8%]
+Walltime:3d6h17m (0s eval)
+ETA:4d22h14m
+Total train time:8d4h30m
+I1201 04:13:13.348945 137274321021824 utils.py:1231] [44900] l2_params = 318.8498719040911
+I1201 04:13:13.349210 137274321021824 utils.py:1231] [44900] train/loss = 2.4823980629444122
+I1201 04:13:13.349326 137274321021824 utils.py:1231] [44900] l2_grads = 1.3807168006896973
+I1201 04:13:13.349394 137274321021824 utils.py:1231] [44900] lr = 0.0007406892623791028
+I1201 04:13:13.349451 137274321021824 utils.py:1231] [44900] uptime = 282182.711812722
+I1201 04:13:13.349518 137274321021824 utils.py:1231] [44900] examples_seen = 45977600.0
+I1201 04:13:13.349567 137274321021824 utils.py:1231] [44900] progress = 0.39874603696171507
+I1201 04:13:13.349615 137274321021824 utils.py:1231] [44900] epoch = 35.887280893123226
+I1201 04:13:13.349666 137274321021824 utils.py:1231] [44900] img/sec/core = 164.2162189914553
+I1201 04:13:13.349733 137274321021824 utils.py:1231] [44900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.34976901228832
+I1201 04:13:13.349793 137274321021824 utils.py:1231] [44900] core_hours = 78.34976901228832
+I1201 04:13:13.349855 137274321021824 train.py:125] NOTE: Steps:44900/112603 [39.9%]
+Walltime:3d6h23m (0s eval)
+ETA:4d22h8m
+Total train time:8d4h29m
+I1201 04:18:25.129523 137274321021824 utils.py:1231] [44950] l2_params = 318.78627359857524
+I1201 04:18:25.129802 137274321021824 utils.py:1231] [44950] train/loss = 2.450736552476883
+I1201 04:18:25.129920 137274321021824 utils.py:1231] [44950] l2_grads = 1.4525585174560547
+I1201 04:18:25.129998 137274321021824 utils.py:1231] [44950] lr = 0.0007400180337680034
+I1201 04:18:25.130051 137274321021824 utils.py:1231] [44950] uptime = 282494.492413719
+I1201 04:18:25.130102 137274321021824 utils.py:1231] [44950] examples_seen = 46028800.0
+I1201 04:18:25.130150 137274321021824 utils.py:1231] [44950] progress = 0.39919007486479047
+I1201 04:18:25.130196 137274321021824 utils.py:1231] [44950] epoch = 35.92724445759218
+I1201 04:18:25.130246 137274321021824 utils.py:1231] [44950] img/sec/core = 164.2180425474468
+I1201 04:18:25.130300 137274321021824 utils.py:1231] [44950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.43637473478749
+I1201 04:18:25.130349 137274321021824 utils.py:1231] [44950] core_hours = 78.43637473478749
+I1201 04:18:25.130406 137274321021824 train.py:125] NOTE: Steps:44950/112603 [39.9%]
+Walltime:3d6h28m (0s eval)
+ETA:4d22h3m
+Total train time:8d4h29m
+I1201 04:23:36.923719 137274321021824 utils.py:1231] [45000] l2_params = 318.73668365655976
+I1201 04:23:36.924002 137274321021824 utils.py:1231] [45000] train/loss = 2.42464742064476
+I1201 04:23:36.924191 137274321021824 utils.py:1231] [45000] l2_grads = 1.3985706567764282
+I1201 04:23:36.924282 137274321021824 utils.py:1231] [45000] lr = 0.0007393462426039304
+I1201 04:23:36.924342 137274321021824 utils.py:1231] [45000] uptime = 282806.286703514
+I1201 04:23:36.924411 137274321021824 utils.py:1231] [45000] examples_seen = 46080000.0
+I1201 04:23:36.924477 137274321021824 utils.py:1231] [45000] progress = 0.39963411276786587
+I1201 04:23:36.924542 137274321021824 utils.py:1231] [45000] epoch = 35.967208022061136
+I1201 04:23:36.924609 137274321021824 utils.py:1231] [45000] img/sec/core = 164.21083283361588
+I1201 04:23:36.924672 137274321021824 utils.py:1231] [45000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.52298425973055
+I1201 04:23:36.924733 137274321021824 utils.py:1231] [45000] core_hours = 78.52298425973055
+I1201 04:23:36.924809 137274321021824 train.py:125] NOTE: Steps:45000/112603 [40.0%]
+Walltime:3d6h33m (0s eval)
+ETA:4d21h58m
+Total train time:8d4h29m
+I1201 04:23:37.273357 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:45000/112603 [40.0%]
+Walltime:3d6h33m (0s eval)
+ETA:4d21h58m
+Total train time:8d4h29m
+I1201 04:25:15.064250 137274321021824 utils.py:1231] [45000] val/acc@1 = 0.6341278698979592
+I1201 04:25:15.064515 137274321021824 utils.py:1231] [45000] val/loss = 1.5028617123560029
+I1201 04:25:15.064719 137274321021824 utils.py:1231] [45000] z/secs/eval/val = 97.79110947600566
+I1201 04:25:15.064808 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.79110947600566
+I1201 04:30:25.806140 137274321021824 utils.py:1231] [45050] l2_params = 318.7095272025993
+I1201 04:30:25.806358 137274321021824 utils.py:1231] [45050] train/loss = 2.4705382585525513
+I1201 04:30:25.806479 137274321021824 utils.py:1231] [45050] l2_grads = 1.5722204446792603
+I1201 04:30:25.806550 137274321021824 utils.py:1231] [45050] lr = 0.0007386738904614249
+I1201 04:30:25.806612 137274321021824 utils.py:1231] [45050] uptime = 283215.168973949
+I1201 04:30:25.806682 137274321021824 utils.py:1231] [45050] examples_seen = 46131200.0
+I1201 04:30:25.806730 137274321021824 utils.py:1231] [45050] progress = 0.40007815067094127
+I1201 04:30:25.806781 137274321021824 utils.py:1231] [45050] epoch = 36.00717158653009
+I1201 04:30:25.806831 137274321021824 utils.py:1231] [45050] img/sec/core = 125.21941816046728
+I1201 04:30:25.806908 137274321021824 utils.py:1231] [45050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.63656266818472
+I1201 04:30:25.806959 137274321021824 utils.py:1231] [45050] core_hours = 78.63656266818472
+I1201 04:30:25.807018 137274321021824 train.py:125] NOTE: Steps:45050/112603 [40.0%]
+Walltime:3d6h40m (0s eval)
+ETA:4d21h55m
+Total train time:8d4h33m
+I1201 04:35:37.523676 137274321021824 utils.py:1231] [45100] l2_params = 318.65684021134473
+I1201 04:35:37.523899 137274321021824 utils.py:1231] [45100] train/loss = 4.109836041927338
+I1201 04:35:37.523993 137274321021824 utils.py:1231] [45100] l2_grads = 1.3421047925949097
+I1201 04:35:37.524052 137274321021824 utils.py:1231] [45100] lr = 0.000738000978916342
+I1201 04:35:37.524104 137274321021824 utils.py:1231] [45100] uptime = 283526.886465384
+I1201 04:35:37.524162 137274321021824 utils.py:1231] [45100] examples_seen = 46182400.0
+I1201 04:35:37.524209 137274321021824 utils.py:1231] [45100] progress = 0.40052218857401667
+I1201 04:35:37.524255 137274321021824 utils.py:1231] [45100] epoch = 36.04713515099905
+I1201 04:35:37.524304 137274321021824 utils.py:1231] [45100] img/sec/core = 164.25128973128466
+I1201 04:35:37.524357 137274321021824 utils.py:1231] [45100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.72315086024999
+I1201 04:35:37.524407 137274321021824 utils.py:1231] [45100] core_hours = 78.72315086024999
+I1201 04:35:37.524466 137274321021824 train.py:125] NOTE: Steps:45100/112603 [40.1%]
+Walltime:3d6h45m (0s eval)
+ETA:4d21h50m
+Total train time:8d4h33m
+I1201 04:40:49.226792 137274321021824 utils.py:1231] [45150] l2_params = 318.5933315461438
+I1201 04:40:49.227077 137274321021824 utils.py:1231] [45150] train/loss = 2.399053692817688
+I1201 04:40:49.227197 137274321021824 utils.py:1231] [45150] l2_grads = 1.3997783660888672
+I1201 04:40:49.227296 137274321021824 utils.py:1231] [45150] lr = 0.000737327509545848
+I1201 04:40:49.227406 137274321021824 utils.py:1231] [45150] uptime = 283838.589760884
+I1201 04:40:49.227487 137274321021824 utils.py:1231] [45150] examples_seen = 46233600.0
+I1201 04:40:49.227578 137274321021824 utils.py:1231] [45150] progress = 0.40096622647709207
+I1201 04:40:49.227648 137274321021824 utils.py:1231] [45150] epoch = 36.08709871546801
+I1201 04:40:49.227741 137274321021824 utils.py:1231] [45150] img/sec/core = 164.2587702445292
+I1201 04:40:49.227817 137274321021824 utils.py:1231] [45150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.809735109
+I1201 04:40:49.227905 137274321021824 utils.py:1231] [45150] core_hours = 78.809735109
+I1201 04:40:49.227978 137274321021824 train.py:125] NOTE: Steps:45150/112603 [40.1%]
+Walltime:3d6h50m (0s eval)
+ETA:4d21h44m
+Total train time:8d4h33m
+I1201 04:46:00.930359 137274321021824 utils.py:1231] [45200] l2_params = 318.5432361711732
+I1201 04:46:00.930643 137274321021824 utils.py:1231] [45200] train/loss = 2.869805544614792
+I1201 04:46:00.930828 137274321021824 utils.py:1231] [45200] l2_grads = 1.2553791999816895
+I1201 04:46:00.930935 137274321021824 utils.py:1231] [45200] lr = 0.0007366534839284169
+I1201 04:46:00.931018 137274321021824 utils.py:1231] [45200] uptime = 284150.29337493697
+I1201 04:46:00.931099 137274321021824 utils.py:1231] [45200] examples_seen = 46284800.0
+I1201 04:46:00.931164 137274321021824 utils.py:1231] [45200] progress = 0.40141026438016747
+I1201 04:46:00.931232 137274321021824 utils.py:1231] [45200] epoch = 36.127062279936965
+I1201 04:46:00.931292 137274321021824 utils.py:1231] [45200] img/sec/core = 164.25860237635771
+I1201 04:46:00.931350 137274321021824 utils.py:1231] [45200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.89631944623693
+I1201 04:46:00.931404 137274321021824 utils.py:1231] [45200] core_hours = 78.89631944623693
+I1201 04:46:00.931469 137274321021824 train.py:125] NOTE: Steps:45200/112603 [40.1%]
+Walltime:3d6h55m (0s eval)
+ETA:4d21h39m
+Total train time:8d4h33m
+I1201 04:51:12.631423 137274321021824 utils.py:1231] [45250] l2_params = 318.481558979077
+I1201 04:51:12.631707 137274321021824 utils.py:1231] [45250] train/loss = 2.4275142550468445
+I1201 04:51:12.631898 137274321021824 utils.py:1231] [45250] l2_grads = 1.4877351522445679
+I1201 04:51:12.631997 137274321021824 utils.py:1231] [45250] lr = 0.000735978903643826
+I1201 04:51:12.632077 137274321021824 utils.py:1231] [45250] uptime = 284461.99443493097
+I1201 04:51:12.632152 137274321021824 utils.py:1231] [45250] examples_seen = 46336000.0
+I1201 04:51:12.632221 137274321021824 utils.py:1231] [45250] progress = 0.40185430228324287
+I1201 04:51:12.632285 137274321021824 utils.py:1231] [45250] epoch = 36.16702584440592
+I1201 04:51:12.632355 137274321021824 utils.py:1231] [45250] img/sec/core = 164.25994830106157
+I1201 04:51:12.632426 137274321021824 utils.py:1231] [45250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 78.98290307401304
+I1201 04:51:12.632488 137274321021824 utils.py:1231] [45250] core_hours = 78.98290307401304
+I1201 04:51:12.632561 137274321021824 train.py:125] NOTE: Steps:45250/112603 [40.2%]
+Walltime:3d7h1m (0s eval)
+ETA:4d21h34m
+Total train time:8d4h33m
+I1201 04:56:24.368148 137274321021824 utils.py:1231] [45300] l2_params = 318.39596704866153
+I1201 04:56:24.368397 137274321021824 utils.py:1231] [45300] train/loss = 2.93537375330925
+I1201 04:56:24.368521 137274321021824 utils.py:1231] [45300] l2_grads = 1.523155927658081
+I1201 04:56:24.368601 137274321021824 utils.py:1231] [45300] lr = 0.0007353037702731531
+I1201 04:56:24.368664 137274321021824 utils.py:1231] [45300] uptime = 284773.731026752
+I1201 04:56:24.368728 137274321021824 utils.py:1231] [45300] examples_seen = 46387200.0
+I1201 04:56:24.368778 137274321021824 utils.py:1231] [45300] progress = 0.4022983401863183
+I1201 04:56:24.368834 137274321021824 utils.py:1231] [45300] epoch = 36.206989408874875
+I1201 04:56:24.368895 137274321021824 utils.py:1231] [45300] img/sec/core = 164.24122590458396
+I1201 04:56:24.368951 137274321021824 utils.py:1231] [45300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.06949657174111
+I1201 04:56:24.369001 137274321021824 utils.py:1231] [45300] core_hours = 79.06949657174111
+I1201 04:56:24.369063 137274321021824 train.py:125] NOTE: Steps:45300/112603 [40.2%]
+Walltime:3d7h6m (0s eval)
+ETA:4d21h28m
+Total train time:8d4h33m
+I1201 05:01:36.149243 137274321021824 utils.py:1231] [45350] l2_params = 318.3195790603395
+I1201 05:01:36.149513 137274321021824 utils.py:1231] [45350] train/loss = 4.4467350244522095
+I1201 05:01:36.149694 137274321021824 utils.py:1231] [45350] l2_grads = 1.3088560104370117
+I1201 05:01:36.149794 137274321021824 utils.py:1231] [45350] lr = 0.000734628085398773
+I1201 05:01:36.149872 137274321021824 utils.py:1231] [45350] uptime = 285085.512232608
+I1201 05:01:36.149963 137274321021824 utils.py:1231] [45350] examples_seen = 46438400.0
+I1201 05:01:36.150047 137274321021824 utils.py:1231] [45350] progress = 0.4027423780893937
+I1201 05:01:36.150111 137274321021824 utils.py:1231] [45350] epoch = 36.24695297334383
+I1201 05:01:36.150175 137274321021824 utils.py:1231] [45350] img/sec/core = 164.21772396263668
+I1201 05:01:36.150239 137274321021824 utils.py:1231] [45350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.15610246225667
+I1201 05:01:36.150293 137274321021824 utils.py:1231] [45350] core_hours = 79.15610246225667
+I1201 05:01:36.150362 137274321021824 train.py:125] NOTE: Steps:45350/112603 [40.3%]
+Walltime:3d7h11m (0s eval)
+ETA:4d21h23m
+Total train time:8d4h33m
+I1201 05:06:47.928617 137274321021824 utils.py:1231] [45400] l2_params = 318.23694155072917
+I1201 05:06:47.928822 137274321021824 utils.py:1231] [45400] train/loss = 2.478846102952957
+I1201 05:06:47.928927 137274321021824 utils.py:1231] [45400] l2_grads = 1.4808160066604614
+I1201 05:06:47.929002 137274321021824 utils.py:1231] [45400] lr = 0.0007339518506043514
+I1201 05:06:47.929062 137274321021824 utils.py:1231] [45400] uptime = 285397.291423662
+I1201 05:06:47.929121 137274321021824 utils.py:1231] [45400] examples_seen = 46489600.0
+I1201 05:06:47.929179 137274321021824 utils.py:1231] [45400] progress = 0.4031864159924691
+I1201 05:06:47.929235 137274321021824 utils.py:1231] [45400] epoch = 36.28691653781279
+I1201 05:06:47.929297 137274321021824 utils.py:1231] [45400] img/sec/core = 164.21878518227598
+I1201 05:06:47.929359 137274321021824 utils.py:1231] [45400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.24270779310498
+I1201 05:06:47.929415 137274321021824 utils.py:1231] [45400] core_hours = 79.24270779310498
+I1201 05:06:47.929489 137274321021824 train.py:125] NOTE: Steps:45400/112603 [40.3%]
+Walltime:3d7h16m (0s eval)
+ETA:4d21h18m
+Total train time:8d4h32m
+I1201 05:11:59.701502 137274321021824 utils.py:1231] [45450] l2_params = 318.1647186838295
+I1201 05:11:59.701740 137274321021824 utils.py:1231] [45450] train/loss = 2.942715883255005
+I1201 05:11:59.701839 137274321021824 utils.py:1231] [45450] l2_grads = 1.408745527267456
+I1201 05:11:59.701915 137274321021824 utils.py:1231] [45450] lr = 0.000733275067474844
+I1201 05:11:59.701977 137274321021824 utils.py:1231] [45450] uptime = 285709.064338066
+I1201 05:11:59.702037 137274321021824 utils.py:1231] [45450] examples_seen = 46540800.0
+I1201 05:11:59.702095 137274321021824 utils.py:1231] [45450] progress = 0.4036304538955445
+I1201 05:11:59.702153 137274321021824 utils.py:1231] [45450] epoch = 36.32688010228175
+I1201 05:11:59.702207 137274321021824 utils.py:1231] [45450] img/sec/core = 164.22209125471272
+I1201 05:11:59.702270 137274321021824 utils.py:1231] [45450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.32931138043942
+I1201 05:11:59.702327 137274321021824 utils.py:1231] [45450] core_hours = 79.32931138043942
+I1201 05:11:59.702392 137274321021824 train.py:125] NOTE: Steps:45450/112603 [40.4%]
+Walltime:3d7h21m (0s eval)
+ETA:4d21h12m
+Total train time:8d4h32m
+I1201 05:17:11.703753 137274321021824 utils.py:1231] [45500] l2_params = 318.1166676459649
+I1201 05:17:11.703983 137274321021824 utils.py:1231] [45500] train/loss = 4.395876169204712
+I1201 05:17:11.704137 137274321021824 utils.py:1231] [45500] l2_grads = 1.2209514379501343
+I1201 05:17:11.704247 137274321021824 utils.py:1231] [45500] lr = 0.0007325977375964911
+I1201 05:17:11.704315 137274321021824 utils.py:1231] [45500] uptime = 286021.066676584
+I1201 05:17:11.704405 137274321021824 utils.py:1231] [45500] examples_seen = 46592000.0
+I1201 05:17:11.704463 137274321021824 utils.py:1231] [45500] progress = 0.40407449179861993
+I1201 05:17:11.704517 137274321021824 utils.py:1231] [45500] epoch = 36.366843666750704
+I1201 05:17:11.704572 137274321021824 utils.py:1231] [45500] img/sec/core = 164.10133412201884
+I1201 05:17:11.704641 137274321021824 utils.py:1231] [45500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.41597869669445
+I1201 05:17:11.704695 137274321021824 utils.py:1231] [45500] core_hours = 79.41597869669445
+I1201 05:17:11.704757 137274321021824 train.py:125] NOTE: Steps:45500/112603 [40.4%]
+Walltime:3d7h27m (0s eval)
+ETA:4d21h7m
+Total train time:8d4h32m
+I1201 05:22:23.479503 137274321021824 utils.py:1231] [45550] l2_params = 318.05100837669704
+I1201 05:22:23.479701 137274321021824 utils.py:1231] [45550] train/loss = 4.350798964500427
+I1201 05:22:23.479805 137274321021824 utils.py:1231] [45550] l2_grads = 1.225569248199463
+I1201 05:22:23.479864 137274321021824 utils.py:1231] [45550] lr = 0.0007319198625568152
+I1201 05:22:23.479922 137274321021824 utils.py:1231] [45550] uptime = 286332.842283691
+I1201 05:22:23.479974 137274321021824 utils.py:1231] [45550] examples_seen = 46643200.0
+I1201 05:22:23.480022 137274321021824 utils.py:1231] [45550] progress = 0.40451852970169533
+I1201 05:22:23.480070 137274321021824 utils.py:1231] [45550] epoch = 36.40680723121966
+I1201 05:22:23.480121 137274321021824 utils.py:1231] [45550] img/sec/core = 164.22067292272996
+I1201 05:22:23.480175 137274321021824 utils.py:1231] [45550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.50258303200194
+I1201 05:22:23.480226 137274321021824 utils.py:1231] [45550] core_hours = 79.50258303200194
+I1201 05:22:23.480301 137274321021824 train.py:125] NOTE: Steps:45550/112603 [40.5%]
+Walltime:3d7h32m (0s eval)
+ETA:4d21h2m
+Total train time:8d4h32m
+I1201 05:27:35.091783 137274321021824 utils.py:1231] [45600] l2_params = 318.0007929903578
+I1201 05:27:35.092007 137274321021824 utils.py:1231] [45600] train/loss = 2.652724415063858
+I1201 05:27:35.092150 137274321021824 utils.py:1231] [45600] l2_grads = 1.4515539407730103
+I1201 05:27:35.092242 137274321021824 utils.py:1231] [45600] lr = 0.0007312414439446163
+I1201 05:27:35.092299 137274321021824 utils.py:1231] [45600] uptime = 286644.454660863
+I1201 05:27:35.092353 137274321021824 utils.py:1231] [45600] examples_seen = 46694400.0
+I1201 05:27:35.092406 137274321021824 utils.py:1231] [45600] progress = 0.40496256760477073
+I1201 05:27:35.092455 137274321021824 utils.py:1231] [45600] epoch = 36.44677079568862
+I1201 05:27:35.092506 137274321021824 utils.py:1231] [45600] img/sec/core = 164.3066955961811
+I1201 05:27:35.092562 137274321021824 utils.py:1231] [45600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.58914202566082
+I1201 05:27:35.092612 137274321021824 utils.py:1231] [45600] core_hours = 79.58914202566082
+I1201 05:27:35.092672 137274321021824 train.py:125] NOTE: Steps:45600/112603 [40.5%]
+Walltime:3d7h37m (0s eval)
+ETA:4d20h57m
+Total train time:8d4h32m
+I1201 05:32:46.862453 137274321021824 utils.py:1231] [45650] l2_params = 317.93928604775397
+I1201 05:32:46.862656 137274321021824 utils.py:1231] [45650] train/loss = 2.417535126209259
+I1201 05:32:46.862767 137274321021824 utils.py:1231] [45650] l2_grads = 1.5541057586669922
+I1201 05:32:46.862856 137274321021824 utils.py:1231] [45650] lr = 0.000730562483349968
+I1201 05:32:46.862915 137274321021824 utils.py:1231] [45650] uptime = 286956.225277929
+I1201 05:32:46.862965 137274321021824 utils.py:1231] [45650] examples_seen = 46745600.0
+I1201 05:32:46.863012 137274321021824 utils.py:1231] [45650] progress = 0.40540660550784613
+I1201 05:32:46.863057 137274321021824 utils.py:1231] [45650] epoch = 36.48673436015758
+I1201 05:32:46.863103 137274321021824 utils.py:1231] [45650] img/sec/core = 164.22330135478967
+I1201 05:32:46.863154 137274321021824 utils.py:1231] [45650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.67574497484583
+I1201 05:32:46.863201 137274321021824 utils.py:1231] [45650] core_hours = 79.67574497484583
+I1201 05:32:46.863255 137274321021824 train.py:125] NOTE: Steps:45650/112603 [40.5%]
+Walltime:3d7h42m (0s eval)
+ETA:4d20h51m
+Total train time:8d4h32m
+I1201 05:37:58.637933 137274321021824 utils.py:1231] [45700] l2_params = 317.85633561599946
+I1201 05:37:58.638143 137274321021824 utils.py:1231] [45700] train/loss = 2.4640650749206543
+I1201 05:37:58.638237 137274321021824 utils.py:1231] [45700] l2_grads = 1.5567185878753662
+I1201 05:37:58.638306 137274321021824 utils.py:1231] [45700] lr = 0.0007298829823642146
+I1201 05:37:58.638366 137274321021824 utils.py:1231] [45700] uptime = 287268.00072767
+I1201 05:37:58.638431 137274321021824 utils.py:1231] [45700] examples_seen = 46796800.0
+I1201 05:37:58.638487 137274321021824 utils.py:1231] [45700] progress = 0.40585064341092153
+I1201 05:37:58.638544 137274321021824 utils.py:1231] [45700] epoch = 36.52669792462653
+I1201 05:37:58.638601 137274321021824 utils.py:1231] [45700] img/sec/core = 164.22075581170543
+I1201 05:37:58.638663 137274321021824 utils.py:1231] [45700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.76234926644055
+I1201 05:37:58.638718 137274321021824 utils.py:1231] [45700] core_hours = 79.76234926644055
+I1201 05:37:58.638784 137274321021824 train.py:125] NOTE: Steps:45700/112603 [40.6%]
+Walltime:3d7h47m (0s eval)
+ETA:4d20h46m
+Total train time:8d4h32m
+I1201 05:43:10.421406 137274321021824 utils.py:1231] [45750] l2_params = 317.7968415592693
+I1201 05:43:10.421671 137274321021824 utils.py:1231] [45750] train/loss = 4.990030765533447
+I1201 05:43:10.421858 137274321021824 utils.py:1231] [45750] l2_grads = 1.3333933353424072
+I1201 05:43:10.421937 137274321021824 utils.py:1231] [45750] lr = 0.0007292029425799669
+I1201 05:43:10.421998 137274321021824 utils.py:1231] [45750] uptime = 287579.784359182
+I1201 05:43:10.422059 137274321021824 utils.py:1231] [45750] examples_seen = 46848000.0
+I1201 05:43:10.422117 137274321021824 utils.py:1231] [45750] progress = 0.406294681313997
+I1201 05:43:10.422173 137274321021824 utils.py:1231] [45750] epoch = 36.56666148909549
+I1201 05:43:10.422232 137274321021824 utils.py:1231] [45750] img/sec/core = 164.21644635961516
+I1201 05:43:10.422294 137274321021824 utils.py:1231] [45750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.84895583074945
+I1201 05:43:10.422351 137274321021824 utils.py:1231] [45750] core_hours = 79.84895583074945
+I1201 05:43:10.422420 137274321021824 train.py:125] NOTE: Steps:45750/112603 [40.6%]
+Walltime:3d7h52m (0s eval)
+ETA:4d20h41m
+Total train time:8d4h32m
+I1201 05:48:22.202168 137274321021824 utils.py:1231] [45800] l2_params = 317.74792366414914
+I1201 05:48:22.202378 137274321021824 utils.py:1231] [45800] train/loss = 4.7455320954322815
+I1201 05:48:22.202479 137274321021824 utils.py:1231] [45800] l2_grads = 1.3133546113967896
+I1201 05:48:22.202548 137274321021824 utils.py:1231] [45800] lr = 0.0007285223655910981
+I1201 05:48:22.202608 137274321021824 utils.py:1231] [45800] uptime = 287891.564969624
+I1201 05:48:22.202668 137274321021824 utils.py:1231] [45800] examples_seen = 46899200.0
+I1201 05:48:22.202726 137274321021824 utils.py:1231] [45800] progress = 0.4067387192170724
+I1201 05:48:22.202781 137274321021824 utils.py:1231] [45800] epoch = 36.60662505356444
+I1201 05:48:22.202838 137274321021824 utils.py:1231] [45800] img/sec/core = 164.218037572695
+I1201 05:48:22.202907 137274321021824 utils.py:1231] [45800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 79.93556155587221
+I1201 05:48:22.202965 137274321021824 utils.py:1231] [45800] core_hours = 79.93556155587221
+I1201 05:48:22.203027 137274321021824 train.py:125] NOTE: Steps:45800/112603 [40.7%]
+Walltime:3d7h58m (0s eval)
+ETA:4d20h35m
+Total train time:8d4h32m
+I1201 05:53:33.972203 137274321021824 utils.py:1231] [45850] l2_params = 317.6922661648741
+I1201 05:53:33.972411 137274321021824 utils.py:1231] [45850] train/loss = 2.4201360046863556
+I1201 05:53:33.972512 137274321021824 utils.py:1231] [45850] l2_grads = 1.5442482233047485
+I1201 05:53:33.972581 137274321021824 utils.py:1231] [45850] lr = 0.0007278412529927411
+I1201 05:53:33.972640 137274321021824 utils.py:1231] [45850] uptime = 288203.33500091097
+I1201 05:53:33.972698 137274321021824 utils.py:1231] [45850] examples_seen = 46950400.0
+I1201 05:53:33.972749 137274321021824 utils.py:1231] [45850] progress = 0.4071827571201478
+I1201 05:53:33.972803 137274321021824 utils.py:1231] [45850] epoch = 36.646588618033405
+I1201 05:53:33.972858 137274321021824 utils.py:1231] [45850] img/sec/core = 164.22360991095448
+I1201 05:53:34.203353 137274321021824 utils.py:1231] [45850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.02216434234083
+I1201 05:53:34.203630 137274321021824 utils.py:1231] [45850] core_hours = 80.02216434234083
+I1201 05:53:34.203724 137274321021824 train.py:125] NOTE: Steps:45850/112603 [40.7%]
+Walltime:3d8h3m (0s eval)
+ETA:4d20h30m
+Total train time:8d4h32m
+I1201 05:58:45.944502 137274321021824 utils.py:1231] [45900] l2_params = 317.6227835202937
+I1201 05:58:45.944746 137274321021824 utils.py:1231] [45900] train/loss = 3.2948841750621796
+I1201 05:58:45.944868 137274321021824 utils.py:1231] [45900] l2_grads = 1.2216142416000366
+I1201 05:58:45.944945 137274321021824 utils.py:1231] [45900] lr = 0.0007271596063812842
+I1201 05:58:45.945000 137274321021824 utils.py:1231] [45900] uptime = 288515.307358343
+I1201 05:58:45.945065 137274321021824 utils.py:1231] [45900] examples_seen = 47001600.0
+I1201 05:58:45.945111 137274321021824 utils.py:1231] [45900] progress = 0.4076267950232232
+I1201 05:58:45.945157 137274321021824 utils.py:1231] [45900] epoch = 36.68655218250236
+I1201 05:58:45.945205 137274321021824 utils.py:1231] [45900] img/sec/core = 164.1171045455799
+I1201 05:58:45.945259 137274321021824 utils.py:1231] [45900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.10882333051637
+I1201 05:58:45.945308 137274321021824 utils.py:1231] [45900] core_hours = 80.10882333051637
+I1201 05:58:45.945365 137274321021824 train.py:125] NOTE: Steps:45900/112603 [40.8%]
+Walltime:3d8h8m (0s eval)
+ETA:4d20h25m
+Total train time:8d4h32m
+I1201 06:03:57.730019 137274321021824 utils.py:1231] [45950] l2_params = 317.54744654664415
+I1201 06:03:57.730237 137274321021824 utils.py:1231] [45950] train/loss = 3.2098982632160187
+I1201 06:03:57.730330 137274321021824 utils.py:1231] [45950] l2_grads = 1.2946213483810425
+I1201 06:03:57.730408 137274321021824 utils.py:1231] [45950] lr = 0.0007264774273543666
+I1201 06:03:57.730503 137274321021824 utils.py:1231] [45950] uptime = 288827.092861127
+I1201 06:03:57.730578 137274321021824 utils.py:1231] [45950] examples_seen = 47052800.0
+I1201 06:03:57.730629 137274321021824 utils.py:1231] [45950] progress = 0.4080708329262986
+I1201 06:03:57.730676 137274321021824 utils.py:1231] [45950] epoch = 36.726515746971316
+I1201 06:03:57.730735 137274321021824 utils.py:1231] [45950] img/sec/core = 164.2154607665274
+I1201 06:03:57.730789 137274321021824 utils.py:1231] [45950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.19543041462306
+I1201 06:03:57.730838 137274321021824 utils.py:1231] [45950] core_hours = 80.19543041462306
+I1201 06:03:57.730900 137274321021824 train.py:125] NOTE: Steps:45950/112603 [40.8%]
+Walltime:3d8h13m (0s eval)
+ETA:4d20h19m
+Total train time:8d4h31m
+I1201 06:09:09.514123 137274321021824 utils.py:1231] [46000] l2_params = 317.49663377116883
+I1201 06:09:09.514347 137274321021824 utils.py:1231] [46000] train/loss = 2.6351684629917145
+I1201 06:09:09.514453 137274321021824 utils.py:1231] [46000] l2_grads = 1.3708924055099487
+I1201 06:09:09.514537 137274321021824 utils.py:1231] [46000] lr = 0.0007257947175108763
+I1201 06:09:09.514606 137274321021824 utils.py:1231] [46000] uptime = 289138.876963595
+I1201 06:09:09.514685 137274321021824 utils.py:1231] [46000] examples_seen = 47104000.0
+I1201 06:09:09.514744 137274321021824 utils.py:1231] [46000] progress = 0.408514870829374
+I1201 06:09:09.514808 137274321021824 utils.py:1231] [46000] epoch = 36.76647931144027
+I1201 06:09:09.514889 137274321021824 utils.py:1231] [46000] img/sec/core = 164.21619830746755
+I1201 06:09:09.514956 137274321021824 utils.py:1231] [46000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.28203710975305
+I1201 06:09:09.515014 137274321021824 utils.py:1231] [46000] core_hours = 80.28203710975305
+I1201 06:09:09.515081 137274321021824 train.py:125] NOTE: Steps:46000/112603 [40.9%]
+Walltime:3d8h18m (0s eval)
+ETA:4d20h14m
+Total train time:8d4h31m
+I1201 06:14:21.549128 137274321021824 utils.py:1231] [46050] l2_params = 317.41540983997106
+I1201 06:14:21.549360 137274321021824 utils.py:1231] [46050] train/loss = 3.70112207531929
+I1201 06:14:21.549492 137274321021824 utils.py:1231] [46050] l2_grads = 1.3042383193969727
+I1201 06:14:21.549577 137274321021824 utils.py:1231] [46050] lr = 0.0007251114784509444
+I1201 06:14:21.549647 137274321021824 utils.py:1231] [46050] uptime = 289450.912009462
+I1201 06:14:21.549720 137274321021824 utils.py:1231] [46050] examples_seen = 47155200.0
+I1201 06:14:21.549777 137274321021824 utils.py:1231] [46050] progress = 0.4089589087324494
+I1201 06:14:21.549838 137274321021824 utils.py:1231] [46050] epoch = 36.80644287590923
+I1201 06:14:21.549896 137274321021824 utils.py:1231] [46050] img/sec/core = 164.0841331066973
+I1201 06:14:21.549956 137274321021824 utils.py:1231] [46050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.36871351138278
+I1201 06:14:21.550009 137274321021824 utils.py:1231] [46050] core_hours = 80.36871351138278
+I1201 06:14:21.550092 137274321021824 train.py:125] NOTE: Steps:46050/112603 [40.9%]
+Walltime:3d8h24m (0s eval)
+ETA:4d20h9m
+Total train time:8d4h31m
+I1201 06:19:33.324163 137274321021824 utils.py:1231] [46100] l2_params = 317.33419537277945
+I1201 06:19:33.324394 137274321021824 utils.py:1231] [46100] train/loss = 2.6797118484973907
+I1201 06:19:33.324501 137274321021824 utils.py:1231] [46100] l2_grads = 1.4508769512176514
+I1201 06:19:33.324591 137274321021824 utils.py:1231] [46100] lr = 0.0007244277117759437
+I1201 06:19:33.324661 137274321021824 utils.py:1231] [46100] uptime = 289762.68702306197
+I1201 06:19:33.324723 137274321021824 utils.py:1231] [46100] examples_seen = 47206400.0
+I1201 06:19:33.324775 137274321021824 utils.py:1231] [46100] progress = 0.4094029466355248
+I1201 06:19:33.324822 137274321021824 utils.py:1231] [46100] epoch = 36.84640644037819
+I1201 06:19:33.324875 137274321021824 utils.py:1231] [46100] img/sec/core = 164.22098553957966
+I1201 06:19:33.324938 137274321021824 utils.py:1231] [46100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.45531768182721
+I1201 06:19:33.324989 137274321021824 utils.py:1231] [46100] core_hours = 80.45531768182721
+I1201 06:19:33.325050 137274321021824 train.py:125] NOTE: Steps:46100/112603 [40.9%]
+Walltime:3d8h29m (0s eval)
+ETA:4d20h4m
+Total train time:8d4h31m
+I1201 06:24:45.110585 137274321021824 utils.py:1231] [46150] l2_params = 317.2587851509655
+I1201 06:24:45.110920 137274321021824 utils.py:1231] [46150] train/loss = 4.969826877117157
+I1201 06:24:45.111059 137274321021824 utils.py:1231] [46150] l2_grads = 1.2816908359527588
+I1201 06:24:45.111130 137274321021824 utils.py:1231] [46150] lr = 0.0007237434190884819
+I1201 06:24:45.111183 137274321021824 utils.py:1231] [46150] uptime = 290074.473545168
+I1201 06:24:45.111239 137274321021824 utils.py:1231] [46150] examples_seen = 47257600.0
+I1201 06:24:45.111288 137274321021824 utils.py:1231] [46150] progress = 0.4098469845386002
+I1201 06:24:45.111336 137274321021824 utils.py:1231] [46150] epoch = 36.886370004847144
+I1201 06:24:45.111387 137274321021824 utils.py:1231] [46150] img/sec/core = 164.21492389779286
+I1201 06:24:45.111441 137274321021824 utils.py:1231] [46150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.54192504907888
+I1201 06:24:45.111490 137274321021824 utils.py:1231] [46150] core_hours = 80.54192504907888
+I1201 06:24:45.111555 137274321021824 train.py:125] NOTE: Steps:46150/112603 [41.0%]
+Walltime:3d8h34m (0s eval)
+ETA:4d19h58m
+Total train time:8d4h31m
+I1201 06:29:56.884836 137274321021824 utils.py:1231] [46200] l2_params = 317.189275091682
+I1201 06:29:56.885128 137274321021824 utils.py:1231] [46200] train/loss = 3.453444629907608
+I1201 06:29:56.885265 137274321021824 utils.py:1231] [46200] l2_grads = 1.2459536790847778
+I1201 06:29:56.885325 137274321021824 utils.py:1231] [46200] lr = 0.0007230586019924008
+I1201 06:29:56.885383 137274321021824 utils.py:1231] [46200] uptime = 290386.247744233
+I1201 06:29:56.885436 137274321021824 utils.py:1231] [46200] examples_seen = 47308800.0
+I1201 06:29:56.885485 137274321021824 utils.py:1231] [46200] progress = 0.41029102244167565
+I1201 06:29:56.885534 137274321021824 utils.py:1231] [46200] epoch = 36.9263335693161
+I1201 06:29:56.885586 137274321021824 utils.py:1231] [46200] img/sec/core = 164.2214145799997
+I1201 06:29:56.885643 137274321021824 utils.py:1231] [46200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.62852899326361
+I1201 06:29:56.885692 137274321021824 utils.py:1231] [46200] core_hours = 80.62852899326361
+I1201 06:29:56.885756 137274321021824 train.py:125] NOTE: Steps:46200/112603 [41.0%]
+Walltime:3d8h39m (0s eval)
+ETA:4d19h53m
+Total train time:8d4h31m
+I1201 06:35:08.670748 137274321021824 utils.py:1231] [46250] l2_params = 317.1351513732078
+I1201 06:35:08.671029 137274321021824 utils.py:1231] [46250] train/loss = 2.451124370098114
+I1201 06:35:08.671150 137274321021824 utils.py:1231] [46250] l2_grads = 1.495671033859253
+I1201 06:35:08.671237 137274321021824 utils.py:1231] [46250] lr = 0.0007223732620927716
+I1201 06:35:08.900113 137274321021824 utils.py:1231] [46250] uptime = 290698.262426398
+I1201 06:35:08.900333 137274321021824 utils.py:1231] [46250] examples_seen = 47360000.0
+I1201 06:35:08.900394 137274321021824 utils.py:1231] [46250] progress = 0.41073506034475105
+I1201 06:35:08.900449 137274321021824 utils.py:1231] [46250] epoch = 36.966297133785055
+I1201 06:35:08.900514 137274321021824 utils.py:1231] [46250] img/sec/core = 164.09484209119594
+I1201 06:35:08.900574 137274321021824 utils.py:1231] [46250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.71519973830944
+I1201 06:35:08.900629 137274321021824 utils.py:1231] [46250] core_hours = 80.71519973830944
+I1201 06:35:08.900695 137274321021824 train.py:125] NOTE: Steps:46250/112603 [41.1%]
+Walltime:3d8h44m (0s eval)
+ETA:4d19h48m
+Total train time:8d4h31m
+I1201 06:40:20.679773 137274321021824 utils.py:1231] [46300] l2_params = 317.06084322501215
+I1201 06:40:20.679996 137274321021824 utils.py:1231] [46300] train/loss = 2.4020816683769226
+I1201 06:40:20.680159 137274321021824 utils.py:1231] [46300] l2_grads = 1.4909971952438354
+I1201 06:40:20.680261 137274321021824 utils.py:1231] [46300] lr = 0.0007216874009958894
+I1201 06:40:20.680355 137274321021824 utils.py:1231] [46300] uptime = 291010.042714239
+I1201 06:40:20.680420 137274321021824 utils.py:1231] [46300] examples_seen = 47411200.0
+I1201 06:40:20.680483 137274321021824 utils.py:1231] [46300] progress = 0.41117909824782645
+I1201 06:40:20.680546 137274321021824 utils.py:1231] [46300] epoch = 37.00626069825401
+I1201 06:40:20.680621 137274321021824 utils.py:1231] [46300] img/sec/core = 164.21820749011894
+I1201 06:40:20.680695 137274321021824 utils.py:1231] [46300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.80180537382083
+I1201 06:40:20.680777 137274321021824 utils.py:1231] [46300] core_hours = 80.80180537382083
+I1201 06:40:20.680850 137274321021824 train.py:125] NOTE: Steps:46300/112603 [41.1%]
+Walltime:3d8h50m (0s eval)
+ETA:4d19h42m
+Total train time:8d4h31m
+I1201 06:45:32.454990 137274321021824 utils.py:1231] [46350] l2_params = 317.0017887358899
+I1201 06:45:32.455211 137274321021824 utils.py:1231] [46350] train/loss = 2.478469282388687
+I1201 06:45:32.455313 137274321021824 utils.py:1231] [46350] l2_grads = 1.561098337173462
+I1201 06:45:32.455411 137274321021824 utils.py:1231] [46350] lr = 0.0007210010203092726
+I1201 06:45:32.455474 137274321021824 utils.py:1231] [46350] uptime = 291321.81783564197
+I1201 06:45:32.455523 137274321021824 utils.py:1231] [46350] examples_seen = 47462400.0
+I1201 06:45:32.455570 137274321021824 utils.py:1231] [46350] progress = 0.41162313615090185
+I1201 06:45:32.455615 137274321021824 utils.py:1231] [46350] epoch = 37.04622426272297
+I1201 06:45:32.455664 137274321021824 utils.py:1231] [46350] img/sec/core = 164.2209287566003
+I1201 06:45:32.455718 137274321021824 utils.py:1231] [46350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.88840957421054
+I1201 06:45:32.455773 137274321021824 utils.py:1231] [46350] core_hours = 80.88840957421054
+I1201 06:45:32.455830 137274321021824 train.py:125] NOTE: Steps:46350/112603 [41.2%]
+Walltime:3d8h55m (0s eval)
+ETA:4d19h37m
+Total train time:8d4h31m
+I1201 06:50:44.235950 137274321021824 utils.py:1231] [46400] l2_params = 316.94442614935167
+I1201 06:50:44.236178 137274321021824 utils.py:1231] [46400] train/loss = 3.3675553798675537
+I1201 06:50:44.236324 137274321021824 utils.py:1231] [46400] l2_grads = 1.2845145463943481
+I1201 06:50:44.236431 137274321021824 utils.py:1231] [46400] lr = 0.0007203141216416557
+I1201 06:50:44.236510 137274321021824 utils.py:1231] [46400] uptime = 291633.59886706097
+I1201 06:50:44.236591 137274321021824 utils.py:1231] [46400] examples_seen = 47513600.0
+I1201 06:50:44.236661 137274321021824 utils.py:1231] [46400] progress = 0.41206717405397725
+I1201 06:50:44.236731 137274321021824 utils.py:1231] [46400] epoch = 37.08618782719193
+I1201 06:50:44.236804 137274321021824 utils.py:1231] [46400] img/sec/core = 164.21781584009474
+I1201 06:50:44.236873 137274321021824 utils.py:1231] [46400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 80.97501541627138
+I1201 06:50:44.236948 137274321021824 utils.py:1231] [46400] core_hours = 80.97501541627138
+I1201 06:50:44.237022 137274321021824 train.py:125] NOTE: Steps:46400/112603 [41.2%]
+Walltime:3d9h0m (0s eval)
+ETA:4d19h32m
+Total train time:8d4h31m
+I1201 06:55:56.005909 137274321021824 utils.py:1231] [46450] l2_params = 316.84815976330253
+I1201 06:55:56.006130 137274321021824 utils.py:1231] [46450] train/loss = 3.9611760079860687
+I1201 06:55:56.006236 137274321021824 utils.py:1231] [46450] l2_grads = 1.2106415033340454
+I1201 06:55:56.006300 137274321021824 utils.py:1231] [46450] lr = 0.0007196267066029882
+I1201 06:55:56.006361 137274321021824 utils.py:1231] [46450] uptime = 291945.36872122897
+I1201 06:55:56.006423 137274321021824 utils.py:1231] [46450] examples_seen = 47564800.0
+I1201 06:55:56.006475 137274321021824 utils.py:1231] [46450] progress = 0.41251121195705265
+I1201 06:55:56.006527 137274321021824 utils.py:1231] [46450] epoch = 37.12615139166088
+I1201 06:55:56.006581 137274321021824 utils.py:1231] [46450] img/sec/core = 164.22370320772063
+I1201 06:55:56.006639 137274321021824 utils.py:1231] [46450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.06161815354027
+I1201 06:55:56.006691 137274321021824 utils.py:1231] [46450] core_hours = 81.06161815354027
+I1201 06:55:56.006753 137274321021824 train.py:125] NOTE: Steps:46450/112603 [41.3%]
+Walltime:3d9h5m (0s eval)
+ETA:4d19h27m
+Total train time:8d4h30m
+I1201 07:01:07.769728 137274321021824 utils.py:1231] [46500] l2_params = 316.7664206557912
+I1201 07:01:07.769961 137274321021824 utils.py:1231] [46500] train/loss = 3.008197605609894
+I1201 07:01:07.770090 137274321021824 utils.py:1231] [46500] l2_grads = 1.3167898654937744
+I1201 07:01:07.770172 137274321021824 utils.py:1231] [46500] lr = 0.0007189387768044304
+I1201 07:01:07.770256 137274321021824 utils.py:1231] [46500] uptime = 292257.13261751697
+I1201 07:01:07.770312 137274321021824 utils.py:1231] [46500] examples_seen = 47616000.0
+I1201 07:01:07.770366 137274321021824 utils.py:1231] [46500] progress = 0.41295524986012805
+I1201 07:01:07.770419 137274321021824 utils.py:1231] [46500] epoch = 37.16611495612984
+I1201 07:01:07.770473 137274321021824 utils.py:1231] [46500] img/sec/core = 164.22684156058438
+I1201 07:01:07.770531 137274321021824 utils.py:1231] [46500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.14821923584249
+I1201 07:01:07.770584 137274321021824 utils.py:1231] [46500] core_hours = 81.14821923584249
+I1201 07:01:07.770643 137274321021824 train.py:125] NOTE: Steps:46500/112603 [41.3%]
+Walltime:3d9h10m (0s eval)
+ETA:4d19h21m
+Total train time:8d4h30m
+I1201 07:06:19.549363 137274321021824 utils.py:1231] [46550] l2_params = 316.69197297499306
+I1201 07:06:19.549593 137274321021824 utils.py:1231] [46550] train/loss = 3.9613885283470154
+I1201 07:06:19.549740 137274321021824 utils.py:1231] [46550] l2_grads = 1.2509171962738037
+I1201 07:06:19.549852 137274321021824 utils.py:1231] [46550] lr = 0.0007182503338583484
+I1201 07:06:19.549950 137274321021824 utils.py:1231] [46550] uptime = 292568.912304219
+I1201 07:06:19.550046 137274321021824 utils.py:1231] [46550] examples_seen = 47667200.0
+I1201 07:06:19.550126 137274321021824 utils.py:1231] [46550] progress = 0.41339928776320345
+I1201 07:06:19.550210 137274321021824 utils.py:1231] [46550] epoch = 37.2060785205988
+I1201 07:06:19.550284 137274321021824 utils.py:1231] [46550] img/sec/core = 164.21852411743495
+I1201 07:06:19.550365 137274321021824 utils.py:1231] [46550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.23482470437082
+I1201 07:06:19.550434 137274321021824 utils.py:1231] [46550] core_hours = 81.23482470437082
+I1201 07:06:19.550515 137274321021824 train.py:125] NOTE: Steps:46550/112603 [41.3%]
+Walltime:3d9h16m (0s eval)
+ETA:4d19h16m
+Total train time:8d4h30m
+I1201 07:11:31.324908 137274321021824 utils.py:1231] [46600] l2_params = 316.6249800766586
+I1201 07:11:31.325169 137274321021824 utils.py:1231] [46600] train/loss = 2.4132902026176453
+I1201 07:11:31.325309 137274321021824 utils.py:1231] [46600] l2_grads = 1.419737696647644
+I1201 07:11:31.325397 137274321021824 utils.py:1231] [46600] lr = 0.0007175613793783111
+I1201 07:11:31.325472 137274321021824 utils.py:1231] [46600] uptime = 292880.68782914197
+I1201 07:11:31.325540 137274321021824 utils.py:1231] [46600] examples_seen = 47718400.0
+I1201 07:11:31.325611 137274321021824 utils.py:1231] [46600] progress = 0.41384332566627885
+I1201 07:11:31.325678 137274321021824 utils.py:1231] [46600] epoch = 37.24604208506776
+I1201 07:11:31.325737 137274321021824 utils.py:1231] [46600] img/sec/core = 164.2207162112731
+I1201 07:11:31.325799 137274321021824 utils.py:1231] [46600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.32142901684944
+I1201 07:11:31.325852 137274321021824 utils.py:1231] [46600] core_hours = 81.32142901684944
+I1201 07:11:31.325923 137274321021824 train.py:125] NOTE: Steps:46600/112603 [41.4%]
+Walltime:3d9h21m (0s eval)
+ETA:4d19h11m
+Total train time:8d4h30m
+I1201 07:16:43.232218 137274321021824 utils.py:1231] [46650] l2_params = 316.57935468262514
+I1201 07:16:43.232435 137274321021824 utils.py:1231] [46650] train/loss = 5.05195164680481
+I1201 07:16:43.232531 137274321021824 utils.py:1231] [46650] l2_grads = 1.2817027568817139
+I1201 07:16:43.232591 137274321021824 utils.py:1231] [46650] lr = 0.0007168719149790859
+I1201 07:16:43.232646 137274321021824 utils.py:1231] [46650] uptime = 293192.595007994
+I1201 07:16:43.232699 137274321021824 utils.py:1231] [46650] examples_seen = 47769600.0
+I1201 07:16:43.232749 137274321021824 utils.py:1231] [46650] progress = 0.4142873635693543
+I1201 07:16:43.232798 137274321021824 utils.py:1231] [46650] epoch = 37.28600564953671
+I1201 07:16:43.232848 137274321021824 utils.py:1231] [46650] img/sec/core = 164.15139974797754
+I1201 07:16:43.232914 137274321021824 utils.py:1231] [46650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.40806989986388
+I1201 07:16:43.232964 137274321021824 utils.py:1231] [46650] core_hours = 81.40806989986388
+I1201 07:16:43.233025 137274321021824 train.py:125] NOTE: Steps:46650/112603 [41.4%]
+Walltime:3d9h26m (0s eval)
+ETA:4d19h5m
+Total train time:8d4h30m
+I1201 07:21:55.004257 137274321021824 utils.py:1231] [46700] l2_params = 316.5080924801813
+I1201 07:21:55.004464 137274321021824 utils.py:1231] [46700] train/loss = 3.264208495616913
+I1201 07:21:55.004570 137274321021824 utils.py:1231] [46700] l2_grads = 1.3134024143218994
+I1201 07:21:55.004656 137274321021824 utils.py:1231] [46700] lr = 0.0007161819422766359
+I1201 07:21:55.004733 137274321021824 utils.py:1231] [46700] uptime = 293504.367093709
+I1201 07:21:55.004817 137274321021824 utils.py:1231] [46700] examples_seen = 47820800.0
+I1201 07:21:55.004865 137274321021824 utils.py:1231] [46700] progress = 0.4147314014724297
+I1201 07:21:55.004923 137274321021824 utils.py:1231] [46700] epoch = 37.32596921400567
+I1201 07:21:55.004974 137274321021824 utils.py:1231] [46700] img/sec/core = 164.22252775637835
+I1201 07:21:55.005028 137274321021824 utils.py:1231] [46700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.49467325700695
+I1201 07:21:55.005079 137274321021824 utils.py:1231] [46700] core_hours = 81.49467325700695
+I1201 07:21:55.005137 137274321021824 train.py:125] NOTE: Steps:46700/112603 [41.5%]
+Walltime:3d9h31m (0s eval)
+ETA:4d19h0m
+Total train time:8d4h30m
+I1201 07:27:06.771090 137274321021824 utils.py:1231] [46750] l2_params = 316.4434175941985
+I1201 07:27:06.771361 137274321021824 utils.py:1231] [46750] train/loss = 3.412907689809799
+I1201 07:27:06.771488 137274321021824 utils.py:1231] [46750] l2_grads = 1.2863526344299316
+I1201 07:27:06.771606 137274321021824 utils.py:1231] [46750] lr = 0.0007154914628881157
+I1201 07:27:06.771692 137274321021824 utils.py:1231] [46750] uptime = 293816.13404841
+I1201 07:27:06.771774 137274321021824 utils.py:1231] [46750] examples_seen = 47872000.0
+I1201 07:27:06.771833 137274321021824 utils.py:1231] [46750] progress = 0.4151754393755051
+I1201 07:27:06.771890 137274321021824 utils.py:1231] [46750] epoch = 37.36593277847462
+I1201 07:27:06.771950 137274321021824 utils.py:1231] [46750] img/sec/core = 164.2252305062465
+I1201 07:27:06.772004 137274321021824 utils.py:1231] [46750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.58127518886833
+I1201 07:27:06.772055 137274321021824 utils.py:1231] [46750] core_hours = 81.58127518886833
+I1201 07:27:06.772119 137274321021824 train.py:125] NOTE: Steps:46750/112603 [41.5%]
+Walltime:3d9h36m (0s eval)
+ETA:4d18h55m
+Total train time:8d4h30m
+I1201 07:32:18.545756 137274321021824 utils.py:1231] [46800] l2_params = 316.37498322997016
+I1201 07:32:18.545994 137274321021824 utils.py:1231] [46800] train/loss = 2.4455469250679016
+I1201 07:32:18.546104 137274321021824 utils.py:1231] [46800] l2_grads = 1.4675904512405396
+I1201 07:32:18.546174 137274321021824 utils.py:1231] [46800] lr = 0.0007148004784318665
+I1201 07:32:18.546236 137274321021824 utils.py:1231] [46800] uptime = 294127.908597578
+I1201 07:32:18.546312 137274321021824 utils.py:1231] [46800] examples_seen = 47923200.0
+I1201 07:32:18.546371 137274321021824 utils.py:1231] [46800] progress = 0.4156194772785805
+I1201 07:32:18.546427 137274321021824 utils.py:1231] [46800] epoch = 37.405896342943585
+I1201 07:32:18.546485 137274321021824 utils.py:1231] [46800] img/sec/core = 164.2212301697823
+I1201 07:32:18.546548 137274321021824 utils.py:1231] [46800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.6678792303039
+I1201 07:32:18.546605 137274321021824 utils.py:1231] [46800] core_hours = 81.6678792303039
+I1201 07:32:18.546673 137274321021824 train.py:125] NOTE: Steps:46800/112603 [41.6%]
+Walltime:3d9h42m (0s eval)
+ETA:4d18h50m
+Total train time:8d4h30m
+I1201 07:37:30.314129 137274321021824 utils.py:1231] [46850] l2_params = 316.3012153601121
+I1201 07:37:30.314346 137274321021824 utils.py:1231] [46850] train/loss = 2.557862490415573
+I1201 07:37:30.314447 137274321021824 utils.py:1231] [46850] l2_grads = 1.5346544981002808
+I1201 07:37:30.314517 137274321021824 utils.py:1231] [46850] lr = 0.0007141089905274145
+I1201 07:37:30.314585 137274321021824 utils.py:1231] [46850] uptime = 294439.676946555
+I1201 07:37:30.314648 137274321021824 utils.py:1231] [46850] examples_seen = 47974400.0
+I1201 07:37:30.314721 137274321021824 utils.py:1231] [46850] progress = 0.4160635151816559
+I1201 07:37:30.314782 137274321021824 utils.py:1231] [46850] epoch = 37.44585990741254
+I1201 07:37:30.314842 137274321021824 utils.py:1231] [46850] img/sec/core = 164.2244960657701
+I1201 07:37:30.314922 137274321021824 utils.py:1231] [46850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.75448154946416
+I1201 07:37:30.314983 137274321021824 utils.py:1231] [46850] core_hours = 81.75448154946416
+I1201 07:37:30.315064 137274321021824 train.py:125] NOTE: Steps:46850/112603 [41.6%]
+Walltime:3d9h47m (0s eval)
+ETA:4d18h44m
+Total train time:8d4h30m
+I1201 07:42:42.094650 137274321021824 utils.py:1231] [46900] l2_params = 316.212436590853
+I1201 07:42:42.094907 137274321021824 utils.py:1231] [46900] train/loss = 4.123679012060165
+I1201 07:42:42.095033 137274321021824 utils.py:1231] [46900] l2_grads = 1.2411452531814575
+I1201 07:42:42.095120 137274321021824 utils.py:1231] [46900] lr = 0.0007134170007954645
+I1201 07:42:42.095201 137274321021824 utils.py:1231] [46900] uptime = 294751.457563242
+I1201 07:42:42.095261 137274321021824 utils.py:1231] [46900] examples_seen = 48025600.0
+I1201 07:42:42.095317 137274321021824 utils.py:1231] [46900] progress = 0.4165075530847313
+I1201 07:42:42.095376 137274321021824 utils.py:1231] [46900] epoch = 37.485823471881496
+I1201 07:42:42.095465 137274321021824 utils.py:1231] [46900] img/sec/core = 164.2180342833677
+I1201 07:42:42.095526 137274321021824 utils.py:1231] [46900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.84108727632166
+I1201 07:42:42.095579 137274321021824 utils.py:1231] [46900] core_hours = 81.84108727632166
+I1201 07:42:42.095643 137274321021824 train.py:125] NOTE: Steps:46900/112603 [41.7%]
+Walltime:3d9h52m (0s eval)
+ETA:4d18h39m
+Total train time:8d4h30m
+I1201 07:47:53.873197 137274321021824 utils.py:1231] [46950] l2_params = 316.1328177555432
+I1201 07:47:53.873397 137274321021824 utils.py:1231] [46950] train/loss = 2.4997394680976868
+I1201 07:47:53.873488 137274321021824 utils.py:1231] [46950] l2_grads = 1.4734059572219849
+I1201 07:47:53.873543 137274321021824 utils.py:1231] [46950] lr = 0.0007127245108578991
+I1201 07:47:53.873592 137274321021824 utils.py:1231] [46950] uptime = 295063.235954359
+I1201 07:47:53.873646 137274321021824 utils.py:1231] [46950] examples_seen = 48076800.0
+I1201 07:47:53.873693 137274321021824 utils.py:1231] [46950] progress = 0.4169515909878067
+I1201 07:47:53.873739 137274321021824 utils.py:1231] [46950] epoch = 37.52578703635045
+I1201 07:47:53.873789 137274321021824 utils.py:1231] [46950] img/sec/core = 164.21920652219546
+I1201 07:47:53.873841 137274321021824 utils.py:1231] [46950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 81.92769238496527
+I1201 07:47:53.873890 137274321021824 utils.py:1231] [46950] core_hours = 81.92769238496527
+I1201 07:47:53.873955 137274321021824 train.py:125] NOTE: Steps:46950/112603 [41.7%]
+Walltime:3d9h57m (0s eval)
+ETA:4d18h34m
+Total train time:8d4h30m
+I1201 07:53:05.652728 137274321021824 utils.py:1231] [47000] l2_params = 316.0699479124346
+I1201 07:53:05.652970 137274321021824 utils.py:1231] [47000] train/loss = 2.5326513051986694
+I1201 07:53:05.653070 137274321021824 utils.py:1231] [47000] l2_grads = 1.4063340425491333
+I1201 07:53:05.653139 137274321021824 utils.py:1231] [47000] lr = 0.0007120315223377724
+I1201 07:53:05.653197 137274321021824 utils.py:1231] [47000] uptime = 295375.015558904
+I1201 07:53:05.653264 137274321021824 utils.py:1231] [47000] examples_seen = 48128000.0
+I1201 07:53:05.653325 137274321021824 utils.py:1231] [47000] progress = 0.4173956288908821
+I1201 07:53:05.653379 137274321021824 utils.py:1231] [47000] epoch = 37.565750600819406
+I1201 07:53:05.653436 137274321021824 utils.py:1231] [47000] img/sec/core = 164.21856739065024
+I1201 07:53:05.653501 137274321021824 utils.py:1231] [47000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.0142978306722
+I1201 07:53:05.653555 137274321021824 utils.py:1231] [47000] core_hours = 82.0142978306722
+I1201 07:53:05.653622 137274321021824 train.py:125] NOTE: Steps:47000/112603 [41.7%]
+Walltime:3d10h2m (0s eval)
+ETA:4d18h28m
+Total train time:8d4h29m
+I1201 07:58:17.611457 137274321021824 utils.py:1231] [47050] l2_params = 316.0249226741215
+I1201 07:58:17.611721 137274321021824 utils.py:1231] [47050] train/loss = 2.4597581028938293
+I1201 07:58:17.611832 137274321021824 utils.py:1231] [47050] l2_grads = 1.559139609336853
+I1201 07:58:17.611921 137274321021824 utils.py:1231] [47050] lr = 0.0007113380368593069
+I1201 07:58:17.611976 137274321021824 utils.py:1231] [47050] uptime = 295686.974336979
+I1201 07:58:17.612030 137274321021824 utils.py:1231] [47050] examples_seen = 48179200.0
+I1201 07:58:17.612082 137274321021824 utils.py:1231] [47050] progress = 0.4178396667939575
+I1201 07:58:17.612129 137274321021824 utils.py:1231] [47050] epoch = 37.60571416528837
+I1201 07:58:17.612181 137274321021824 utils.py:1231] [47050] img/sec/core = 164.12424845339842
+I1201 07:58:17.612237 137274321021824 utils.py:1231] [47050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.10095304680415
+I1201 07:58:17.612285 137274321021824 utils.py:1231] [47050] core_hours = 82.10095304680415
+I1201 07:58:17.612346 137274321021824 train.py:125] NOTE: Steps:47050/112603 [41.8%]
+Walltime:3d10h8m (0s eval)
+ETA:4d18h23m
+Total train time:8d4h29m
+I1201 08:03:29.386325 137274321021824 utils.py:1231] [47100] l2_params = 315.95316253414
+I1201 08:03:29.386628 137274321021824 utils.py:1231] [47100] train/loss = 2.735156923532486
+I1201 08:03:29.386817 137274321021824 utils.py:1231] [47100] l2_grads = 1.493613362312317
+I1201 08:03:29.386914 137274321021824 utils.py:1231] [47100] lr = 0.0007106440560478909
+I1201 08:03:29.386976 137274321021824 utils.py:1231] [47100] uptime = 295998.749337267
+I1201 08:03:29.387037 137274321021824 utils.py:1231] [47100] examples_seen = 48230400.0
+I1201 08:03:29.387095 137274321021824 utils.py:1231] [47100] progress = 0.41828370469703297
+I1201 08:03:29.387152 137274321021824 utils.py:1231] [47100] epoch = 37.645677729757324
+I1201 08:03:29.387216 137274321021824 utils.py:1231] [47100] img/sec/core = 164.22099255135478
+I1201 08:03:29.387284 137274321021824 utils.py:1231] [47100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.18755721355085
+I1201 08:03:29.387361 137274321021824 utils.py:1231] [47100] core_hours = 82.18755721355085
+I1201 08:03:29.387478 137274321021824 train.py:125] NOTE: Steps:47100/112603 [41.8%]
+Walltime:3d10h13m (0s eval)
+ETA:4d18h18m
+Total train time:8d4h29m
+I1201 08:08:41.153607 137274321021824 utils.py:1231] [47150] l2_params = 315.87256347594354
+I1201 08:08:41.153807 137274321021824 utils.py:1231] [47150] train/loss = 2.458952561020851
+I1201 08:08:41.153925 137274321021824 utils.py:1231] [47150] l2_grads = 1.5664713382720947
+I1201 08:08:41.153999 137274321021824 utils.py:1231] [47150] lr = 0.000709949581530072
+I1201 08:08:41.154061 137274321021824 utils.py:1231] [47150] uptime = 296310.516422529
+I1201 08:08:41.154121 137274321021824 utils.py:1231] [47150] examples_seen = 48281600.0
+I1201 08:08:41.154184 137274321021824 utils.py:1231] [47150] progress = 0.41872774260010837
+I1201 08:08:41.154239 137274321021824 utils.py:1231] [47150] epoch = 37.68564129422628
+I1201 08:08:41.154297 137274321021824 utils.py:1231] [47150] img/sec/core = 164.22516173243335
+I1201 08:08:41.154360 137274321021824 utils.py:1231] [47150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.27415918167917
+I1201 08:08:41.154418 137274321021824 utils.py:1231] [47150] core_hours = 82.27415918167917
+I1201 08:08:41.154484 137274321021824 train.py:125] NOTE: Steps:47150/112603 [41.9%]
+Walltime:3d10h18m (0s eval)
+ETA:4d18h13m
+Total train time:8d4h29m
+I1201 08:13:52.932543 137274321021824 utils.py:1231] [47200] l2_params = 315.8129229688086
+I1201 08:13:52.932763 137274321021824 utils.py:1231] [47200] train/loss = 2.4535407423973083
+I1201 08:13:52.932891 137274321021824 utils.py:1231] [47200] l2_grads = 1.5411553382873535
+I1201 08:13:52.932980 137274321021824 utils.py:1231] [47200] lr = 0.0007092546149335559
+I1201 08:13:52.933038 137274321021824 utils.py:1231] [47200] uptime = 296622.295400867
+I1201 08:13:52.933099 137274321021824 utils.py:1231] [47200] examples_seen = 48332800.0
+I1201 08:13:52.933151 137274321021824 utils.py:1231] [47200] progress = 0.41917178050318377
+I1201 08:13:52.933204 137274321021824 utils.py:1231] [47200] epoch = 37.725604858695235
+I1201 08:13:52.933259 137274321021824 utils.py:1231] [47200] img/sec/core = 164.21889722306454
+I1201 08:13:52.933315 137274321021824 utils.py:1231] [47200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.36076445343973
+I1201 08:13:52.933364 137274321021824 utils.py:1231] [47200] core_hours = 82.36076445343973
+I1201 08:13:52.933423 137274321021824 train.py:125] NOTE: Steps:47200/112603 [41.9%]
+Walltime:3d10h23m (0s eval)
+ETA:4d18h7m
+Total train time:8d4h29m
+I1201 08:19:04.714236 137274321021824 utils.py:1231] [47250] l2_params = 315.75526566637706
+I1201 08:19:04.714450 137274321021824 utils.py:1231] [47250] train/loss = 3.157207876443863
+I1201 08:19:04.714554 137274321021824 utils.py:1231] [47250] l2_grads = 1.270267367362976
+I1201 08:19:04.714627 137274321021824 utils.py:1231] [47250] lr = 0.0007085591578872015
+I1201 08:19:04.714676 137274321021824 utils.py:1231] [47250] uptime = 296934.077039056
+I1201 08:19:04.714742 137274321021824 utils.py:1231] [47250] examples_seen = 48384000.0
+I1201 08:19:04.714788 137274321021824 utils.py:1231] [47250] progress = 0.41961581840625917
+I1201 08:19:04.714834 137274321021824 utils.py:1231] [47250] epoch = 37.76556842316419
+I1201 08:19:04.714886 137274321021824 utils.py:1231] [47250] img/sec/core = 164.21749624961933
+I1201 08:19:04.714940 137274321021824 utils.py:1231] [47250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.44737046404776
+I1201 08:19:04.714990 137274321021824 utils.py:1231] [47250] core_hours = 82.44737046404776
+I1201 08:19:04.715053 137274321021824 train.py:125] NOTE: Steps:47250/112603 [42.0%]
+Walltime:3d10h28m (0s eval)
+ETA:4d18h2m
+Total train time:8d4h29m
+I1201 08:24:16.484423 137274321021824 utils.py:1231] [47300] l2_params = 315.67527347412596
+I1201 08:24:16.484645 137274321021824 utils.py:1231] [47300] train/loss = 4.877057790756226
+I1201 08:24:16.484740 137274321021824 utils.py:1231] [47300] l2_grads = 1.4316717386245728
+I1201 08:24:16.484808 137274321021824 utils.py:1231] [47300] lr = 0.0007078632120210179
+I1201 08:24:16.484859 137274321021824 utils.py:1231] [47300] uptime = 297245.847220682
+I1201 08:24:16.484917 137274321021824 utils.py:1231] [47300] examples_seen = 48435200.0
+I1201 08:24:16.484966 137274321021824 utils.py:1231] [47300] progress = 0.42005985630933457
+I1201 08:24:16.485014 137274321021824 utils.py:1231] [47300] epoch = 37.80553198763315
+I1201 08:24:16.485067 137274321021824 utils.py:1231] [47300] img/sec/core = 164.22353072050933
+I1201 08:24:16.485123 137274321021824 utils.py:1231] [47300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.53397329227722
+I1201 08:24:16.485174 137274321021824 utils.py:1231] [47300] core_hours = 82.53397329227722
+I1201 08:24:16.485235 137274321021824 train.py:125] NOTE: Steps:47300/112603 [42.0%]
+Walltime:3d10h34m (0s eval)
+ETA:4d17h57m
+Total train time:8d4h29m
+I1201 08:29:28.261557 137274321021824 utils.py:1231] [47350] l2_params = 315.6061635806078
+I1201 08:29:28.261789 137274321021824 utils.py:1231] [47350] train/loss = 2.314809024333954
+I1201 08:29:28.261920 137274321021824 utils.py:1231] [47350] l2_grads = 1.4436458349227905
+I1201 08:29:28.262002 137274321021824 utils.py:1231] [47350] lr = 0.0007071667789661591
+I1201 08:29:28.262078 137274321021824 utils.py:1231] [47350] uptime = 297557.62443649396
+I1201 08:29:28.262147 137274321021824 utils.py:1231] [47350] examples_seen = 48486400.0
+I1201 08:29:28.262212 137274321021824 utils.py:1231] [47350] progress = 0.42050389421240997
+I1201 08:29:28.262293 137274321021824 utils.py:1231] [47350] epoch = 37.84549555210211
+I1201 08:29:28.262361 137274321021824 utils.py:1231] [47350] img/sec/core = 164.21982557853713
+I1201 08:29:28.262428 137274321021824 utils.py:1231] [47350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.62057807444721
+I1201 08:29:28.262507 137274321021824 utils.py:1231] [47350] core_hours = 82.62057807444721
+I1201 08:29:28.262588 137274321021824 train.py:125] NOTE: Steps:47350/112603 [42.1%]
+Walltime:3d10h39m (0s eval)
+ETA:4d17h51m
+Total train time:8d4h29m
+I1201 08:34:40.032791 137274321021824 utils.py:1231] [47400] l2_params = 315.53354746371355
+I1201 08:34:40.033025 137274321021824 utils.py:1231] [47400] train/loss = 2.430378347635269
+I1201 08:34:40.033132 137274321021824 utils.py:1231] [47400] l2_grads = 1.4800453186035156
+I1201 08:34:40.033202 137274321021824 utils.py:1231] [47400] lr = 0.0007064698603549206
+I1201 08:34:40.033282 137274321021824 utils.py:1231] [47400] uptime = 297869.395640683
+I1201 08:34:40.033360 137274321021824 utils.py:1231] [47400] examples_seen = 48537600.0
+I1201 08:34:40.033432 137274321021824 utils.py:1231] [47400] progress = 0.42094793211548537
+I1201 08:34:40.033495 137274321021824 utils.py:1231] [47400] epoch = 37.88545911657106
+I1201 08:34:40.033553 137274321021824 utils.py:1231] [47400] img/sec/core = 164.22299209183524
+I1201 08:34:40.033614 137274321021824 utils.py:1231] [47400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.70718118672195
+I1201 08:34:40.033676 137274321021824 utils.py:1231] [47400] core_hours = 82.70718118672195
+I1201 08:34:40.033760 137274321021824 train.py:125] NOTE: Steps:47400/112603 [42.1%]
+Walltime:3d10h44m (0s eval)
+ETA:4d17h46m
+Total train time:8d4h29m
+I1201 08:39:51.820872 137274321021824 utils.py:1231] [47450] l2_params = 315.4600950502062
+I1201 08:39:51.821133 137274321021824 utils.py:1231] [47450] train/loss = 3.5915231704711914
+I1201 08:39:51.821243 137274321021824 utils.py:1231] [47450] l2_grads = 1.3099015951156616
+I1201 08:39:51.821312 137274321021824 utils.py:1231] [47450] lr = 0.000705772457820737
+I1201 08:39:51.821376 137274321021824 utils.py:1231] [47450] uptime = 298181.183737207
+I1201 08:39:51.821434 137274321021824 utils.py:1231] [47450] examples_seen = 48588800.0
+I1201 08:39:51.821491 137274321021824 utils.py:1231] [47450] progress = 0.42139197001856077
+I1201 08:39:51.821544 137274321021824 utils.py:1231] [47450] epoch = 37.92542268104002
+I1201 08:39:51.821603 137274321021824 utils.py:1231] [47450] img/sec/core = 164.2140946713962
+I1201 08:39:51.821664 137274321021824 utils.py:1231] [47450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.79378899131193
+I1201 08:39:51.821718 137274321021824 utils.py:1231] [47450] core_hours = 82.79378899131193
+I1201 08:39:51.821783 137274321021824 train.py:125] NOTE: Steps:47450/112603 [42.1%]
+Walltime:3d10h49m (0s eval)
+ETA:4d17h41m
+Total train time:8d4h29m
+I1201 08:45:03.587932 137274321021824 utils.py:1231] [47500] l2_params = 315.390497977926
+I1201 08:45:03.588153 137274321021824 utils.py:1231] [47500] train/loss = 4.010518431663513
+I1201 08:45:03.588281 137274321021824 utils.py:1231] [47500] l2_grads = 1.3287179470062256
+I1201 08:45:03.588358 137274321021824 utils.py:1231] [47500] lr = 0.0007050745729981757
+I1201 08:45:03.588415 137274321021824 utils.py:1231] [47500] uptime = 298492.950776531
+I1201 08:45:03.588470 137274321021824 utils.py:1231] [47500] examples_seen = 48640000.0
+I1201 08:45:03.588517 137274321021824 utils.py:1231] [47500] progress = 0.42183600792163617
+I1201 08:45:03.588565 137274321021824 utils.py:1231] [47500] epoch = 37.96538624550898
+I1201 08:45:03.588615 137274321021824 utils.py:1231] [47500] img/sec/core = 164.22518593052547
+I1201 08:45:03.588670 137274321021824 utils.py:1231] [47500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.88039094667973
+I1201 08:45:03.588719 137274321021824 utils.py:1231] [47500] core_hours = 82.88039094667973
+I1201 08:45:03.588777 137274321021824 train.py:125] NOTE: Steps:47500/112603 [42.2%]
+Walltime:3d10h54m (0s eval)
+ETA:4d17h35m
+Total train time:8d4h29m
+I1201 08:45:03.588870 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:47500/112603 [42.2%]
+Walltime:3d10h54m (0s eval)
+ETA:4d17h35m
+Total train time:8d4h29m
+I1201 08:46:39.398530 137274321021824 utils.py:1231] [47500] val/acc@1 = 0.6411431760204082
+I1201 08:46:39.398811 137274321021824 utils.py:1231] [47500] val/loss = 1.5237073700646966
+I1201 08:46:39.398979 137274321021824 utils.py:1231] [47500] z/secs/eval/val = 95.81003068597056
+I1201 08:46:39.399058 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 95.81003068597056
+I1201 08:51:49.429159 137274321021824 utils.py:1231] [47550] l2_params = 315.2984087769811
+I1201 08:51:49.429372 137274321021824 utils.py:1231] [47550] train/loss = 2.326855316758156
+I1201 08:51:49.429467 137274321021824 utils.py:1231] [47550] l2_grads = 1.4874560832977295
+I1201 08:51:49.429536 137274321021824 utils.py:1231] [47550] lr = 0.0007043762075229363
+I1201 08:51:49.429604 137274321021824 utils.py:1231] [47550] uptime = 298898.79196615
+I1201 08:51:49.429673 137274321021824 utils.py:1231] [47550] examples_seen = 48691200.0
+I1201 08:51:49.429731 137274321021824 utils.py:1231] [47550] progress = 0.4222800458247116
+I1201 08:51:49.429785 137274321021824 utils.py:1231] [47550] epoch = 38.005349809977936
+I1201 08:51:49.429844 137274321021824 utils.py:1231] [47550] img/sec/core = 126.15772205888891
+I1201 08:51:49.429911 137274321021824 utils.py:1231] [47550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 82.99312461046277
+I1201 08:51:49.429969 137274321021824 utils.py:1231] [47550] core_hours = 82.99312461046277
+I1201 08:51:49.430061 137274321021824 train.py:125] NOTE: Steps:47550/112603 [42.2%]
+Walltime:3d11h1m (0s eval)
+ETA:4d17h32m
+Total train time:8d4h32m
+I1201 08:57:01.215261 137274321021824 utils.py:1231] [47600] l2_params = 315.2183649390342
+I1201 08:57:01.215546 137274321021824 utils.py:1231] [47600] train/loss = 2.6583633422851562
+I1201 08:57:01.215714 137274321021824 utils.py:1231] [47600] l2_grads = 1.459077000617981
+I1201 08:57:01.215812 137274321021824 utils.py:1231] [47600] lr = 0.0007036773630318435
+I1201 08:57:01.215917 137274321021824 utils.py:1231] [47600] uptime = 299210.578277454
+I1201 08:57:01.215989 137274321021824 utils.py:1231] [47600] examples_seen = 48742400.0
+I1201 08:57:01.216053 137274321021824 utils.py:1231] [47600] progress = 0.422724083727787
+I1201 08:57:01.216114 137274321021824 utils.py:1231] [47600] epoch = 38.04531337444689
+I1201 08:57:01.216175 137274321021824 utils.py:1231] [47600] img/sec/core = 164.21503492523442
+I1201 08:57:01.216242 137274321021824 utils.py:1231] [47600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.07973191915833
+I1201 08:57:01.216302 137274321021824 utils.py:1231] [47600] core_hours = 83.07973191915833
+I1201 08:57:01.216369 137274321021824 train.py:125] NOTE: Steps:47600/112603 [42.3%]
+Walltime:3d11h6m (0s eval)
+ETA:4d17h27m
+Total train time:8d4h32m
+I1201 09:02:12.992563 137274321021824 utils.py:1231] [47650] l2_params = 315.135406605838
+I1201 09:02:12.992800 137274321021824 utils.py:1231] [47650] train/loss = 2.5884416103363037
+I1201 09:02:12.992922 137274321021824 utils.py:1231] [47650] l2_grads = 1.418962836265564
+I1201 09:02:12.992985 137274321021824 utils.py:1231] [47650] lr = 0.000702978041162846
+I1201 09:02:12.993044 137274321021824 utils.py:1231] [47650] uptime = 299522.355400855
+I1201 09:02:12.993093 137274321021824 utils.py:1231] [47650] examples_seen = 48793600.0
+I1201 09:02:12.993140 137274321021824 utils.py:1231] [47650] progress = 0.4231681216308624
+I1201 09:02:12.993187 137274321021824 utils.py:1231] [47650] epoch = 38.08527693891585
+I1201 09:02:12.993236 137274321021824 utils.py:1231] [47650] img/sec/core = 164.2198742534034
+I1201 09:02:12.993292 137274321021824 utils.py:1231] [47650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.16633667565861
+I1201 09:02:12.993341 137274321021824 utils.py:1231] [47650] core_hours = 83.16633667565861
+I1201 09:02:12.993407 137274321021824 train.py:125] NOTE: Steps:47650/112603 [42.3%]
+Walltime:3d11h12m (0s eval)
+ETA:4d17h22m
+Total train time:8d4h32m
+I1201 09:07:24.762906 137274321021824 utils.py:1231] [47700] l2_params = 315.07600957007577
+I1201 09:07:24.763114 137274321021824 utils.py:1231] [47700] train/loss = 2.467342495918274
+I1201 09:07:24.763211 137274321021824 utils.py:1231] [47700] l2_grads = 1.5476797819137573
+I1201 09:07:24.763278 137274321021824 utils.py:1231] [47700] lr = 0.0007022782435550099
+I1201 09:07:24.763329 137274321021824 utils.py:1231] [47700] uptime = 299834.125691321
+I1201 09:07:24.763381 137274321021824 utils.py:1231] [47700] examples_seen = 48844800.0
+I1201 09:07:24.763429 137274321021824 utils.py:1231] [47700] progress = 0.4236121595339378
+I1201 09:07:24.763477 137274321021824 utils.py:1231] [47700] epoch = 38.1252405033848
+I1201 09:07:24.763526 137274321021824 utils.py:1231] [47700] img/sec/core = 164.22347338958193
+I1201 09:07:24.763586 137274321021824 utils.py:1231] [47700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.25293953412138
+I1201 09:07:24.763647 137274321021824 utils.py:1231] [47700] core_hours = 83.25293953412138
+I1201 09:07:24.763705 137274321021824 train.py:125] NOTE: Steps:47700/112603 [42.4%]
+Walltime:3d11h17m (0s eval)
+ETA:4d17h16m
+Total train time:8d4h32m
+I1201 09:12:36.512113 137274321021824 utils.py:1231] [47750] l2_params = 315.0229510416678
+I1201 09:12:36.512316 137274321021824 utils.py:1231] [47750] train/loss = 3.7705387473106384
+I1201 09:12:36.512417 137274321021824 utils.py:1231] [47750] l2_grads = 1.2418360710144043
+I1201 09:12:36.512508 137274321021824 utils.py:1231] [47750] lr = 0.0007015779718485177
+I1201 09:12:36.512564 137274321021824 utils.py:1231] [47750] uptime = 300145.874925635
+I1201 09:12:36.512621 137274321021824 utils.py:1231] [47750] examples_seen = 48896000.0
+I1201 09:12:36.512682 137274321021824 utils.py:1231] [47750] progress = 0.4240561974370132
+I1201 09:12:36.512733 137274321021824 utils.py:1231] [47750] epoch = 38.165204067853765
+I1201 09:12:36.512812 137274321021824 utils.py:1231] [47750] img/sec/core = 164.23456536360672
+I1201 09:12:36.512869 137274321021824 utils.py:1231] [47750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.33953654365305
+I1201 09:12:36.512964 137274321021824 utils.py:1231] [47750] core_hours = 83.33953654365305
+I1201 09:12:36.513066 137274321021824 train.py:125] NOTE: Steps:47750/112603 [42.4%]
+Walltime:3d11h22m (0s eval)
+ETA:4d17h11m
+Total train time:8d4h32m
+I1201 09:17:48.273760 137274321021824 utils.py:1231] [47800] l2_params = 314.9746637845139
+I1201 09:17:48.274032 137274321021824 utils.py:1231] [47800] train/loss = 3.5514771044254303
+I1201 09:17:48.274152 137274321021824 utils.py:1231] [47800] l2_grads = 1.3169316053390503
+I1201 09:17:48.274234 137274321021824 utils.py:1231] [47800] lr = 0.0007008772276846621
+I1201 09:17:48.274301 137274321021824 utils.py:1231] [47800] uptime = 300457.63665783
+I1201 09:17:48.274358 137274321021824 utils.py:1231] [47800] examples_seen = 48947200.0
+I1201 09:17:48.274417 137274321021824 utils.py:1231] [47800] progress = 0.4245002353400886
+I1201 09:17:48.274475 137274321021824 utils.py:1231] [47800] epoch = 38.20516763232272
+I1201 09:17:48.274529 137274321021824 utils.py:1231] [47800] img/sec/core = 164.22798154065555
+I1201 09:17:48.274585 137274321021824 utils.py:1231] [47800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.42613702481832
+I1201 09:17:48.274635 137274321021824 utils.py:1231] [47800] core_hours = 83.42613702481832
+I1201 09:17:48.274695 137274321021824 train.py:125] NOTE: Steps:47800/112603 [42.5%]
+Walltime:3d11h27m (0s eval)
+ETA:4d17h6m
+Total train time:8d4h32m
+I1201 09:22:58.572647 137274321021824 utils.py:1231] [47850] l2_params = 314.91528708462045
+I1201 09:22:58.572849 137274321021824 utils.py:1231] [47850] train/loss = 2.4320835769176483
+I1201 09:22:58.572954 137274321021824 utils.py:1231] [47850] l2_grads = 1.6256959438323975
+I1201 09:22:58.573026 137274321021824 utils.py:1231] [47850] lr = 0.0007001760127058427
+I1201 09:22:58.573091 137274321021824 utils.py:1231] [47850] uptime = 300767.935453
+I1201 09:22:58.573153 137274321021824 utils.py:1231] [47850] examples_seen = 48998400.0
+I1201 09:22:58.573211 137274321021824 utils.py:1231] [47850] progress = 0.42494427324316403
+I1201 09:22:58.573268 137274321021824 utils.py:1231] [47850] epoch = 38.245131196791675
+I1201 09:22:58.573326 137274321021824 utils.py:1231] [47850] img/sec/core = 165.0022520130765
+I1201 09:22:58.573388 137274321021824 utils.py:1231] [47850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.51233113458778
+I1201 09:22:58.573447 137274321021824 utils.py:1231] [47850] core_hours = 83.51233113458778
+I1201 09:22:58.573509 137274321021824 train.py:125] NOTE: Steps:47850/112603 [42.5%]
+Walltime:3d11h32m (0s eval)
+ETA:4d17h1m
+Total train time:8d4h32m
+I1201 09:28:10.344764 137274321021824 utils.py:1231] [47900] l2_params = 314.85552632715724
+I1201 09:28:10.345088 137274321021824 utils.py:1231] [47900] train/loss = 2.3538960218429565
+I1201 09:28:10.345315 137274321021824 utils.py:1231] [47900] l2_grads = 1.6459952592849731
+I1201 09:28:10.345431 137274321021824 utils.py:1231] [47900] lr = 0.000699474328555565
+I1201 09:28:10.345518 137274321021824 utils.py:1231] [47900] uptime = 301079.707875013
+I1201 09:28:10.345608 137274321021824 utils.py:1231] [47900] examples_seen = 49049600.0
+I1201 09:28:10.345687 137274321021824 utils.py:1231] [47900] progress = 0.42538831114623943
+I1201 09:28:10.345759 137274321021824 utils.py:1231] [47900] epoch = 38.28509476126063
+I1201 09:28:10.345831 137274321021824 utils.py:1231] [47900] img/sec/core = 164.22235061530506
+I1201 09:28:10.345917 137274321021824 utils.py:1231] [47900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.59893458514694
+I1201 09:28:10.345988 137274321021824 utils.py:1231] [47900] core_hours = 83.59893458514694
+I1201 09:28:10.346069 137274321021824 train.py:125] NOTE: Steps:47900/112603 [42.5%]
+Walltime:3d11h37m (0s eval)
+ETA:4d16h55m
+Total train time:8d4h31m
+I1201 09:33:22.125175 137274321021824 utils.py:1231] [47950] l2_params = 314.78625236093643
+I1201 09:33:22.125388 137274321021824 utils.py:1231] [47950] train/loss = 3.161150813102722
+I1201 09:33:22.125488 137274321021824 utils.py:1231] [47950] l2_grads = 1.30668044090271
+I1201 09:33:22.125559 137274321021824 utils.py:1231] [47950] lr = 0.0006987721768784309
+I1201 09:33:22.125625 137274321021824 utils.py:1231] [47950] uptime = 301391.487986565
+I1201 09:33:22.125685 137274321021824 utils.py:1231] [47950] examples_seen = 49100800.0
+I1201 09:33:22.125749 137274321021824 utils.py:1231] [47950] progress = 0.42583234904931483
+I1201 09:33:22.125804 137274321021824 utils.py:1231] [47950] epoch = 38.325058325729586
+I1201 09:33:22.125864 137274321021824 utils.py:1231] [47950] img/sec/core = 164.21830034358493
+I1201 09:33:22.125933 137274321021824 utils.py:1231] [47950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.68554017168915
+I1201 09:33:22.125990 137274321021824 utils.py:1231] [47950] core_hours = 83.68554017168915
+I1201 09:33:22.126055 137274321021824 train.py:125] NOTE: Steps:47950/112603 [42.6%]
+Walltime:3d11h43m (0s eval)
+ETA:4d16h50m
+Total train time:8d4h31m
+I1201 09:38:33.900947 137274321021824 utils.py:1231] [48000] l2_params = 314.74415833392123
+I1201 09:38:33.901174 137274321021824 utils.py:1231] [48000] train/loss = 2.901117652654648
+I1201 09:38:33.901279 137274321021824 utils.py:1231] [48000] l2_grads = 1.3267675638198853
+I1201 09:38:33.901349 137274321021824 utils.py:1231] [48000] lr = 0.0006980695593201414
+I1201 09:38:33.901406 137274321021824 utils.py:1231] [48000] uptime = 301703.263767501
+I1201 09:38:33.901470 137274321021824 utils.py:1231] [48000] examples_seen = 49152000.0
+I1201 09:38:33.901516 137274321021824 utils.py:1231] [48000] progress = 0.42627638695239023
+I1201 09:38:33.901562 137274321021824 utils.py:1231] [48000] epoch = 38.36502189019855
+I1201 09:38:33.901613 137274321021824 utils.py:1231] [48000] img/sec/core = 164.2205813623136
+I1201 09:38:33.901684 137274321021824 utils.py:1231] [48000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.7721445552825
+I1201 09:38:33.901743 137274321021824 utils.py:1231] [48000] core_hours = 83.7721445552825
+I1201 09:38:33.901800 137274321021824 train.py:125] NOTE: Steps:48000/112603 [42.6%]
+Walltime:3d11h48m (0s eval)
+ETA:4d16h45m
+Total train time:8d4h31m
+I1201 09:43:46.014365 137274321021824 utils.py:1231] [48050] l2_params = 314.68574225731277
+I1201 09:43:46.014592 137274321021824 utils.py:1231] [48050] train/loss = 2.4809485375881195
+I1201 09:43:46.014695 137274321021824 utils.py:1231] [48050] l2_grads = 1.4553314447402954
+I1201 09:43:46.014756 137274321021824 utils.py:1231] [48050] lr = 0.0006973664775274861
+I1201 09:43:46.014816 137274321021824 utils.py:1231] [48050] uptime = 302015.377172869
+I1201 09:43:46.014873 137274321021824 utils.py:1231] [48050] examples_seen = 49203200.0
+I1201 09:43:46.014933 137274321021824 utils.py:1231] [48050] progress = 0.4267204248554657
+I1201 09:43:46.014982 137274321021824 utils.py:1231] [48050] epoch = 38.404985454667504
+I1201 09:43:46.015034 137274321021824 utils.py:1231] [48050] img/sec/core = 164.0429379815719
+I1201 09:43:46.015089 137274321021824 utils.py:1231] [48050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.85884272344028
+I1201 09:43:46.015140 137274321021824 utils.py:1231] [48050] core_hours = 83.85884272344028
+I1201 09:43:46.015200 137274321021824 train.py:125] NOTE: Steps:48050/112603 [42.7%]
+Walltime:3d11h53m (0s eval)
+ETA:4d16h39m
+Total train time:8d4h31m
+I1201 09:48:57.779401 137274321021824 utils.py:1231] [48100] l2_params = 314.5970361305303
+I1201 09:48:57.779646 137274321021824 utils.py:1231] [48100] train/loss = 2.4241887629032135
+I1201 09:48:57.779753 137274321021824 utils.py:1231] [48100] l2_grads = 1.4520533084869385
+I1201 09:48:57.779820 137274321021824 utils.py:1231] [48100] lr = 0.0006966629331483452
+I1201 09:48:57.779871 137274321021824 utils.py:1231] [48100] uptime = 302327.142233709
+I1201 09:48:57.779932 137274321021824 utils.py:1231] [48100] examples_seen = 49254400.0
+I1201 09:48:57.779985 137274321021824 utils.py:1231] [48100] progress = 0.4271644627585411
+I1201 09:48:57.780038 137274321021824 utils.py:1231] [48100] epoch = 38.44494901913646
+I1201 09:48:57.780089 137274321021824 utils.py:1231] [48100] img/sec/core = 164.22622811565216
+I1201 09:48:57.780145 137274321021824 utils.py:1231] [48100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 83.94544412922916
+I1201 09:48:57.780195 137274321021824 utils.py:1231] [48100] core_hours = 83.94544412922916
+I1201 09:48:57.780264 137274321021824 train.py:125] NOTE: Steps:48100/112603 [42.7%]
+Walltime:3d11h58m (0s eval)
+ETA:4d16h34m
+Total train time:8d4h31m
+I1201 09:54:09.558922 137274321021824 utils.py:1231] [48150] l2_params = 314.529232953403
+I1201 09:54:09.559146 137274321021824 utils.py:1231] [48150] train/loss = 4.322412014007568
+I1201 09:54:09.559256 137274321021824 utils.py:1231] [48150] l2_grads = 1.3579224348068237
+I1201 09:54:09.559363 137274321021824 utils.py:1231] [48150] lr = 0.0006959589278316822
+I1201 09:54:09.559431 137274321021824 utils.py:1231] [48150] uptime = 302638.921791573
+I1201 09:54:09.559493 137274321021824 utils.py:1231] [48150] examples_seen = 49305600.0
+I1201 09:54:09.559550 137274321021824 utils.py:1231] [48150] progress = 0.4276085006616165
+I1201 09:54:09.559606 137274321021824 utils.py:1231] [48150] epoch = 38.484912583605414
+I1201 09:54:09.559676 137274321021824 utils.py:1231] [48150] img/sec/core = 164.21859197815837
+I1201 09:54:09.559743 137274321021824 utils.py:1231] [48150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.03204956196916
+I1201 09:54:09.559812 137274321021824 utils.py:1231] [48150] core_hours = 84.03204956196916
+I1201 09:54:09.559891 137274321021824 train.py:125] NOTE: Steps:48150/112603 [42.8%]
+Walltime:3d12h3m (0s eval)
+ETA:4d16h29m
+Total train time:8d4h31m
+I1201 09:59:21.323131 137274321021824 utils.py:1231] [48200] l2_params = 314.46294520806396
+I1201 09:59:21.323364 137274321021824 utils.py:1231] [48200] train/loss = 3.863252490758896
+I1201 09:59:21.323492 137274321021824 utils.py:1231] [48200] l2_grads = 1.2622400522232056
+I1201 09:59:21.323575 137274321021824 utils.py:1231] [48200] lr = 0.0006952544632275414
+I1201 09:59:21.323642 137274321021824 utils.py:1231] [48200] uptime = 302950.68600421597
+I1201 09:59:21.323712 137274321021824 utils.py:1231] [48200] examples_seen = 49356800.0
+I1201 09:59:21.323785 137274321021824 utils.py:1231] [48200] progress = 0.4280525385646919
+I1201 09:59:21.323843 137274321021824 utils.py:1231] [48200] epoch = 38.52487614807437
+I1201 09:59:21.323923 137274321021824 utils.py:1231] [48200] img/sec/core = 164.22667491548714
+I1201 09:59:21.323999 137274321021824 utils.py:1231] [48200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.11865073214776
+I1201 09:59:21.324057 137274321021824 utils.py:1231] [48200] core_hours = 84.11865073214776
+I1201 09:59:21.324128 137274321021824 train.py:125] NOTE: Steps:48200/112603 [42.8%]
+Walltime:3d12h9m (0s eval)
+ETA:4d16h24m
+Total train time:8d4h31m
+I1201 10:04:33.262825 137274321021824 utils.py:1231] [48250] l2_params = 314.384954947777
+I1201 10:04:33.263078 137274321021824 utils.py:1231] [48250] train/loss = 2.418149083852768
+I1201 10:04:33.263205 137274321021824 utils.py:1231] [48250] l2_grads = 1.4915159940719604
+I1201 10:04:33.263267 137274321021824 utils.py:1231] [48250] lr = 0.0006945495409870422
+I1201 10:04:33.263315 137274321021824 utils.py:1231] [48250] uptime = 303262.62567788496
+I1201 10:04:33.263364 137274321021824 utils.py:1231] [48250] examples_seen = 49408000.0
+I1201 10:04:33.263411 137274321021824 utils.py:1231] [48250] progress = 0.4284965764677673
+I1201 10:04:33.263456 137274321021824 utils.py:1231] [48250] epoch = 38.56483971254333
+I1201 10:04:33.263504 137274321021824 utils.py:1231] [48250] img/sec/core = 164.13430006447052
+I1201 10:04:33.263556 137274321021824 utils.py:1231] [48250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.20530064150027
+I1201 10:04:33.263607 137274321021824 utils.py:1231] [48250] core_hours = 84.20530064150027
+I1201 10:04:33.263663 137274321021824 train.py:125] NOTE: Steps:48250/112603 [42.8%]
+Walltime:3d12h14m (0s eval)
+ETA:4d16h18m
+Total train time:8d4h31m
+I1201 10:09:45.030866 137274321021824 utils.py:1231] [48300] l2_params = 314.3269261195418
+I1201 10:09:45.031077 137274321021824 utils.py:1231] [48300] train/loss = 2.4490241408348083
+I1201 10:09:45.031181 137274321021824 utils.py:1231] [48300] l2_grads = 1.440463662147522
+I1201 10:09:45.031257 137274321021824 utils.py:1231] [48300] lr = 0.0006938441627623782
+I1201 10:09:45.031317 137274321021824 utils.py:1231] [48300] uptime = 303574.393676747
+I1201 10:09:45.031385 137274321021824 utils.py:1231] [48300] examples_seen = 49459200.0
+I1201 10:09:45.031435 137274321021824 utils.py:1231] [48300] progress = 0.4289406143708427
+I1201 10:09:45.031480 137274321021824 utils.py:1231] [48300] epoch = 38.60480327701229
+I1201 10:09:45.031527 137274321021824 utils.py:1231] [48300] img/sec/core = 164.22468048960786
+I1201 10:09:45.031597 137274321021824 utils.py:1231] [48300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.29190286340639
+I1201 10:09:45.031642 137274321021824 utils.py:1231] [48300] core_hours = 84.29190286340639
+I1201 10:09:45.031698 137274321021824 train.py:125] NOTE: Steps:48300/112603 [42.9%]
+Walltime:3d12h19m (0s eval)
+ETA:4d16h13m
+Total train time:8d4h31m
+I1201 10:14:56.807281 137274321021824 utils.py:1231] [48350] l2_params = 314.2545527660996
+I1201 10:14:56.807511 137274321021824 utils.py:1231] [48350] train/loss = 2.4061860740184784
+I1201 10:14:56.807612 137274321021824 utils.py:1231] [48350] l2_grads = 1.4354451894760132
+I1201 10:14:56.807676 137274321021824 utils.py:1231] [48350] lr = 0.0006931383302068104
+I1201 10:14:56.807730 137274321021824 utils.py:1231] [48350] uptime = 303886.17009254097
+I1201 10:14:56.807787 137274321021824 utils.py:1231] [48350] examples_seen = 49510400.0
+I1201 10:14:56.807840 137274321021824 utils.py:1231] [48350] progress = 0.4293846522739181
+I1201 10:14:56.807897 137274321021824 utils.py:1231] [48350] epoch = 38.64476684148124
+I1201 10:14:56.807954 137274321021824 utils.py:1231] [48350] img/sec/core = 164.2202469664426
+I1201 10:14:56.808013 137274321021824 utils.py:1231] [48350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.37850742334916
+I1201 10:14:56.808064 137274321021824 utils.py:1231] [48350] core_hours = 84.37850742334916
+I1201 10:14:56.808125 137274321021824 train.py:125] NOTE: Steps:48350/112603 [42.9%]
+Walltime:3d12h24m (0s eval)
+ETA:4d16h8m
+Total train time:8d4h31m
+I1201 10:20:08.601580 137274321021824 utils.py:1231] [48400] l2_params = 314.20534989756175
+I1201 10:20:08.601783 137274321021824 utils.py:1231] [48400] train/loss = 2.3366892635822296
+I1201 10:20:08.601892 137274321021824 utils.py:1231] [48400] l2_grads = 1.3749750852584839
+I1201 10:20:08.601966 137274321021824 utils.py:1231] [48400] lr = 0.0006924320449746662
+I1201 10:20:08.602039 137274321021824 utils.py:1231] [48400] uptime = 304197.964400713
+I1201 10:20:08.602112 137274321021824 utils.py:1231] [48400] examples_seen = 49561600.0
+I1201 10:20:08.602169 137274321021824 utils.py:1231] [48400] progress = 0.4298286901769935
+I1201 10:20:08.602223 137274321021824 utils.py:1231] [48400] epoch = 38.6847304059502
+I1201 10:20:08.602284 137274321021824 utils.py:1231] [48400] img/sec/core = 164.21082315509517
+I1201 10:20:08.602379 137274321021824 utils.py:1231] [48400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.46511695339694
+I1201 10:20:08.602432 137274321021824 utils.py:1231] [48400] core_hours = 84.46511695339694
+I1201 10:20:08.602496 137274321021824 train.py:125] NOTE: Steps:48400/112603 [43.0%]
+Walltime:3d12h29m (0s eval)
+ETA:4d16h2m
+Total train time:8d4h31m
+I1201 10:25:20.387284 137274321021824 utils.py:1231] [48450] l2_params = 314.1416663914638
+I1201 10:25:20.387523 137274321021824 utils.py:1231] [48450] train/loss = 2.399618625640869
+I1201 10:25:20.387660 137274321021824 utils.py:1231] [48450] l2_grads = 1.584885597229004
+I1201 10:25:20.387765 137274321021824 utils.py:1231] [48450] lr = 0.0006917253087213326
+I1201 10:25:20.387830 137274321021824 utils.py:1231] [48450] uptime = 304509.7501919
+I1201 10:25:20.387906 137274321021824 utils.py:1231] [48450] examples_seen = 49612800.0
+I1201 10:25:20.387968 137274321021824 utils.py:1231] [48450] progress = 0.4302727280800689
+I1201 10:25:20.388020 137274321021824 utils.py:1231] [48450] epoch = 38.72469397041916
+I1201 10:25:20.388075 137274321021824 utils.py:1231] [48450] img/sec/core = 164.21530886661836
+I1201 10:25:20.388131 137274321021824 utils.py:1231] [48450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.55172411761555
+I1201 10:25:20.388184 137274321021824 utils.py:1231] [48450] core_hours = 84.55172411761555
+I1201 10:25:20.388250 137274321021824 train.py:125] NOTE: Steps:48450/112603 [43.0%]
+Walltime:3d12h35m (0s eval)
+ETA:4d15h57m
+Total train time:8d4h30m
+I1201 10:30:32.178654 137274321021824 utils.py:1231] [48500] l2_params = 314.0648613523085
+I1201 10:30:32.178952 137274321021824 utils.py:1231] [48500] train/loss = 4.520925521850586
+I1201 10:30:32.179125 137274321021824 utils.py:1231] [48500] l2_grads = 1.3778412342071533
+I1201 10:30:32.179224 137274321021824 utils.py:1231] [48500] lr = 0.0006910181231032539
+I1201 10:30:32.179296 137274321021824 utils.py:1231] [48500] uptime = 304821.541657548
+I1201 10:30:32.179367 137274321021824 utils.py:1231] [48500] examples_seen = 49664000.0
+I1201 10:30:32.179431 137274321021824 utils.py:1231] [48500] progress = 0.43071676598314435
+I1201 10:30:32.179491 137274321021824 utils.py:1231] [48500] epoch = 38.764657534888116
+I1201 10:30:32.179552 137274321021824 utils.py:1231] [48500] img/sec/core = 164.2123202236886
+I1201 10:30:32.179618 137274321021824 utils.py:1231] [48500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.63833285807333
+I1201 10:30:32.179682 137274321021824 utils.py:1231] [48500] core_hours = 84.63833285807333
+I1201 10:30:32.179766 137274321021824 train.py:125] NOTE: Steps:48500/112603 [43.1%]
+Walltime:3d12h40m (0s eval)
+ETA:4d15h52m
+Total train time:8d4h30m
+I1201 10:35:43.959691 137274321021824 utils.py:1231] [48550] l2_params = 313.97604015238267
+I1201 10:35:43.959944 137274321021824 utils.py:1231] [48550] train/loss = 2.5692822635173798
+I1201 10:35:43.960078 137274321021824 utils.py:1231] [48550] l2_grads = 1.4716219902038574
+I1201 10:35:43.960169 137274321021824 utils.py:1231] [48550] lr = 0.0006903104897779284
+I1201 10:35:43.960234 137274321021824 utils.py:1231] [48550] uptime = 305133.322592023
+I1201 10:35:43.960292 137274321021824 utils.py:1231] [48550] examples_seen = 49715200.0
+I1201 10:35:43.960343 137274321021824 utils.py:1231] [48550] progress = 0.43116080388621975
+I1201 10:35:43.960392 137274321021824 utils.py:1231] [48550] epoch = 38.80462109935707
+I1201 10:35:43.960444 137274321021824 utils.py:1231] [48550] img/sec/core = 164.21786690136213
+I1201 10:35:43.960513 137274321021824 utils.py:1231] [48550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.72493867320527
+I1201 10:35:43.960566 137274321021824 utils.py:1231] [48550] core_hours = 84.72493867320527
+I1201 10:35:43.960628 137274321021824 train.py:125] NOTE: Steps:48550/112603 [43.1%]
+Walltime:3d12h45m (0s eval)
+ETA:4d15h47m
+Total train time:8d4h30m
+I1201 10:40:55.738079 137274321021824 utils.py:1231] [48600] l2_params = 313.93884886366993
+I1201 10:40:55.738372 137274321021824 utils.py:1231] [48600] train/loss = 2.5434380769729614
+I1201 10:40:55.738482 137274321021824 utils.py:1231] [48600] l2_grads = 1.5680853128433228
+I1201 10:40:55.738552 137274321021824 utils.py:1231] [48600] lr = 0.0006896024104039033
+I1201 10:40:55.738608 137274321021824 utils.py:1231] [48600] uptime = 305445.100970415
+I1201 10:40:55.738671 137274321021824 utils.py:1231] [48600] examples_seen = 49766400.0
+I1201 10:40:55.738725 137274321021824 utils.py:1231] [48600] progress = 0.43160484178929515
+I1201 10:40:55.738780 137274321021824 utils.py:1231] [48600] epoch = 38.844584663826026
+I1201 10:40:55.738833 137274321021824 utils.py:1231] [48600] img/sec/core = 164.2192132246839
+I1201 10:40:55.738895 137274321021824 utils.py:1231] [48600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.81154377831416
+I1201 10:40:55.738958 137274321021824 utils.py:1231] [48600] core_hours = 84.81154377831416
+I1201 10:40:55.739023 137274321021824 train.py:125] NOTE: Steps:48600/112603 [43.2%]
+Walltime:3d12h50m (0s eval)
+ETA:4d15h41m
+Total train time:8d4h30m
+I1201 10:46:07.515199 137274321021824 utils.py:1231] [48650] l2_params = 313.8449389140715
+I1201 10:46:07.515422 137274321021824 utils.py:1231] [48650] train/loss = 3.6767291128635406
+I1201 10:46:07.515535 137274321021824 utils.py:1231] [48650] l2_grads = 1.2691452503204346
+I1201 10:46:07.515610 137274321021824 utils.py:1231] [48650] lr = 0.000688893886640771
+I1201 10:46:07.515671 137274321021824 utils.py:1231] [48650] uptime = 305756.878032721
+I1201 10:46:07.515737 137274321021824 utils.py:1231] [48650] examples_seen = 49817600.0
+I1201 10:46:07.515795 137274321021824 utils.py:1231] [48650] progress = 0.43204887969237055
+I1201 10:46:07.515852 137274321021824 utils.py:1231] [48650] epoch = 38.88454822829498
+I1201 10:46:07.515916 137274321021824 utils.py:1231] [48650] img/sec/core = 164.2199064334878
+I1201 10:46:07.515976 137274321021824 utils.py:1231] [48650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.89814851784361
+I1201 10:46:07.516027 137274321021824 utils.py:1231] [48650] core_hours = 84.89814851784361
+I1201 10:46:07.516098 137274321021824 train.py:125] NOTE: Steps:48650/112603 [43.2%]
+Walltime:3d12h55m (0s eval)
+ETA:4d15h36m
+Total train time:8d4h30m
+I1201 10:51:19.299325 137274321021824 utils.py:1231] [48700] l2_params = 313.77257171952886
+I1201 10:51:19.299593 137274321021824 utils.py:1231] [48700] train/loss = 2.3764727413654327
+I1201 10:51:19.299719 137274321021824 utils.py:1231] [48700] l2_grads = 1.4554526805877686
+I1201 10:51:19.299813 137274321021824 utils.py:1231] [48700] lr = 0.0006881849201491658
+I1201 10:51:19.299879 137274321021824 utils.py:1231] [48700] uptime = 306068.662240538
+I1201 10:51:19.299947 137274321021824 utils.py:1231] [48700] examples_seen = 49868800.0
+I1201 10:51:19.300003 137274321021824 utils.py:1231] [48700] progress = 0.43249291759544595
+I1201 10:51:19.300058 137274321021824 utils.py:1231] [48700] epoch = 38.924511792763944
+I1201 10:51:19.300113 137274321021824 utils.py:1231] [48700] img/sec/core = 164.21614282033426
+I1201 10:51:19.300177 137274321021824 utils.py:1231] [48700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 84.98475524223721
+I1201 10:51:19.300236 137274321021824 utils.py:1231] [48700] core_hours = 84.98475524223721
+I1201 10:51:19.300298 137274321021824 train.py:125] NOTE: Steps:48700/112603 [43.2%]
+Walltime:3d13h1m (0s eval)
+ETA:4d15h31m
+Total train time:8d4h30m
+I1201 10:56:31.072190 137274321021824 utils.py:1231] [48750] l2_params = 313.68624472789725
+I1201 10:56:31.072447 137274321021824 utils.py:1231] [48750] train/loss = 2.461132049560547
+I1201 10:56:31.072638 137274321021824 utils.py:1231] [48750] l2_grads = 1.5752238035202026
+I1201 10:56:31.072716 137274321021824 utils.py:1231] [48750] lr = 0.0006874755125907596
+I1201 10:56:31.072786 137274321021824 utils.py:1231] [48750] uptime = 306380.43514366896
+I1201 10:56:31.072846 137274321021824 utils.py:1231] [48750] examples_seen = 49920000.0
+I1201 10:56:31.072912 137274321021824 utils.py:1231] [48750] progress = 0.43293695549852135
+I1201 10:56:31.072971 137274321021824 utils.py:1231] [48750] epoch = 38.9644753572329
+I1201 10:56:31.073029 137274321021824 utils.py:1231] [48750] img/sec/core = 164.2220971926208
+I1201 10:56:31.073091 137274321021824 utils.py:1231] [48750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.07135882644026
+I1201 10:56:31.073146 137274321021824 utils.py:1231] [48750] core_hours = 85.07135882644026
+I1201 10:56:31.073216 137274321021824 train.py:125] NOTE: Steps:48750/112603 [43.3%]
+Walltime:3d13h6m (0s eval)
+ETA:4d15h25m
+Total train time:8d4h30m
+I1201 11:01:39.162437 137274321021824 utils.py:1231] [48800] l2_params = 313.6331580611147
+I1201 11:01:39.162645 137274321021824 utils.py:1231] [48800] train/loss = 4.795711874961853
+I1201 11:01:39.162738 137274321021824 utils.py:1231] [48800] l2_grads = 1.3117287158966064
+I1201 11:01:39.162833 137274321021824 utils.py:1231] [48800] lr = 0.0006867656656282581
+I1201 11:01:39.162895 137274321021824 utils.py:1231] [48800] uptime = 306688.525256916
+I1201 11:01:39.162965 137274321021824 utils.py:1231] [48800] examples_seen = 49971200.0
+I1201 11:01:39.163021 137274321021824 utils.py:1231] [48800] progress = 0.43338099340159675
+I1201 11:01:39.163073 137274321021824 utils.py:1231] [48800] epoch = 39.004438921701855
+I1201 11:01:39.163141 137274321021824 utils.py:1231] [48800] img/sec/core = 166.1851445357631
+I1201 11:01:39.163210 137274321021824 utils.py:1231] [48800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.15693941345333
+I1201 11:01:39.163265 137274321021824 utils.py:1231] [48800] core_hours = 85.15693941345333
+I1201 11:01:39.163324 137274321021824 train.py:125] NOTE: Steps:48800/112603 [43.3%]
+Walltime:3d13h11m (0s eval)
+ETA:4d15h20m
+Total train time:8d4h30m
+I1201 11:06:50.951173 137274321021824 utils.py:1231] [48850] l2_params = 313.55578175533554
+I1201 11:06:50.951390 137274321021824 utils.py:1231] [48850] train/loss = 3.4350481927394867
+I1201 11:06:50.951516 137274321021824 utils.py:1231] [48850] l2_grads = 1.3139797449111938
+I1201 11:06:50.951602 137274321021824 utils.py:1231] [48850] lr = 0.0006860553809253967
+I1201 11:06:50.951688 137274321021824 utils.py:1231] [48850] uptime = 307000.314044393
+I1201 11:06:50.951747 137274321021824 utils.py:1231] [48850] examples_seen = 50022400.0
+I1201 11:06:50.951804 137274321021824 utils.py:1231] [48850] progress = 0.43382503130467215
+I1201 11:06:50.951860 137274321021824 utils.py:1231] [48850] epoch = 39.04440248617081
+I1201 11:06:50.951935 137274321021824 utils.py:1231] [48850] img/sec/core = 164.21373075763785
+I1201 11:06:50.951998 137274321021824 utils.py:1231] [48850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.24354740997472
+I1201 11:06:50.952054 137274321021824 utils.py:1231] [48850] core_hours = 85.24354740997472
+I1201 11:06:50.952118 137274321021824 train.py:125] NOTE: Steps:48850/112603 [43.4%]
+Walltime:3d13h16m (0s eval)
+ETA:4d15h15m
+Total train time:8d4h30m
+I1201 11:12:02.733205 137274321021824 utils.py:1231] [48900] l2_params = 313.4904357236061
+I1201 11:12:02.733431 137274321021824 utils.py:1231] [48900] train/loss = 2.453496664762497
+I1201 11:12:02.733543 137274321021824 utils.py:1231] [48900] l2_grads = 1.516906499862671
+I1201 11:12:02.733618 137274321021824 utils.py:1231] [48900] lr = 0.000685344660146937
+I1201 11:12:02.733680 137274321021824 utils.py:1231] [48900] uptime = 307312.096042576
+I1201 11:12:02.733738 137274321021824 utils.py:1231] [48900] examples_seen = 50073600.0
+I1201 11:12:02.733788 137274321021824 utils.py:1231] [48900] progress = 0.43426906920774755
+I1201 11:12:02.733835 137274321021824 utils.py:1231] [48900] epoch = 39.084366050639765
+I1201 11:12:02.733890 137274321021824 utils.py:1231] [48900] img/sec/core = 164.21730663855988
+I1201 11:12:02.733963 137274321021824 utils.py:1231] [48900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.3301535205811
+I1201 11:12:02.734022 137274321021824 utils.py:1231] [48900] core_hours = 85.3301535205811
+I1201 11:12:02.734084 137274321021824 train.py:125] NOTE: Steps:48900/112603 [43.4%]
+Walltime:3d13h21m (0s eval)
+ETA:4d15h9m
+Total train time:8d4h29m
+I1201 11:17:14.503100 137274321021824 utils.py:1231] [48950] l2_params = 313.42331722666904
+I1201 11:17:14.503300 137274321021824 utils.py:1231] [48950] train/loss = 4.904091119766235
+I1201 11:17:14.503392 137274321021824 utils.py:1231] [48950] l2_grads = 1.3848475217819214
+I1201 11:17:14.503451 137274321021824 utils.py:1231] [48950] lr = 0.0006846335049586619
+I1201 11:17:14.503504 137274321021824 utils.py:1231] [48950] uptime = 307623.865866707
+I1201 11:17:14.503555 137274321021824 utils.py:1231] [48950] examples_seen = 50124800.0
+I1201 11:17:14.503602 137274321021824 utils.py:1231] [48950] progress = 0.434713107110823
+I1201 11:17:14.503648 137274321021824 utils.py:1231] [48950] epoch = 39.12432961510873
+I1201 11:17:14.503696 137274321021824 utils.py:1231] [48950] img/sec/core = 164.2237190296173
+I1201 11:17:14.503753 137274321021824 utils.py:1231] [48950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.41675624950638
+I1201 11:17:14.503800 137274321021824 utils.py:1231] [48950] core_hours = 85.41675624950638
+I1201 11:17:14.503856 137274321021824 train.py:125] NOTE: Steps:48950/112603 [43.5%]
+Walltime:3d13h27m (0s eval)
+ETA:4d15h4m
+Total train time:8d4h29m
+I1201 11:22:26.285560 137274321021824 utils.py:1231] [49000] l2_params = 313.34153066779174
+I1201 11:22:26.285782 137274321021824 utils.py:1231] [49000] train/loss = 2.342273712158203
+I1201 11:22:26.285887 137274321021824 utils.py:1231] [49000] l2_grads = 1.6058335304260254
+I1201 11:22:26.285952 137274321021824 utils.py:1231] [49000] lr = 0.0006839219170273734
+I1201 11:22:26.286018 137274321021824 utils.py:1231] [49000] uptime = 307935.64837536396
+I1201 11:22:26.286072 137274321021824 utils.py:1231] [49000] examples_seen = 50176000.0
+I1201 11:22:26.286121 137274321021824 utils.py:1231] [49000] progress = 0.4351571450138984
+I1201 11:22:26.286169 137274321021824 utils.py:1231] [49000] epoch = 39.16429317957768
+I1201 11:22:26.286220 137274321021824 utils.py:1231] [49000] img/sec/core = 164.21703776953137
+I1201 11:22:26.286277 137274321021824 utils.py:1231] [49000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.5033625019111
+I1201 11:22:26.286329 137274321021824 utils.py:1231] [49000] core_hours = 85.5033625019111
+I1201 11:22:26.286388 137274321021824 train.py:125] NOTE: Steps:49000/112603 [43.5%]
+Walltime:3d13h32m (0s eval)
+ETA:4d14h59m
+Total train time:8d4h29m
+I1201 11:27:37.897496 137274321021824 utils.py:1231] [49050] l2_params = 313.29082910119496
+I1201 11:27:37.897711 137274321021824 utils.py:1231] [49050] train/loss = 3.4315770268440247
+I1201 11:27:37.897809 137274321021824 utils.py:1231] [49050] l2_grads = 1.3271864652633667
+I1201 11:27:37.897878 137274321021824 utils.py:1231] [49050] lr = 0.0006832098980208878
+I1201 11:27:37.897940 137274321021824 utils.py:1231] [49050] uptime = 308247.260302303
+I1201 11:27:37.898014 137274321021824 utils.py:1231] [49050] examples_seen = 50227200.0
+I1201 11:27:37.898073 137274321021824 utils.py:1231] [49050] progress = 0.4356011829169738
+I1201 11:27:37.898136 137274321021824 utils.py:1231] [49050] epoch = 39.20425674404664
+I1201 11:27:37.898191 137274321021824 utils.py:1231] [49050] img/sec/core = 164.30693299494578
+I1201 11:27:37.898251 137274321021824 utils.py:1231] [49050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.58992137050527
+I1201 11:27:37.898301 137274321021824 utils.py:1231] [49050] core_hours = 85.58992137050527
+I1201 11:27:37.898363 137274321021824 train.py:125] NOTE: Steps:49050/112603 [43.6%]
+Walltime:3d13h37m (0s eval)
+ETA:4d14h54m
+Total train time:8d4h29m
+I1201 11:32:49.659517 137274321021824 utils.py:1231] [49100] l2_params = 313.229334201214
+I1201 11:32:49.659738 137274321021824 utils.py:1231] [49100] train/loss = 2.3562064170837402
+I1201 11:32:49.659839 137274321021824 utils.py:1231] [49100] l2_grads = 1.4249670505523682
+I1201 11:32:49.659917 137274321021824 utils.py:1231] [49100] lr = 0.0006824974496080312
+I1201 11:32:49.659984 137274321021824 utils.py:1231] [49100] uptime = 308559.02234128397
+I1201 11:32:49.660043 137274321021824 utils.py:1231] [49100] examples_seen = 50278400.0
+I1201 11:32:49.660100 137274321021824 utils.py:1231] [49100] progress = 0.4360452208200492
+I1201 11:32:49.660151 137274321021824 utils.py:1231] [49100] epoch = 39.244220308515594
+I1201 11:32:49.660209 137274321021824 utils.py:1231] [49100] img/sec/core = 164.22781993393016
+I1201 11:32:49.660283 137274321021824 utils.py:1231] [49100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.67652193688888
+I1201 11:32:49.660351 137274321021824 utils.py:1231] [49100] core_hours = 85.67652193688888
+I1201 11:32:49.660412 137274321021824 train.py:125] NOTE: Steps:49100/112603 [43.6%]
+Walltime:3d13h42m (0s eval)
+ETA:4d14h48m
+Total train time:8d4h29m
+I1201 11:38:01.436115 137274321021824 utils.py:1231] [49150] l2_params = 313.1490637849561
+I1201 11:38:01.436361 137274321021824 utils.py:1231] [49150] train/loss = 2.2834435552358627
+I1201 11:38:01.436468 137274321021824 utils.py:1231] [49150] l2_grads = 1.4563491344451904
+I1201 11:38:01.436536 137274321021824 utils.py:1231] [49150] lr = 0.0006817845734586362
+I1201 11:38:01.436598 137274321021824 utils.py:1231] [49150] uptime = 308870.798958949
+I1201 11:38:01.436655 137274321021824 utils.py:1231] [49150] examples_seen = 50329600.0
+I1201 11:38:01.436712 137274321021824 utils.py:1231] [49150] progress = 0.4364892587231246
+I1201 11:38:01.436769 137274321021824 utils.py:1231] [49150] epoch = 39.28418387298455
+I1201 11:38:01.436826 137274321021824 utils.py:1231] [49150] img/sec/core = 164.22014063610993
+I1201 11:38:01.436893 137274321021824 utils.py:1231] [49150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.76312655290694
+I1201 11:38:01.436949 137274321021824 utils.py:1231] [49150] core_hours = 85.76312655290694
+I1201 11:38:01.437016 137274321021824 train.py:125] NOTE: Steps:49150/112603 [43.6%]
+Walltime:3d13h47m (0s eval)
+ETA:4d14h43m
+Total train time:8d4h29m
+I1201 11:43:13.220152 137274321021824 utils.py:1231] [49200] l2_params = 313.08911259403004
+I1201 11:43:13.220370 137274321021824 utils.py:1231] [49200] train/loss = 2.689150631427765
+I1201 11:43:13.220476 137274321021824 utils.py:1231] [49200] l2_grads = 1.3417073488235474
+I1201 11:43:13.220567 137274321021824 utils.py:1231] [49200] lr = 0.0006810712712435382
+I1201 11:43:13.220635 137274321021824 utils.py:1231] [49200] uptime = 309182.582995573
+I1201 11:43:13.220692 137274321021824 utils.py:1231] [49200] examples_seen = 50380800.0
+I1201 11:43:13.220746 137274321021824 utils.py:1231] [49200] progress = 0.4369332966262
+I1201 11:43:13.220799 137274321021824 utils.py:1231] [49200] epoch = 39.32414743745351
+I1201 11:43:13.220855 137274321021824 utils.py:1231] [49200] img/sec/core = 164.21623298739712
+I1201 11:43:13.220925 137274321021824 utils.py:1231] [49200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.84973322974695
+I1201 11:43:13.220984 137274321021824 utils.py:1231] [49200] core_hours = 85.84973322974695
+I1201 11:43:13.221050 137274321021824 train.py:125] NOTE: Steps:49200/112603 [43.7%]
+Walltime:3d13h53m (0s eval)
+ETA:4d14h38m
+Total train time:8d4h29m
+I1201 11:48:24.883099 137274321021824 utils.py:1231] [49250] l2_params = 312.99666500590484
+I1201 11:48:24.883304 137274321021824 utils.py:1231] [49250] train/loss = 2.3144869208335876
+I1201 11:48:24.883398 137274321021824 utils.py:1231] [49250] l2_grads = 1.5034576654434204
+I1201 11:48:24.883459 137274321021824 utils.py:1231] [49250] lr = 0.0006803575446345713
+I1201 11:48:24.883512 137274321021824 utils.py:1231] [49250] uptime = 309494.245873957
+I1201 11:48:24.883565 137274321021824 utils.py:1231] [49250] examples_seen = 50432000.0
+I1201 11:48:24.883617 137274321021824 utils.py:1231] [49250] progress = 0.4373773345292754
+I1201 11:48:24.883665 137274321021824 utils.py:1231] [49250] epoch = 39.36411100192247
+I1201 11:48:24.883717 137274321021824 utils.py:1231] [49250] img/sec/core = 164.28007167705312
+I1201 11:48:24.883774 137274321021824 utils.py:1231] [49250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 85.93630625152028
+I1201 11:48:24.883823 137274321021824 utils.py:1231] [49250] core_hours = 85.93630625152028
+I1201 11:48:24.883888 137274321021824 train.py:125] NOTE: Steps:49250/112603 [43.7%]
+Walltime:3d13h58m (0s eval)
+ETA:4d14h32m
+Total train time:8d4h29m
+I1201 11:53:36.652513 137274321021824 utils.py:1231] [49300] l2_params = 312.9198721580012
+I1201 11:53:36.652781 137274321021824 utils.py:1231] [49300] train/loss = 2.296733111143112
+I1201 11:53:36.652921 137274321021824 utils.py:1231] [49300] l2_grads = 1.5634644031524658
+I1201 11:53:36.653014 137274321021824 utils.py:1231] [49300] lr = 0.0006796433953045632
+I1201 11:53:36.653084 137274321021824 utils.py:1231] [49300] uptime = 309806.0154459
+I1201 11:53:36.653159 137274321021824 utils.py:1231] [49300] examples_seen = 50483200.0
+I1201 11:53:36.653232 137274321021824 utils.py:1231] [49300] progress = 0.4378213724323508
+I1201 11:53:36.653295 137274321021824 utils.py:1231] [49300] epoch = 39.40407456639142
+I1201 11:53:36.653352 137274321021824 utils.py:1231] [49300] img/sec/core = 164.2238518689205
+I1201 11:53:36.653432 137274321021824 utils.py:1231] [49300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.02290891039333
+I1201 11:53:36.653501 137274321021824 utils.py:1231] [49300] core_hours = 86.02290891039333
+I1201 11:53:36.653567 137274321021824 train.py:125] NOTE: Steps:49300/112603 [43.8%]
+Walltime:3d14h3m (0s eval)
+ETA:4d14h27m
+Total train time:8d4h29m
+I1201 11:58:48.426445 137274321021824 utils.py:1231] [49350] l2_params = 312.83316832066225
+I1201 11:58:48.426671 137274321021824 utils.py:1231] [49350] train/loss = 2.4866059720516205
+I1201 11:58:48.426780 137274321021824 utils.py:1231] [49350] l2_grads = 1.472379207611084
+I1201 11:58:48.426849 137274321021824 utils.py:1231] [49350] lr = 0.0006789288249273346
+I1201 11:58:48.426910 137274321021824 utils.py:1231] [49350] uptime = 310117.78927151
+I1201 11:58:48.426961 137274321021824 utils.py:1231] [49350] examples_seen = 50534400.0
+I1201 11:58:48.427009 137274321021824 utils.py:1231] [49350] progress = 0.4382654103354262
+I1201 11:58:48.427056 137274321021824 utils.py:1231] [49350] epoch = 39.44403813086038
+I1201 11:58:48.427106 137274321021824 utils.py:1231] [49350] img/sec/core = 164.22161129089363
+I1201 11:58:48.427160 137274321021824 utils.py:1231] [49350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.10951275084055
+I1201 11:58:48.427207 137274321021824 utils.py:1231] [49350] core_hours = 86.10951275084055
+I1201 11:58:48.427267 137274321021824 train.py:125] NOTE: Steps:49350/112603 [43.8%]
+Walltime:3d14h8m (0s eval)
+ETA:4d14h22m
+Total train time:8d4h29m
+I1201 12:04:00.204523 137274321021824 utils.py:1231] [49400] l2_params = 312.77016213680287
+I1201 12:04:00.204770 137274321021824 utils.py:1231] [49400] train/loss = 2.3901089429855347
+I1201 12:04:00.205000 137274321021824 utils.py:1231] [49400] l2_grads = 1.4593276977539062
+I1201 12:04:00.205103 137274321021824 utils.py:1231] [49400] lr = 0.0006782138351776912
+I1201 12:04:00.205201 137274321021824 utils.py:1231] [49400] uptime = 310429.56755927997
+I1201 12:04:00.205277 137274321021824 utils.py:1231] [49400] examples_seen = 50585600.0
+I1201 12:04:00.205374 137274321021824 utils.py:1231] [49400] progress = 0.43870944823850166
+I1201 12:04:00.205447 137274321021824 utils.py:1231] [49400] epoch = 39.48400169532933
+I1201 12:04:00.205514 137274321021824 utils.py:1231] [49400] img/sec/core = 164.2192609569126
+I1201 12:04:00.205604 137274321021824 utils.py:1231] [49400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.19611783077666
+I1201 12:04:00.205668 137274321021824 utils.py:1231] [49400] core_hours = 86.19611783077666
+I1201 12:04:00.205755 137274321021824 train.py:125] NOTE: Steps:49400/112603 [43.9%]
+Walltime:3d14h13m (0s eval)
+ETA:4d14h17m
+Total train time:8d4h29m
+I1201 12:09:12.010191 137274321021824 utils.py:1231] [49450] l2_params = 312.67361326870144
+I1201 12:09:12.010475 137274321021824 utils.py:1231] [49450] train/loss = 2.6951706409454346
+I1201 12:09:12.010750 137274321021824 utils.py:1231] [49450] l2_grads = 1.3331186771392822
+I1201 12:09:12.010889 137274321021824 utils.py:1231] [49450] lr = 0.0006774984277314216
+I1201 12:09:12.010986 137274321021824 utils.py:1231] [49450] uptime = 310741.37334127596
+I1201 12:09:12.011067 137274321021824 utils.py:1231] [49450] examples_seen = 50636800.0
+I1201 12:09:12.011139 137274321021824 utils.py:1231] [49450] progress = 0.43915348614157707
+I1201 12:09:12.011217 137274321021824 utils.py:1231] [49450] epoch = 39.523965259798295
+I1201 12:09:12.011323 137274321021824 utils.py:1231] [49450] img/sec/core = 164.20478052795633
+I1201 12:09:12.011409 137274321021824 utils.py:1231] [49450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.28273054799776
+I1201 12:09:12.011487 137274321021824 utils.py:1231] [49450] core_hours = 86.28273054799776
+I1201 12:09:12.011587 137274321021824 train.py:125] NOTE: Steps:49450/112603 [43.9%]
+Walltime:3d14h19m (0s eval)
+ETA:4d14h11m
+Total train time:8d4h28m
+I1201 12:14:23.790166 137274321021824 utils.py:1231] [49500] l2_params = 312.5895726431904
+I1201 12:14:23.790446 137274321021824 utils.py:1231] [49500] train/loss = 2.506605476140976
+I1201 12:14:23.790623 137274321021824 utils.py:1231] [49500] l2_grads = 1.4939976930618286
+I1201 12:14:23.790714 137274321021824 utils.py:1231] [49500] lr = 0.0006767826042652946
+I1201 12:14:23.790789 137274321021824 utils.py:1231] [49500] uptime = 311053.153144031
+I1201 12:14:23.790848 137274321021824 utils.py:1231] [49500] examples_seen = 50688000.0
+I1201 12:14:23.790917 137274321021824 utils.py:1231] [49500] progress = 0.43959752404465247
+I1201 12:14:23.791039 137274321021824 utils.py:1231] [49500] epoch = 39.56392882426725
+I1201 12:14:23.791124 137274321021824 utils.py:1231] [49500] img/sec/core = 164.2184629907733
+I1201 12:14:23.791191 137274321021824 utils.py:1231] [49500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.36933604876305
+I1201 12:14:23.791249 137274321021824 utils.py:1231] [49500] core_hours = 86.36933604876305
+I1201 12:14:23.791329 137274321021824 train.py:125] NOTE: Steps:49500/112603 [44.0%]
+Walltime:3d14h24m (0s eval)
+ETA:4d14h6m
+Total train time:8d4h28m
+I1201 12:19:35.645683 137274321021824 utils.py:1231] [49550] l2_params = 312.52207259751776
+I1201 12:19:35.645953 137274321021824 utils.py:1231] [49550] train/loss = 2.4526360034942627
+I1201 12:19:35.646193 137274321021824 utils.py:1231] [49550] l2_grads = 1.51970636844635
+I1201 12:19:35.646341 137274321021824 utils.py:1231] [49550] lr = 0.0006760663664570523
+I1201 12:19:35.646453 137274321021824 utils.py:1231] [49550] uptime = 311365.008804638
+I1201 12:19:35.646560 137274321021824 utils.py:1231] [49550] examples_seen = 50739200.0
+I1201 12:19:35.646674 137274321021824 utils.py:1231] [49550] progress = 0.44004156194772787
+I1201 12:19:35.646767 137274321021824 utils.py:1231] [49550] epoch = 39.603892388736206
+I1201 12:19:35.646865 137274321021824 utils.py:1231] [49550] img/sec/core = 164.17851739598117
+I1201 12:19:35.646980 137274321021824 utils.py:1231] [49550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.45596262115389
+I1201 12:19:35.647064 137274321021824 utils.py:1231] [49550] core_hours = 86.45596262115389
+I1201 12:19:35.647162 137274321021824 train.py:125] NOTE: Steps:49550/112603 [44.0%]
+Walltime:3d14h29m (0s eval)
+ETA:4d14h1m
+Total train time:8d4h28m
+I1201 12:24:47.463222 137274321021824 utils.py:1231] [49600] l2_params = 312.42585985504644
+I1201 12:24:47.463504 137274321021824 utils.py:1231] [49600] train/loss = 2.257215142250061
+I1201 12:24:47.463620 137274321021824 utils.py:1231] [49600] l2_grads = 1.4101365804672241
+I1201 12:24:47.463695 137274321021824 utils.py:1231] [49600] lr = 0.00067534971598541
+I1201 12:24:47.463762 137274321021824 utils.py:1231] [49600] uptime = 311676.826119682
+I1201 12:24:47.463822 137274321021824 utils.py:1231] [49600] examples_seen = 50790400.0
+I1201 12:24:47.463877 137274321021824 utils.py:1231] [49600] progress = 0.44048559985080327
+I1201 12:24:47.463936 137274321021824 utils.py:1231] [49600] epoch = 39.64385595320516
+I1201 12:24:47.463987 137274321021824 utils.py:1231] [49600] img/sec/core = 164.1987071589517
+I1201 12:24:47.464042 137274321021824 utils.py:1231] [49600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.54257854199945
+I1201 12:24:47.464090 137274321021824 utils.py:1231] [49600] core_hours = 86.54257854199945
+I1201 12:24:47.464150 137274321021824 train.py:125] NOTE: Steps:49600/112603 [44.0%]
+Walltime:3d14h34m (0s eval)
+ETA:4d13h55m
+Total train time:8d4h28m
+I1201 12:29:59.256549 137274321021824 utils.py:1231] [49650] l2_params = 312.3441446629102
+I1201 12:29:59.256758 137274321021824 utils.py:1231] [49650] train/loss = 2.332817941904068
+I1201 12:29:59.256865 137274321021824 utils.py:1231] [49650] l2_grads = 1.4220972061157227
+I1201 12:29:59.256944 137274321021824 utils.py:1231] [49650] lr = 0.0006746326545300494
+I1201 12:29:59.257005 137274321021824 utils.py:1231] [49650] uptime = 311988.619366555
+I1201 12:29:59.257065 137274321021824 utils.py:1231] [49650] examples_seen = 50841600.0
+I1201 12:29:59.257122 137274321021824 utils.py:1231] [49650] progress = 0.44092963775387867
+I1201 12:29:59.257194 137274321021824 utils.py:1231] [49650] epoch = 39.683819517674124
+I1201 12:29:59.257266 137274321021824 utils.py:1231] [49650] img/sec/core = 164.21138210492583
+I1201 12:29:59.257344 137274321021824 utils.py:1231] [49650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.62918777724194
+I1201 12:29:59.257408 137274321021824 utils.py:1231] [49650] core_hours = 86.62918777724194
+I1201 12:29:59.257483 137274321021824 train.py:125] NOTE: Steps:49650/112603 [44.1%]
+Walltime:3d14h39m (0s eval)
+ETA:4d13h50m
+Total train time:8d4h28m
+I1201 12:35:11.037601 137274321021824 utils.py:1231] [49700] l2_params = 312.2913803058662
+I1201 12:35:11.037819 137274321021824 utils.py:1231] [49700] train/loss = 2.3745903968811035
+I1201 12:35:11.037926 137274321021824 utils.py:1231] [49700] l2_grads = 1.5705209970474243
+I1201 12:35:11.037995 137274321021824 utils.py:1231] [49700] lr = 0.0006739151837716144
+I1201 12:35:11.038056 137274321021824 utils.py:1231] [49700] uptime = 312300.40041803196
+I1201 12:35:11.038115 137274321021824 utils.py:1231] [49700] examples_seen = 50892800.0
+I1201 12:35:11.038169 137274321021824 utils.py:1231] [49700] progress = 0.44137367565695407
+I1201 12:35:11.038222 137274321021824 utils.py:1231] [49700] epoch = 39.72378308214308
+I1201 12:35:11.038276 137274321021824 utils.py:1231] [49700] img/sec/core = 164.21780527539514
+I1201 12:35:11.038342 137274321021824 utils.py:1231] [49700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.71579362487442
+I1201 12:35:11.038397 137274321021824 utils.py:1231] [49700] core_hours = 86.71579362487442
+I1201 12:35:11.038460 137274321021824 train.py:125] NOTE: Steps:49700/112603 [44.1%]
+Walltime:3d14h45m (0s eval)
+ETA:4d13h45m
+Total train time:8d4h28m
+I1201 12:40:22.824622 137274321021824 utils.py:1231] [49750] l2_params = 312.204159365831
+I1201 12:40:22.824848 137274321021824 utils.py:1231] [49750] train/loss = 2.2749973982572556
+I1201 12:40:22.824972 137274321021824 utils.py:1231] [49750] l2_grads = 1.4937520027160645
+I1201 12:40:22.825044 137274321021824 utils.py:1231] [49750] lr = 0.0006731973053917094
+I1201 12:40:22.825105 137274321021824 utils.py:1231] [49750] uptime = 312612.18746683
+I1201 12:40:22.825174 137274321021824 utils.py:1231] [49750] examples_seen = 50944000.0
+I1201 12:40:22.825226 137274321021824 utils.py:1231] [49750] progress = 0.44181771356002947
+I1201 12:40:22.825280 137274321021824 utils.py:1231] [49750] epoch = 39.763746646612034
+I1201 12:40:22.825335 137274321021824 utils.py:1231] [49750] img/sec/core = 164.21464649470548
+I1201 12:40:22.825390 137274321021824 utils.py:1231] [49750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.80240113842946
+I1201 12:40:22.825443 137274321021824 utils.py:1231] [49750] core_hours = 86.80240113842946
+I1201 12:40:22.825521 137274321021824 train.py:125] NOTE: Steps:49750/112603 [44.2%]
+Walltime:3d14h50m (0s eval)
+ETA:4d13h40m
+Total train time:8d4h28m
+I1201 12:45:34.601225 137274321021824 utils.py:1231] [49800] l2_params = 312.1374188172707
+I1201 12:45:34.601432 137274321021824 utils.py:1231] [49800] train/loss = 2.4776520133018494
+I1201 12:45:34.601521 137274321021824 utils.py:1231] [49800] l2_grads = 1.4833711385726929
+I1201 12:45:34.601580 137274321021824 utils.py:1231] [49800] lr = 0.0006724790210728938
+I1201 12:45:34.601634 137274321021824 utils.py:1231] [49800] uptime = 312923.963996155
+I1201 12:45:34.601687 137274321021824 utils.py:1231] [49800] examples_seen = 50995200.0
+I1201 12:45:34.601734 137274321021824 utils.py:1231] [49800] progress = 0.44226175146310487
+I1201 12:45:34.601781 137274321021824 utils.py:1231] [49800] epoch = 39.80371021108099
+I1201 12:45:34.601831 137274321021824 utils.py:1231] [49800] img/sec/core = 164.2201871669077
+I1201 12:45:34.601891 137274321021824 utils.py:1231] [49800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.88900572990862
+I1201 12:45:34.601941 137274321021824 utils.py:1231] [49800] core_hours = 86.88900572990862
+I1201 12:45:34.602000 137274321021824 train.py:125] NOTE: Steps:49800/112603 [44.2%]
+Walltime:3d14h55m (0s eval)
+ETA:4d13h34m
+Total train time:8d4h28m
+I1201 12:50:46.314291 137274321021824 utils.py:1231] [49850] l2_params = 312.07580026425273
+I1201 12:50:46.314522 137274321021824 utils.py:1231] [49850] train/loss = 4.018667459487915
+I1201 12:50:46.314627 137274321021824 utils.py:1231] [49850] l2_grads = 1.2741284370422363
+I1201 12:50:46.314686 137274321021824 utils.py:1231] [49850] lr = 0.0006717603324986787
+I1201 12:50:46.314738 137274321021824 utils.py:1231] [49850] uptime = 313235.677099753
+I1201 12:50:46.314791 137274321021824 utils.py:1231] [49850] examples_seen = 51046400.0
+I1201 12:50:46.314839 137274321021824 utils.py:1231] [49850] progress = 0.4427057893661803
+I1201 12:50:46.314896 137274321021824 utils.py:1231] [49850] epoch = 39.843673775549945
+I1201 12:50:46.314947 137274321021824 utils.py:1231] [49850] img/sec/core = 164.25360181852548
+I1201 12:50:46.315002 137274321021824 utils.py:1231] [49850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 86.97559270313027
+I1201 12:50:46.315052 137274321021824 utils.py:1231] [49850] core_hours = 86.97559270313027
+I1201 12:50:46.315111 137274321021824 train.py:125] NOTE: Steps:49850/112603 [44.3%]
+Walltime:3d15h0m (0s eval)
+ETA:4d13h29m
+Total train time:8d4h28m
+I1201 12:55:58.069731 137274321021824 utils.py:1231] [49900] l2_params = 311.9836956513886
+I1201 12:55:58.069960 137274321021824 utils.py:1231] [49900] train/loss = 4.099775046110153
+I1201 12:55:58.070066 137274321021824 utils.py:1231] [49900] l2_grads = 1.3660249710083008
+I1201 12:55:58.070128 137274321021824 utils.py:1231] [49900] lr = 0.0006710412413535229
+I1201 12:55:58.070179 137274321021824 utils.py:1231] [49900] uptime = 313547.432541502
+I1201 12:55:58.070232 137274321021824 utils.py:1231] [49900] examples_seen = 51097600.0
+I1201 12:55:58.070284 137274321021824 utils.py:1231] [49900] progress = 0.4431498272692557
+I1201 12:55:58.070333 137274321021824 utils.py:1231] [49900] epoch = 39.88363734001891
+I1201 12:55:58.070384 137274321021824 utils.py:1231] [49900] img/sec/core = 164.2312952510436
+I1201 12:55:58.070439 137274321021824 utils.py:1231] [49900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.06219143694945
+I1201 12:55:58.070526 137274321021824 utils.py:1231] [49900] core_hours = 87.06219143694945
+I1201 12:55:58.070593 137274321021824 train.py:125] NOTE: Steps:49900/112603 [44.3%]
+Walltime:3d15h5m (0s eval)
+ETA:4d13h24m
+Total train time:8d4h28m
+I1201 13:01:09.857081 137274321021824 utils.py:1231] [49950] l2_params = 311.90484638823204
+I1201 13:01:09.857403 137274321021824 utils.py:1231] [49950] train/loss = 2.4091476052999496
+I1201 13:01:09.857579 137274321021824 utils.py:1231] [49950] l2_grads = 1.5206735134124756
+I1201 13:01:09.857669 137274321021824 utils.py:1231] [49950] lr = 0.0006703217493228278
+I1201 13:01:09.857744 137274321021824 utils.py:1231] [49950] uptime = 313859.220101388
+I1201 13:01:09.857823 137274321021824 utils.py:1231] [49950] examples_seen = 51148800.0
+I1201 13:01:09.857886 137274321021824 utils.py:1231] [49950] progress = 0.4435938651723311
+I1201 13:01:09.857963 137274321021824 utils.py:1231] [49950] epoch = 39.92360090448786
+I1201 13:01:09.858028 137274321021824 utils.py:1231] [49950] img/sec/core = 164.21437731101548
+I1201 13:01:09.858099 137274321021824 utils.py:1231] [49950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.14879909247334
+I1201 13:01:09.858206 137274321021824 utils.py:1231] [49950] core_hours = 87.14879909247334
+I1201 13:01:09.858310 137274321021824 train.py:125] NOTE: Steps:49950/112603 [44.4%]
+Walltime:3d15h10m (0s eval)
+ETA:4d13h18m
+Total train time:8d4h28m
+I1201 13:06:21.637768 137274321021824 utils.py:1231] [50000] l2_params = 311.83690505567563
+I1201 13:06:21.638024 137274321021824 utils.py:1231] [50000] train/loss = 2.3824846744537354
+I1201 13:06:21.638169 137274321021824 utils.py:1231] [50000] l2_grads = 1.4862955808639526
+I1201 13:06:21.638258 137274321021824 utils.py:1231] [50000] lr = 0.0006696018580929353
+I1201 13:06:21.638331 137274321021824 utils.py:1231] [50000] uptime = 314171.000693022
+I1201 13:06:21.638391 137274321021824 utils.py:1231] [50000] examples_seen = 51200000.0
+I1201 13:06:21.638446 137274321021824 utils.py:1231] [50000] progress = 0.4440379030754065
+I1201 13:06:21.638501 137274321021824 utils.py:1231] [50000] epoch = 39.96356446895682
+I1201 13:06:21.638557 137274321021824 utils.py:1231] [50000] img/sec/core = 164.21804747906228
+I1201 13:06:21.638616 137274321021824 utils.py:1231] [50000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.23540481237166
+I1201 13:06:21.638668 137274321021824 utils.py:1231] [50000] core_hours = 87.23540481237166
+I1201 13:06:21.638731 137274321021824 train.py:125] NOTE: Steps:50000/112603 [44.4%]
+Walltime:3d15h16m (0s eval)
+ETA:4d13h13m
+Total train time:8d4h28m
+I1201 13:06:21.985487 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:50000/112603 [44.4%]
+Walltime:3d15h16m (0s eval)
+ETA:4d13h13m
+Total train time:8d4h28m
+I1201 13:07:59.485255 137274321021824 utils.py:1231] [50000] val/acc@1 = 0.642578125
+I1201 13:07:59.485580 137274321021824 utils.py:1231] [50000] val/loss = 1.4801594344328861
+I1201 13:07:59.485856 137274321021824 utils.py:1231] [50000] z/secs/eval/val = 97.5001055339817
+I1201 13:07:59.485947 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.5001055339817
+I1201 13:13:11.262992 137274321021824 utils.py:1231] [50050] l2_params = 311.7789603417195
+I1201 13:13:11.263187 137274321021824 utils.py:1231] [50050] train/loss = 3.007276803255081
+I1201 13:13:11.263287 137274321021824 utils.py:1231] [50050] l2_grads = 1.3848490715026855
+I1201 13:13:11.263362 137274321021824 utils.py:1231] [50050] lr = 0.0006688815693511231
+I1201 13:13:11.263453 137274321021824 utils.py:1231] [50050] uptime = 314580.62581181596
+I1201 13:13:11.263525 137274321021824 utils.py:1231] [50050] examples_seen = 51251200.0
+I1201 13:13:11.263586 137274321021824 utils.py:1231] [50050] progress = 0.4444819409784819
+I1201 13:13:11.263647 137274321021824 utils.py:1231] [50050] epoch = 40.00352803342577
+I1201 13:13:11.263703 137274321021824 utils.py:1231] [50050] img/sec/core = 124.99233482249214
+I1201 13:13:11.263764 137274321021824 utils.py:1231] [50050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.3491895675922
+I1201 13:13:11.263828 137274321021824 utils.py:1231] [50050] core_hours = 87.3491895675922
+I1201 13:13:11.263897 137274321021824 train.py:125] NOTE: Steps:50050/112603 [44.4%]
+Walltime:3d15h23m (0s eval)
+ETA:4d13h10m
+Total train time:8d4h31m
+I1201 13:18:23.041786 137274321021824 utils.py:1231] [50100] l2_params = 311.6830713666178
+I1201 13:18:23.042032 137274321021824 utils.py:1231] [50100] train/loss = 2.3286951780319214
+I1201 13:18:23.042160 137274321021824 utils.py:1231] [50100] l2_grads = 1.483980417251587
+I1201 13:18:23.042244 137274321021824 utils.py:1231] [50100] lr = 0.0006681608847855986
+I1201 13:18:23.042306 137274321021824 utils.py:1231] [50100] uptime = 314892.4046678
+I1201 13:18:23.042369 137274321021824 utils.py:1231] [50100] examples_seen = 51302400.0
+I1201 13:18:23.042418 137274321021824 utils.py:1231] [50100] progress = 0.4449259788815573
+I1201 13:18:23.042467 137274321021824 utils.py:1231] [50100] epoch = 40.04349159789473
+I1201 13:18:23.042518 137274321021824 utils.py:1231] [50100] img/sec/core = 164.21896166884636
+I1201 13:18:23.042574 137274321021824 utils.py:1231] [50100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.43579480536555
+I1201 13:18:23.042627 137274321021824 utils.py:1231] [50100] core_hours = 87.43579480536555
+I1201 13:18:23.042699 137274321021824 train.py:125] NOTE: Steps:50100/112603 [44.5%]
+Walltime:3d15h28m (0s eval)
+ETA:4d13h5m
+Total train time:8d4h31m
+I1201 13:23:34.829001 137274321021824 utils.py:1231] [50150] l2_params = 311.61377976793347
+I1201 13:23:34.829197 137274321021824 utils.py:1231] [50150] train/loss = 3.3562644720077515
+I1201 13:23:34.829305 137274321021824 utils.py:1231] [50150] l2_grads = 1.4146021604537964
+I1201 13:23:34.829378 137274321021824 utils.py:1231] [50150] lr = 0.0006674398060854989
+I1201 13:23:34.829438 137274321021824 utils.py:1231] [50150] uptime = 315204.191799649
+I1201 13:23:34.829497 137274321021824 utils.py:1231] [50150] examples_seen = 51353600.0
+I1201 13:23:34.829552 137274321021824 utils.py:1231] [50150] progress = 0.4453700167846327
+I1201 13:23:34.829603 137274321021824 utils.py:1231] [50150] epoch = 40.08345516236369
+I1201 13:23:34.829659 137274321021824 utils.py:1231] [50150] img/sec/core = 164.2146027527526
+I1201 13:23:34.829720 137274321021824 utils.py:1231] [50150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.52240234199027
+I1201 13:23:34.829780 137274321021824 utils.py:1231] [50150] core_hours = 87.52240234199027
+I1201 13:23:34.829853 137274321021824 train.py:125] NOTE: Steps:50150/112603 [44.5%]
+Walltime:3d15h33m (0s eval)
+ETA:4d12h59m
+Total train time:8d4h31m
+I1201 13:28:46.707807 137274321021824 utils.py:1231] [50200] l2_params = 311.5438739618296
+I1201 13:28:46.708089 137274321021824 utils.py:1231] [50200] train/loss = 4.162283420562744
+I1201 13:28:46.708260 137274321021824 utils.py:1231] [50200] l2_grads = 1.2404506206512451
+I1201 13:28:46.708340 137274321021824 utils.py:1231] [50200] lr = 0.000666718334940885
+I1201 13:28:46.708413 137274321021824 utils.py:1231] [50200] uptime = 315516.07077577
+I1201 13:28:46.708465 137274321021824 utils.py:1231] [50200] examples_seen = 51404800.0
+I1201 13:28:46.708512 137274321021824 utils.py:1231] [50200] progress = 0.44581405468770813
+I1201 13:28:46.708559 137274321021824 utils.py:1231] [50200] epoch = 40.12341872683265
+I1201 13:28:46.708609 137274321021824 utils.py:1231] [50200] img/sec/core = 164.16624370387967
+I1201 13:28:46.708662 137274321021824 utils.py:1231] [50200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.60903539091278
+I1201 13:28:46.708711 137274321021824 utils.py:1231] [50200] core_hours = 87.60903539091278
+I1201 13:28:46.708770 137274321021824 train.py:125] NOTE: Steps:50200/112603 [44.6%]
+Walltime:3d15h38m (0s eval)
+ETA:4d12h54m
+Total train time:8d4h31m
+I1201 13:33:58.503394 137274321021824 utils.py:1231] [50250] l2_params = 311.46417954043295
+I1201 13:33:58.503592 137274321021824 utils.py:1231] [50250] train/loss = 2.434307873249054
+I1201 13:33:58.503696 137274321021824 utils.py:1231] [50250] l2_grads = 1.5186986923217773
+I1201 13:33:58.503768 137274321021824 utils.py:1231] [50250] lr = 0.0006659964730427364
+I1201 13:33:58.503836 137274321021824 utils.py:1231] [50250] uptime = 315827.866197543
+I1201 13:33:58.503902 137274321021824 utils.py:1231] [50250] examples_seen = 51456000.0
+I1201 13:33:58.503961 137274321021824 utils.py:1231] [50250] progress = 0.44625809259078353
+I1201 13:33:58.504017 137274321021824 utils.py:1231] [50250] epoch = 40.1633822913016
+I1201 13:33:58.504073 137274321021824 utils.py:1231] [50250] img/sec/core = 164.21023666369953
+I1201 13:33:58.504131 137274321021824 utils.py:1231] [50250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.69564523029416
+I1201 13:33:58.504190 137274321021824 utils.py:1231] [50250] core_hours = 87.69564523029416
+I1201 13:33:58.504255 137274321021824 train.py:125] NOTE: Steps:50250/112603 [44.6%]
+Walltime:3d15h43m (0s eval)
+ETA:4d12h49m
+Total train time:8d4h31m
+I1201 13:39:10.298043 137274321021824 utils.py:1231] [50300] l2_params = 311.39437204237237
+I1201 13:39:10.298397 137274321021824 utils.py:1231] [50300] train/loss = 2.5683362185955048
+I1201 13:39:10.298593 137274321021824 utils.py:1231] [50300] l2_grads = 1.6043564081192017
+I1201 13:39:10.298684 137274321021824 utils.py:1231] [50300] lr = 0.0006652742220829495
+I1201 13:39:10.298764 137274321021824 utils.py:1231] [50300] uptime = 316139.66112546297
+I1201 13:39:10.298833 137274321021824 utils.py:1231] [50300] examples_seen = 51507200.0
+I1201 13:39:10.298898 137274321021824 utils.py:1231] [50300] progress = 0.44670213049385893
+I1201 13:39:10.298957 137274321021824 utils.py:1231] [50300] epoch = 40.20334585577056
+I1201 13:39:10.299016 137274321021824 utils.py:1231] [50300] img/sec/core = 164.2104967568292
+I1201 13:39:10.299074 137274321021824 utils.py:1231] [50300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.78225493249415
+I1201 13:39:10.299129 137274321021824 utils.py:1231] [50300] core_hours = 87.78225493249415
+I1201 13:39:10.299194 137274321021824 train.py:125] NOTE: Steps:50300/112603 [44.7%]
+Walltime:3d15h48m (0s eval)
+ETA:4d12h44m
+Total train time:8d4h31m
+I1201 13:44:22.094842 137274321021824 utils.py:1231] [50350] l2_params = 311.33081982536385
+I1201 13:44:22.095053 137274321021824 utils.py:1231] [50350] train/loss = 2.527290105819702
+I1201 13:44:22.095170 137274321021824 utils.py:1231] [50350] l2_grads = 1.5420715808868408
+I1201 13:44:22.095238 137274321021824 utils.py:1231] [50350] lr = 0.0006645515837543321
+I1201 13:44:22.095304 137274321021824 utils.py:1231] [50350] uptime = 316451.457666155
+I1201 13:44:22.095369 137274321021824 utils.py:1231] [50350] examples_seen = 51558400.0
+I1201 13:44:22.095422 137274321021824 utils.py:1231] [50350] progress = 0.4471461683969344
+I1201 13:44:22.095477 137274321021824 utils.py:1231] [50350] epoch = 40.24330942023951
+I1201 13:44:22.095544 137274321021824 utils.py:1231] [50350] img/sec/core = 164.20964737568983
+I1201 13:44:22.095604 137274321021824 utils.py:1231] [50350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.86886508268638
+I1201 13:44:22.095670 137274321021824 utils.py:1231] [50350] core_hours = 87.86886508268638
+I1201 13:44:22.095744 137274321021824 train.py:125] NOTE: Steps:50350/112603 [44.7%]
+Walltime:3d15h54m (0s eval)
+ETA:4d12h38m
+Total train time:8d4h31m
+I1201 13:49:33.818542 137274321021824 utils.py:1231] [50400] l2_params = 311.2583398821922
+I1201 13:49:33.818786 137274321021824 utils.py:1231] [50400] train/loss = 2.5225872695446014
+I1201 13:49:33.818912 137274321021824 utils.py:1231] [50400] l2_grads = 1.5249717235565186
+I1201 13:49:33.818986 137274321021824 utils.py:1231] [50400] lr = 0.0006638285597506002
+I1201 13:49:33.819054 137274321021824 utils.py:1231] [50400] uptime = 316763.18141584296
+I1201 13:49:33.819108 137274321021824 utils.py:1231] [50400] examples_seen = 51609600.0
+I1201 13:49:33.819165 137274321021824 utils.py:1231] [50400] progress = 0.4475902063000098
+I1201 13:49:33.819217 137274321021824 utils.py:1231] [50400] epoch = 40.283272984708475
+I1201 13:49:33.819271 137274321021824 utils.py:1231] [50400] img/sec/core = 164.2479921765763
+I1201 13:49:33.819329 137274321021824 utils.py:1231] [50400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 87.95545501315526
+I1201 13:49:33.819387 137274321021824 utils.py:1231] [50400] core_hours = 87.95545501315526
+I1201 13:49:33.819458 137274321021824 train.py:125] NOTE: Steps:50400/112603 [44.8%]
+Walltime:3d15h59m (0s eval)
+ETA:4d12h33m
+Total train time:8d4h31m
+I1201 13:54:45.538801 137274321021824 utils.py:1231] [50450] l2_params = 311.1914377645262
+I1201 13:54:45.539054 137274321021824 utils.py:1231] [50450] train/loss = 2.577472597360611
+I1201 13:54:45.539171 137274321021824 utils.py:1231] [50450] l2_grads = 1.4526726007461548
+I1201 13:54:45.539241 137274321021824 utils.py:1231] [50450] lr = 0.0006631051517663737
+I1201 13:54:45.539306 137274321021824 utils.py:1231] [50450] uptime = 317074.90166399296
+I1201 13:54:45.539378 137274321021824 utils.py:1231] [50450] examples_seen = 51660800.0
+I1201 13:54:45.539428 137274321021824 utils.py:1231] [50450] progress = 0.4480342442030852
+I1201 13:54:45.539492 137274321021824 utils.py:1231] [50450] epoch = 40.32323654917743
+I1201 13:54:45.539545 137274321021824 utils.py:1231] [50450] img/sec/core = 164.24983716605632
+I1201 13:54:45.539601 137274321021824 utils.py:1231] [50450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.0420439709747
+I1201 13:54:45.539651 137274321021824 utils.py:1231] [50450] core_hours = 88.0420439709747
+I1201 13:54:45.539734 137274321021824 train.py:125] NOTE: Steps:50450/112603 [44.8%]
+Walltime:3d16h4m (0s eval)
+ETA:4d12h28m
+Total train time:8d4h30m
+I1201 13:59:57.332153 137274321021824 utils.py:1231] [50500] l2_params = 311.12940187911283
+I1201 13:59:57.332380 137274321021824 utils.py:1231] [50500] train/loss = 2.706224352121353
+I1201 13:59:57.332468 137274321021824 utils.py:1231] [50500] l2_grads = 1.4605485200881958
+I1201 13:59:57.332523 137274321021824 utils.py:1231] [50500] lr = 0.0006623813614971713
+I1201 13:59:57.332573 137274321021824 utils.py:1231] [50500] uptime = 317386.694934875
+I1201 13:59:57.332622 137274321021824 utils.py:1231] [50500] examples_seen = 51712000.0
+I1201 13:59:57.332668 137274321021824 utils.py:1231] [50500] progress = 0.4484782821061606
+I1201 13:59:57.332714 137274321021824 utils.py:1231] [50500] epoch = 40.363200113646386
+I1201 13:59:57.332760 137274321021824 utils.py:1231] [50500] img/sec/core = 164.2113694601499
+I1201 13:59:57.332812 137274321021824 utils.py:1231] [50500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.12865321288638
+I1201 13:59:57.332859 137274321021824 utils.py:1231] [50500] core_hours = 88.12865321288638
+I1201 13:59:57.332922 137274321021824 train.py:125] NOTE: Steps:50500/112603 [44.8%]
+Walltime:3d16h9m (0s eval)
+ETA:4d12h22m
+Total train time:8d4h30m
+I1201 14:05:09.122004 137274321021824 utils.py:1231] [50550] l2_params = 311.07455127300943
+I1201 14:05:09.122239 137274321021824 utils.py:1231] [50550] train/loss = 2.3740032613277435
+I1201 14:05:09.122383 137274321021824 utils.py:1231] [50550] l2_grads = 1.5257041454315186
+I1201 14:05:09.122490 137274321021824 utils.py:1231] [50550] lr = 0.0006616571906394099
+I1201 14:05:09.122577 137274321021824 utils.py:1231] [50550] uptime = 317698.484933577
+I1201 14:05:09.122664 137274321021824 utils.py:1231] [50550] examples_seen = 51763200.0
+I1201 14:05:09.122744 137274321021824 utils.py:1231] [50550] progress = 0.448922320009236
+I1201 14:05:09.122827 137274321021824 utils.py:1231] [50550] epoch = 40.40316367811534
+I1201 14:05:09.122941 137274321021824 utils.py:1231] [50550] img/sec/core = 164.21309282897957
+I1201 14:05:09.123027 137274321021824 utils.py:1231] [50550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.21526154585916
+I1201 14:05:09.123097 137274321021824 utils.py:1231] [50550] core_hours = 88.21526154585916
+I1201 14:05:09.123177 137274321021824 train.py:125] NOTE: Steps:50550/112603 [44.9%]
+Walltime:3d16h14m (0s eval)
+ETA:4d12h17m
+Total train time:8d4h30m
+I1201 14:10:20.915721 137274321021824 utils.py:1231] [50600] l2_params = 310.98397393318874
+I1201 14:10:20.915941 137274321021824 utils.py:1231] [50600] train/loss = 4.661018133163452
+I1201 14:10:20.916078 137274321021824 utils.py:1231] [50600] l2_grads = 1.515096664428711
+I1201 14:10:20.916151 137274321021824 utils.py:1231] [50600] lr = 0.0006609326408903966
+I1201 14:10:20.916214 137274321021824 utils.py:1231] [50600] uptime = 318010.278576691
+I1201 14:10:20.916272 137274321021824 utils.py:1231] [50600] examples_seen = 51814400.0
+I1201 14:10:20.916325 137274321021824 utils.py:1231] [50600] progress = 0.4493663579123114
+I1201 14:10:20.916378 137274321021824 utils.py:1231] [50600] epoch = 40.4431272425843
+I1201 14:10:20.916434 137274321021824 utils.py:1231] [50600] img/sec/core = 164.21117341793533
+I1201 14:10:20.916497 137274321021824 utils.py:1231] [50600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.3018708911686
+I1201 14:10:20.916548 137274321021824 utils.py:1231] [50600] core_hours = 88.3018708911686
+I1201 14:10:20.916610 137274321021824 train.py:125] NOTE: Steps:50600/112603 [44.9%]
+Walltime:3d16h20m (0s eval)
+ETA:4d12h12m
+Total train time:8d4h30m
+I1201 14:15:32.693215 137274321021824 utils.py:1231] [50650] l2_params = 310.9062259692084
+I1201 14:15:32.693460 137274321021824 utils.py:1231] [50650] train/loss = 2.29231296479702
+I1201 14:15:32.693580 137274321021824 utils.py:1231] [50650] l2_grads = 1.5195120573043823
+I1201 14:15:32.693652 137274321021824 utils.py:1231] [50650] lr = 0.0006602077139483273
+I1201 14:15:32.693715 137274321021824 utils.py:1231] [50650] uptime = 318322.056076395
+I1201 14:15:32.693789 137274321021824 utils.py:1231] [50650] examples_seen = 51865600.0
+I1201 14:15:32.693856 137274321021824 utils.py:1231] [50650] progress = 0.4498103958153868
+I1201 14:15:32.693922 137274321021824 utils.py:1231] [50650] epoch = 40.48309080705326
+I1201 14:15:32.693983 137274321021824 utils.py:1231] [50650] img/sec/core = 164.21967604657831
+I1201 14:15:32.694046 137274321021824 utils.py:1231] [50650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.3884757521975
+I1201 14:15:32.694102 137274321021824 utils.py:1231] [50650] core_hours = 88.3884757521975
+I1201 14:15:32.694167 137274321021824 train.py:125] NOTE: Steps:50650/112603 [45.0%]
+Walltime:3d16h25m (0s eval)
+ETA:4d12h7m
+Total train time:8d4h30m
+I1201 14:20:44.457707 137274321021824 utils.py:1231] [50700] l2_params = 310.8281592035061
+I1201 14:20:44.457951 137274321021824 utils.py:1231] [50700] train/loss = 3.7634719908237457
+I1201 14:20:44.458069 137274321021824 utils.py:1231] [50700] l2_grads = 1.31930410861969
+I1201 14:20:44.458143 137274321021824 utils.py:1231] [50700] lr = 0.0006594824115122826
+I1201 14:20:44.458191 137274321021824 utils.py:1231] [50700] uptime = 318633.82055413997
+I1201 14:20:44.458240 137274321021824 utils.py:1231] [50700] examples_seen = 51916800.0
+I1201 14:20:44.458285 137274321021824 utils.py:1231] [50700] progress = 0.4502544337184622
+I1201 14:20:44.458330 137274321021824 utils.py:1231] [50700] epoch = 40.523054371522214
+I1201 14:20:44.458377 137274321021824 utils.py:1231] [50700] img/sec/core = 164.22653526897147
+I1201 14:20:44.458428 137274321021824 utils.py:1231] [50700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.47507699601555
+I1201 14:20:44.458475 137274321021824 utils.py:1231] [50700] core_hours = 88.47507699601555
+I1201 14:20:44.458536 137274321021824 train.py:125] NOTE: Steps:50700/112603 [45.0%]
+Walltime:3d16h30m (0s eval)
+ETA:4d12h1m
+Total train time:8d4h30m
+I1201 14:25:56.183190 137274321021824 utils.py:1231] [50750] l2_params = 310.7492832909626
+I1201 14:25:56.183447 137274321021824 utils.py:1231] [50750] train/loss = 3.082583099603653
+I1201 14:25:56.183573 137274321021824 utils.py:1231] [50750] l2_grads = 1.424689769744873
+I1201 14:25:56.183665 137274321021824 utils.py:1231] [50750] lr = 0.0006587567352822212
+I1201 14:25:56.183719 137274321021824 utils.py:1231] [50750] uptime = 318945.546081189
+I1201 14:25:56.183773 137274321021824 utils.py:1231] [50750] examples_seen = 51968000.0
+I1201 14:25:56.183822 137274321021824 utils.py:1231] [50750] progress = 0.4506984716215376
+I1201 14:25:56.183871 137274321021824 utils.py:1231] [50750] epoch = 40.56301793599117
+I1201 14:25:56.183932 137274321021824 utils.py:1231] [50750] img/sec/core = 164.24705568609065
+I1201 14:25:56.183987 137274321021824 utils.py:1231] [50750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.56166742019583
+I1201 14:25:56.184036 137274321021824 utils.py:1231] [50750] core_hours = 88.56166742019583
+I1201 14:25:56.184097 137274321021824 train.py:125] NOTE: Steps:50750/112603 [45.1%]
+Walltime:3d16h35m (0s eval)
+ETA:4d11h56m
+Total train time:8d4h30m
+I1201 14:31:07.969130 137274321021824 utils.py:1231] [50800] l2_params = 310.67102086963223
+I1201 14:31:07.969349 137274321021824 utils.py:1231] [50800] train/loss = 3.1415742337703705
+I1201 14:31:07.969452 137274321021824 utils.py:1231] [50800] l2_grads = 1.3070147037506104
+I1201 14:31:07.969527 137274321021824 utils.py:1231] [50800] lr = 0.0006580306869589797
+I1201 14:31:07.969589 137274321021824 utils.py:1231] [50800] uptime = 319257.33195053396
+I1201 14:31:07.969651 137274321021824 utils.py:1231] [50800] examples_seen = 52019200.0
+I1201 14:31:07.969710 137274321021824 utils.py:1231] [50800] progress = 0.45114250952461304
+I1201 14:31:07.969769 137274321021824 utils.py:1231] [50800] epoch = 40.602981500460125
+I1201 14:31:07.969829 137274321021824 utils.py:1231] [50800] img/sec/core = 164.2152677014157
+I1201 14:31:07.969898 137274321021824 utils.py:1231] [50800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.64827460612499
+I1201 14:31:07.969957 137274321021824 utils.py:1231] [50800] core_hours = 88.64827460612499
+I1201 14:31:07.970025 137274321021824 train.py:125] NOTE: Steps:50800/112603 [45.1%]
+Walltime:3d16h40m (0s eval)
+ETA:4d11h51m
+Total train time:8d4h30m
+I1201 14:36:19.759202 137274321021824 utils.py:1231] [50850] l2_params = 310.60856528325104
+I1201 14:36:19.759443 137274321021824 utils.py:1231] [50850] train/loss = 3.704569160938263
+I1201 14:36:19.759571 137274321021824 utils.py:1231] [50850] l2_grads = 1.3321982622146606
+I1201 14:36:19.759663 137274321021824 utils.py:1231] [50850] lr = 0.0006573042682442664
+I1201 14:36:19.759731 137274321021824 utils.py:1231] [50850] uptime = 319569.122092938
+I1201 14:36:19.759799 137274321021824 utils.py:1231] [50850] examples_seen = 52070400.0
+I1201 14:36:19.759872 137274321021824 utils.py:1231] [50850] progress = 0.45158654742768845
+I1201 14:36:19.759950 137274321021824 utils.py:1231] [50850] epoch = 40.64294506492909
+I1201 14:36:19.760019 137274321021824 utils.py:1231] [50850] img/sec/core = 164.21301714424536
+I1201 14:36:19.760095 137274321021824 utils.py:1231] [50850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.734882979015
+I1201 14:36:19.760149 137274321021824 utils.py:1231] [50850] core_hours = 88.734882979015
+I1201 14:36:19.760228 137274321021824 train.py:125] NOTE: Steps:50850/112603 [45.2%]
+Walltime:3d16h46m (0s eval)
+ETA:4d11h45m
+Total train time:8d4h30m
+I1201 14:41:31.555577 137274321021824 utils.py:1231] [50900] l2_params = 310.52206109002674
+I1201 14:41:31.555808 137274321021824 utils.py:1231] [50900] train/loss = 2.2644460201263428
+I1201 14:41:31.555924 137274321021824 utils.py:1231] [50900] l2_grads = 1.5021837949752808
+I1201 14:41:31.555997 137274321021824 utils.py:1231] [50900] lr = 0.0006565774808406574
+I1201 14:41:31.556058 137274321021824 utils.py:1231] [50900] uptime = 319880.918419006
+I1201 14:41:31.556119 137274321021824 utils.py:1231] [50900] examples_seen = 52121600.0
+I1201 14:41:31.556182 137274321021824 utils.py:1231] [50900] progress = 0.45203058533076385
+I1201 14:41:31.556244 137274321021824 utils.py:1231] [50900] epoch = 40.68290862939804
+I1201 14:41:31.556310 137274321021824 utils.py:1231] [50900] img/sec/core = 164.20976040889656
+I1201 14:41:31.556374 137274321021824 utils.py:1231] [50900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.82149306958944
+I1201 14:41:31.556435 137274321021824 utils.py:1231] [50900] core_hours = 88.82149306958944
+I1201 14:41:31.556503 137274321021824 train.py:125] NOTE: Steps:50900/112603 [45.2%]
+Walltime:3d16h51m (0s eval)
+ETA:4d11h40m
+Total train time:8d4h30m
+I1201 14:46:43.350278 137274321021824 utils.py:1231] [50950] l2_params = 310.44205121372744
+I1201 14:46:43.350486 137274321021824 utils.py:1231] [50950] train/loss = 2.3721644580364227
+I1201 14:46:43.350598 137274321021824 utils.py:1231] [50950] l2_grads = 1.5004349946975708
+I1201 14:46:43.350673 137274321021824 utils.py:1231] [50950] lr = 0.0006558503264515929
+I1201 14:46:43.350730 137274321021824 utils.py:1231] [50950] uptime = 320192.713091779
+I1201 14:46:43.350785 137274321021824 utils.py:1231] [50950] examples_seen = 52172800.0
+I1201 14:46:43.350836 137274321021824 utils.py:1231] [50950] progress = 0.45247462323383925
+I1201 14:46:43.350895 137274321021824 utils.py:1231] [50950] epoch = 40.722872193867
+I1201 14:46:43.350949 137274321021824 utils.py:1231] [50950] img/sec/core = 164.21063113313028
+I1201 14:46:43.351010 137274321021824 utils.py:1231] [50950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.90810270091528
+I1201 14:46:43.351063 137274321021824 utils.py:1231] [50950] core_hours = 88.90810270091528
+I1201 14:46:43.351125 137274321021824 train.py:125] NOTE: Steps:50950/112603 [45.2%]
+Walltime:3d16h56m (0s eval)
+ETA:4d11h35m
+Total train time:8d4h30m
+I1201 14:51:55.146651 137274321021824 utils.py:1231] [51000] l2_params = 310.37876267760333
+I1201 14:51:55.146895 137274321021824 utils.py:1231] [51000] train/loss = 2.4750353693962097
+I1201 14:51:55.147021 137274321021824 utils.py:1231] [51000] l2_grads = 1.5326491594314575
+I1201 14:51:55.147091 137274321021824 utils.py:1231] [51000] lr = 0.0006551228067813735
+I1201 14:51:55.147144 137274321021824 utils.py:1231] [51000] uptime = 320504.50950605597
+I1201 14:51:55.147196 137274321021824 utils.py:1231] [51000] examples_seen = 52224000.0
+I1201 14:51:55.147246 137274321021824 utils.py:1231] [51000] progress = 0.45291866113691465
+I1201 14:51:55.147294 137274321021824 utils.py:1231] [51000] epoch = 40.76283575833595
+I1201 14:51:55.147346 137274321021824 utils.py:1231] [51000] img/sec/core = 164.2097139530128
+I1201 14:51:55.147414 137274321021824 utils.py:1231] [51000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 88.99471281599222
+I1201 14:51:55.147469 137274321021824 utils.py:1231] [51000] core_hours = 88.99471281599222
+I1201 14:51:55.147546 137274321021824 train.py:125] NOTE: Steps:51000/112603 [45.3%]
+Walltime:3d17h1m (0s eval)
+ETA:4d11h30m
+Total train time:8d4h29m
+I1201 14:57:07.320754 137274321021824 utils.py:1231] [51050] l2_params = 310.29614681225877
+I1201 14:57:07.321029 137274321021824 utils.py:1231] [51050] train/loss = 4.336691796779633
+I1201 14:57:07.321167 137274321021824 utils.py:1231] [51050] l2_grads = 1.2629342079162598
+I1201 14:57:07.321256 137274321021824 utils.py:1231] [51050] lr = 0.0006543949235351564
+I1201 14:57:07.321327 137274321021824 utils.py:1231] [51050] uptime = 320816.683687807
+I1201 14:57:07.321398 137274321021824 utils.py:1231] [51050] examples_seen = 52275200.0
+I1201 14:57:07.321463 137274321021824 utils.py:1231] [51050] progress = 0.45336269903999005
+I1201 14:57:07.321526 137274321021824 utils.py:1231] [51050] epoch = 40.80279932280491
+I1201 14:57:07.321586 137274321021824 utils.py:1231] [51050] img/sec/core = 164.01100088678442
+I1201 14:57:07.321672 137274321021824 utils.py:1231] [51050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.0814278664786
+I1201 14:57:07.321738 137274321021824 utils.py:1231] [51050] core_hours = 89.0814278664786
+I1201 14:57:07.321815 137274321021824 train.py:125] NOTE: Steps:51050/112603 [45.3%]
+Walltime:3d17h6m (0s eval)
+ETA:4d11h24m
+Total train time:8d4h29m
+I1201 15:02:19.102359 137274321021824 utils.py:1231] [51100] l2_params = 310.21432016644684
+I1201 15:02:19.102592 137274321021824 utils.py:1231] [51100] train/loss = 2.5129427313804626
+I1201 15:02:19.102739 137274321021824 utils.py:1231] [51100] l2_grads = 1.4471714496612549
+I1201 15:02:19.102839 137274321021824 utils.py:1231] [51100] lr = 0.00065366667841895
+I1201 15:02:19.102934 137274321021824 utils.py:1231] [51100] uptime = 321128.465289745
+I1201 15:02:19.103021 137274321021824 utils.py:1231] [51100] examples_seen = 52326400.0
+I1201 15:02:19.103101 137274321021824 utils.py:1231] [51100] progress = 0.45380673694306545
+I1201 15:02:19.103188 137274321021824 utils.py:1231] [51100] epoch = 40.84276288727387
+I1201 15:02:19.103261 137274321021824 utils.py:1231] [51100] img/sec/core = 164.21751534325486
+I1201 15:02:19.103335 137274321021824 utils.py:1231] [51100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.16803386701694
+I1201 15:02:19.103407 137274321021824 utils.py:1231] [51100] core_hours = 89.16803386701694
+I1201 15:02:19.103488 137274321021824 train.py:125] NOTE: Steps:51100/112603 [45.4%]
+Walltime:3d17h12m (0s eval)
+ETA:4d11h19m
+Total train time:8d4h29m
+I1201 15:07:30.903170 137274321021824 utils.py:1231] [51150] l2_params = 310.1341881430696
+I1201 15:07:30.903375 137274321021824 utils.py:1231] [51150] train/loss = 2.3934989869594574
+I1201 15:07:30.903471 137274321021824 utils.py:1231] [51150] l2_grads = 1.5507417917251587
+I1201 15:07:30.903540 137274321021824 utils.py:1231] [51150] lr = 0.0006529380731396121
+I1201 15:07:30.903599 137274321021824 utils.py:1231] [51150] uptime = 321440.265960826
+I1201 15:07:30.903663 137274321021824 utils.py:1231] [51150] examples_seen = 52377600.0
+I1201 15:07:30.903720 137274321021824 utils.py:1231] [51150] progress = 0.45425077484614085
+I1201 15:07:30.903777 137274321021824 utils.py:1231] [51150] epoch = 40.882726451742826
+I1201 15:07:30.903836 137274321021824 utils.py:1231] [51150] img/sec/core = 164.20747210867648
+I1201 15:07:30.903903 137274321021824 utils.py:1231] [51150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.25464516453944
+I1201 15:07:30.903960 137274321021824 utils.py:1231] [51150] core_hours = 89.25464516453944
+I1201 15:07:30.904023 137274321021824 train.py:125] NOTE: Steps:51150/112603 [45.4%]
+Walltime:3d17h17m (0s eval)
+ETA:4d11h14m
+Total train time:8d4h29m
+I1201 15:12:42.698074 137274321021824 utils.py:1231] [51200] l2_params = 310.05729161488
+I1201 15:12:42.698292 137274321021824 utils.py:1231] [51200] train/loss = 2.3504397571086884
+I1201 15:12:42.698394 137274321021824 utils.py:1231] [51200] l2_grads = 1.4727979898452759
+I1201 15:12:42.698467 137274321021824 utils.py:1231] [51200] lr = 0.0006522091094048428
+I1201 15:12:42.698531 137274321021824 utils.py:1231] [51200] uptime = 321752.060891842
+I1201 15:12:42.698593 137274321021824 utils.py:1231] [51200] examples_seen = 52428800.0
+I1201 15:12:42.698649 137274321021824 utils.py:1231] [51200] progress = 0.45469481274921625
+I1201 15:12:42.698705 137274321021824 utils.py:1231] [51200] epoch = 40.92269001621178
+I1201 15:12:42.698764 137274321021824 utils.py:1231] [51200] img/sec/core = 164.21049512628028
+I1201 15:12:42.698824 137274321021824 utils.py:1231] [51200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.34125486759945
+I1201 15:12:42.698878 137274321021824 utils.py:1231] [51200] core_hours = 89.34125486759945
+I1201 15:12:42.698955 137274321021824 train.py:125] NOTE: Steps:51200/112603 [45.5%]
+Walltime:3d17h22m (0s eval)
+ETA:4d11h8m
+Total train time:8d4h29m
+I1201 15:17:54.493698 137274321021824 utils.py:1231] [51250] l2_params = 309.98781139038647
+I1201 15:17:54.493918 137274321021824 utils.py:1231] [51250] train/loss = 3.843371242284775
+I1201 15:17:54.494023 137274321021824 utils.py:1231] [51250] l2_grads = 1.366582989692688
+I1201 15:17:54.494089 137274321021824 utils.py:1231] [51250] lr = 0.0006514797889231838
+I1201 15:17:54.494142 137274321021824 utils.py:1231] [51250] uptime = 322063.856503241
+I1201 15:17:54.494195 137274321021824 utils.py:1231] [51250] examples_seen = 52480000.0
+I1201 15:17:54.494254 137274321021824 utils.py:1231] [51250] progress = 0.4551388506522917
+I1201 15:17:54.494310 137274321021824 utils.py:1231] [51250] epoch = 40.96265358068074
+I1201 15:17:54.494366 137274321021824 utils.py:1231] [51250] img/sec/core = 164.21013679529085
+I1201 15:17:54.494427 137274321021824 utils.py:1231] [51250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.42786475965472
+I1201 15:17:54.494482 137274321021824 utils.py:1231] [51250] core_hours = 89.42786475965472
+I1201 15:17:54.494548 137274321021824 train.py:125] NOTE: Steps:51250/112603 [45.5%]
+Walltime:3d17h27m (0s eval)
+ETA:4d11h3m
+Total train time:8d4h29m
+I1201 15:23:06.280856 137274321021824 utils.py:1231] [51300] l2_params = 309.890335379366
+I1201 15:23:06.281063 137274321021824 utils.py:1231] [51300] train/loss = 2.3300329744815826
+I1201 15:23:06.281176 137274321021824 utils.py:1231] [51300] l2_grads = 1.562292456626892
+I1201 15:23:06.281254 137274321021824 utils.py:1231] [51300] lr = 0.0006507501134040118
+I1201 15:23:06.281328 137274321021824 utils.py:1231] [51300] uptime = 322375.64368867397
+I1201 15:23:06.281398 137274321021824 utils.py:1231] [51300] examples_seen = 52531200.0
+I1201 15:23:06.281455 137274321021824 utils.py:1231] [51300] progress = 0.4555828885553671
+I1201 15:23:06.281518 137274321021824 utils.py:1231] [51300] epoch = 41.00261714514969
+I1201 15:23:06.281581 137274321021824 utils.py:1231] [51300] img/sec/core = 164.21457453070042
+I1201 15:23:06.281672 137274321021824 utils.py:1231] [51300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.51447231116389
+I1201 15:23:06.281733 137274321021824 utils.py:1231] [51300] core_hours = 89.51447231116389
+I1201 15:23:06.281797 137274321021824 train.py:125] NOTE: Steps:51300/112603 [45.6%]
+Walltime:3d17h32m (0s eval)
+ETA:4d10h58m
+Total train time:8d4h29m
+I1201 15:28:18.050032 137274321021824 utils.py:1231] [51350] l2_params = 309.80886006665196
+I1201 15:28:18.050307 137274321021824 utils.py:1231] [51350] train/loss = 2.6476220786571503
+I1201 15:28:18.050421 137274321021824 utils.py:1231] [51350] l2_grads = 1.5397285223007202
+I1201 15:28:18.050501 137274321021824 utils.py:1231] [51350] lr = 0.0006500200845575373
+I1201 15:28:18.050562 137274321021824 utils.py:1231] [51350] uptime = 322687.412923761
+I1201 15:28:18.050616 137274321021824 utils.py:1231] [51350] examples_seen = 52582400.0
+I1201 15:28:18.050666 137274321021824 utils.py:1231] [51350] progress = 0.4560269264584425
+I1201 15:28:18.050720 137274321021824 utils.py:1231] [51350] epoch = 41.042580709618655
+I1201 15:28:18.050771 137274321021824 utils.py:1231] [51350] img/sec/core = 164.224029307147
+I1201 15:28:18.050835 137274321021824 utils.py:1231] [51350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.60107487646583
+I1201 15:28:18.050896 137274321021824 utils.py:1231] [51350] core_hours = 89.60107487646583
+I1201 15:28:18.050966 137274321021824 train.py:125] NOTE: Steps:51350/112603 [45.6%]
+Walltime:3d17h38m (0s eval)
+ETA:4d10h53m
+Total train time:8d4h29m
+I1201 15:33:29.855088 137274321021824 utils.py:1231] [51400] l2_params = 309.7399417078872
+I1201 15:33:29.855320 137274321021824 utils.py:1231] [51400] train/loss = 2.3984117209911346
+I1201 15:33:29.855439 137274321021824 utils.py:1231] [51400] l2_grads = 1.5069547891616821
+I1201 15:33:29.855530 137274321021824 utils.py:1231] [51400] lr = 0.0006492897040947973
+I1201 15:33:29.855586 137274321021824 utils.py:1231] [51400] uptime = 322999.217948629
+I1201 15:33:29.855648 137274321021824 utils.py:1231] [51400] examples_seen = 52633600.0
+I1201 15:33:29.855702 137274321021824 utils.py:1231] [51400] progress = 0.4564709643615179
+I1201 15:33:29.855756 137274321021824 utils.py:1231] [51400] epoch = 41.08254427408761
+I1201 15:33:29.855812 137274321021824 utils.py:1231] [51400] img/sec/core = 164.20517925160476
+I1201 15:33:29.855870 137274321021824 utils.py:1231] [51400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.68768738337361
+I1201 15:33:29.855923 137274321021824 utils.py:1231] [51400] core_hours = 89.68768738337361
+I1201 15:33:29.855986 137274321021824 train.py:125] NOTE: Steps:51400/112603 [45.6%]
+Walltime:3d17h43m (0s eval)
+ETA:4d10h47m
+Total train time:8d4h29m
+I1201 15:38:41.629202 137274321021824 utils.py:1231] [51450] l2_params = 309.66064661296184
+I1201 15:38:41.629432 137274321021824 utils.py:1231] [51450] train/loss = 2.3635638058185577
+I1201 15:38:41.629532 137274321021824 utils.py:1231] [51450] l2_grads = 1.5225179195404053
+I1201 15:38:41.629610 137274321021824 utils.py:1231] [51450] lr = 0.0006485589737276544
+I1201 15:38:41.629670 137274321021824 utils.py:1231] [51450] uptime = 323310.99203197
+I1201 15:38:41.629730 137274321021824 utils.py:1231] [51450] examples_seen = 52684800.0
+I1201 15:38:41.629787 137274321021824 utils.py:1231] [51450] progress = 0.4569150022645933
+I1201 15:38:41.629843 137274321021824 utils.py:1231] [51450] epoch = 41.122507838556565
+I1201 15:38:41.629914 137274321021824 utils.py:1231] [51450] img/sec/core = 164.22147553553853
+I1201 15:38:41.629980 137274321021824 utils.py:1231] [51450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.77429129541277
+I1201 15:38:41.630032 137274321021824 utils.py:1231] [51450] core_hours = 89.77429129541277
+I1201 15:38:41.630097 137274321021824 train.py:125] NOTE: Steps:51450/112603 [45.7%]
+Walltime:3d17h48m (0s eval)
+ETA:4d10h42m
+Total train time:8d4h29m
+I1201 15:43:53.415775 137274321021824 utils.py:1231] [51500] l2_params = 309.58643613194624
+I1201 15:43:53.415982 137274321021824 utils.py:1231] [51500] train/loss = 2.664214789867401
+I1201 15:43:53.416087 137274321021824 utils.py:1231] [51500] l2_grads = 1.4087719917297363
+I1201 15:43:53.416147 137274321021824 utils.py:1231] [51500] lr = 0.0006478278951687896
+I1201 15:43:53.416198 137274321021824 utils.py:1231] [51500] uptime = 323622.778560791
+I1201 15:43:53.416251 137274321021824 utils.py:1231] [51500] examples_seen = 52736000.0
+I1201 15:43:53.416300 137274321021824 utils.py:1231] [51500] progress = 0.4573590401676687
+I1201 15:43:53.416351 137274321021824 utils.py:1231] [51500] epoch = 41.16247140302552
+I1201 15:43:53.416401 137274321021824 utils.py:1231] [51500] img/sec/core = 164.21492036109305
+I1201 15:43:53.416457 137274321021824 utils.py:1231] [51500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.86089866452971
+I1201 15:43:53.416510 137274321021824 utils.py:1231] [51500] core_hours = 89.86089866452971
+I1201 15:43:53.416575 137274321021824 train.py:125] NOTE: Steps:51500/112603 [45.7%]
+Walltime:3d17h53m (0s eval)
+ETA:4d10h37m
+Total train time:8d4h29m
+I1201 15:49:05.145840 137274321021824 utils.py:1231] [51550] l2_params = 309.5279564176293
+I1201 15:49:05.146106 137274321021824 utils.py:1231] [51550] train/loss = 3.9881982803344727
+I1201 15:49:05.146312 137274321021824 utils.py:1231] [51550] l2_grads = 1.269601821899414
+I1201 15:49:05.146426 137274321021824 utils.py:1231] [51550] lr = 0.0006470964701317015
+I1201 15:49:05.146511 137274321021824 utils.py:1231] [51550] uptime = 323934.50886028097
+I1201 15:49:05.146586 137274321021824 utils.py:1231] [51550] examples_seen = 52787200.0
+I1201 15:49:05.146660 137274321021824 utils.py:1231] [51550] progress = 0.4578030780707441
+I1201 15:49:05.146740 137274321021824 utils.py:1231] [51550] epoch = 41.20243496749448
+I1201 15:49:05.146806 137274321021824 utils.py:1231] [51550] img/sec/core = 164.2445411426736
+I1201 15:49:05.146874 137274321021824 utils.py:1231] [51550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 89.94749041438804
+I1201 15:49:05.146971 137274321021824 utils.py:1231] [51550] core_hours = 89.94749041438804
+I1201 15:49:05.147051 137274321021824 train.py:125] NOTE: Steps:51550/112603 [45.8%]
+Walltime:3d17h58m (0s eval)
+ETA:4d10h31m
+Total train time:8d4h29m
+I1201 15:54:16.929479 137274321021824 utils.py:1231] [51600] l2_params = 309.4463340200968
+I1201 15:54:16.929726 137274321021824 utils.py:1231] [51600] train/loss = 2.4610188007354736
+I1201 15:54:16.929847 137274321021824 utils.py:1231] [51600] l2_grads = 1.4466480016708374
+I1201 15:54:16.929922 137274321021824 utils.py:1231] [51600] lr = 0.0006463647003307003
+I1201 15:54:16.930018 137274321021824 utils.py:1231] [51600] uptime = 324246.29237293696
+I1201 15:54:16.930098 137274321021824 utils.py:1231] [51600] examples_seen = 52838400.0
+I1201 15:54:16.930164 137274321021824 utils.py:1231] [51600] progress = 0.4582471159738195
+I1201 15:54:16.930212 137274321021824 utils.py:1231] [51600] epoch = 41.24239853196344
+I1201 15:54:16.930262 137274321021824 utils.py:1231] [51600] img/sec/core = 164.2165089611115
+I1201 15:54:16.930316 137274321021824 utils.py:1231] [51600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.03409694568137
+I1201 15:54:16.930376 137274321021824 utils.py:1231] [51600] core_hours = 90.03409694568137
+I1201 15:54:16.930433 137274321021824 train.py:125] NOTE: Steps:51600/112603 [45.8%]
+Walltime:3d18h4m (0s eval)
+ETA:4d10h26m
+Total train time:8d4h28m
+I1201 15:59:28.715329 137274321021824 utils.py:1231] [51650] l2_params = 309.3605398936198
+I1201 15:59:28.715594 137274321021824 utils.py:1231] [51650] train/loss = 2.318826586008072
+I1201 15:59:28.715725 137274321021824 utils.py:1231] [51650] l2_grads = 1.5869176387786865
+I1201 15:59:28.715834 137274321021824 utils.py:1231] [51650] lr = 0.0006456325874809035
+I1201 15:59:28.715903 137274321021824 utils.py:1231] [51650] uptime = 324558.078263801
+I1201 15:59:28.715969 137274321021824 utils.py:1231] [51650] examples_seen = 52889600.0
+I1201 15:59:28.716025 137274321021824 utils.py:1231] [51650] progress = 0.4586911538768949
+I1201 15:59:28.716081 137274321021824 utils.py:1231] [51650] epoch = 41.282362096432394
+I1201 15:59:28.716137 137274321021824 utils.py:1231] [51650] img/sec/core = 164.2152563674786
+I1201 15:59:28.716198 137274321021824 utils.py:1231] [51650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.12070413758805
+I1201 15:59:28.716253 137274321021824 utils.py:1231] [51650] core_hours = 90.12070413758805
+I1201 15:59:28.716320 137274321021824 train.py:125] NOTE: Steps:51650/112603 [45.9%]
+Walltime:3d18h9m (0s eval)
+ETA:4d10h21m
+Total train time:8d4h28m
+I1201 16:04:40.509082 137274321021824 utils.py:1231] [51700] l2_params = 309.2841044039181
+I1201 16:04:40.509298 137274321021824 utils.py:1231] [51700] train/loss = 2.6267766654491425
+I1201 16:04:40.509405 137274321021824 utils.py:1231] [51700] l2_grads = 1.508642554283142
+I1201 16:04:40.509482 137274321021824 utils.py:1231] [51700] lr = 0.0006449001332982336
+I1201 16:04:40.509552 137274321021824 utils.py:1231] [51700] uptime = 324869.871913124
+I1201 16:04:40.509616 137274321021824 utils.py:1231] [51700] examples_seen = 52940800.0
+I1201 16:04:40.509674 137274321021824 utils.py:1231] [51700] progress = 0.45913519177997036
+I1201 16:04:40.509732 137274321021824 utils.py:1231] [51700] epoch = 41.32232566090135
+I1201 16:04:40.509793 137274321021824 utils.py:1231] [51700] img/sec/core = 164.21117014785906
+I1201 16:04:40.509859 137274321021824 utils.py:1231] [51700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.20731348462222
+I1201 16:04:40.509919 137274321021824 utils.py:1231] [51700] core_hours = 90.20731348462222
+I1201 16:04:40.509987 137274321021824 train.py:125] NOTE: Steps:51700/112603 [45.9%]
+Walltime:3d18h14m (0s eval)
+ETA:4d10h16m
+Total train time:8d4h28m
+I1201 16:09:52.295327 137274321021824 utils.py:1231] [51750] l2_params = 309.1973925676044
+I1201 16:09:52.295545 137274321021824 utils.py:1231] [51750] train/loss = 3.9450021386146545
+I1201 16:09:52.295687 137274321021824 utils.py:1231] [51750] l2_grads = 1.2321738004684448
+I1201 16:09:52.295790 137274321021824 utils.py:1231] [51750] lr = 0.0006441673394994124
+I1201 16:09:52.295876 137274321021824 utils.py:1231] [51750] uptime = 325181.658232619
+I1201 16:09:52.295969 137274321021824 utils.py:1231] [51750] examples_seen = 52992000.0
+I1201 16:09:52.296055 137274321021824 utils.py:1231] [51750] progress = 0.45957922968304576
+I1201 16:09:52.296132 137274321021824 utils.py:1231] [51750] epoch = 41.362289225370304
+I1201 16:09:52.296204 137274321021824 utils.py:1231] [51750] img/sec/core = 164.21503061111827
+I1201 16:09:52.296279 137274321021824 utils.py:1231] [51750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.29392079559305
+I1201 16:09:52.296348 137274321021824 utils.py:1231] [51750] core_hours = 90.29392079559305
+I1201 16:09:52.296427 137274321021824 train.py:125] NOTE: Steps:51750/112603 [46.0%]
+Walltime:3d18h19m (0s eval)
+ETA:4d10h10m
+Total train time:8d4h28m
+I1201 16:15:04.074530 137274321021824 utils.py:1231] [51800] l2_params = 309.1284335121579
+I1201 16:15:04.074747 137274321021824 utils.py:1231] [51800] train/loss = 2.3011531978845596
+I1201 16:15:04.074874 137274321021824 utils.py:1231] [51800] l2_grads = 1.486930012702942
+I1201 16:15:04.074951 137274321021824 utils.py:1231] [51800] lr = 0.0006434342078019578
+I1201 16:15:04.075010 137274321021824 utils.py:1231] [51800] uptime = 325493.43737118
+I1201 16:15:04.075069 137274321021824 utils.py:1231] [51800] examples_seen = 53043200.0
+I1201 16:15:04.075139 137274321021824 utils.py:1231] [51800] progress = 0.46002326758612117
+I1201 16:15:04.075194 137274321021824 utils.py:1231] [51800] epoch = 41.40225278983927
+I1201 16:15:04.075251 137274321021824 utils.py:1231] [51800] img/sec/core = 164.21881283113646
+I1201 16:15:04.075313 137274321021824 utils.py:1231] [51800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.38052611186
+I1201 16:15:04.075368 137274321021824 utils.py:1231] [51800] core_hours = 90.38052611186
+I1201 16:15:04.075433 137274321021824 train.py:125] NOTE: Steps:51800/112603 [46.0%]
+Walltime:3d18h24m (0s eval)
+ETA:4d10h5m
+Total train time:8d4h28m
+I1201 16:20:15.793449 137274321021824 utils.py:1231] [51850] l2_params = 309.0364214622317
+I1201 16:20:15.793674 137274321021824 utils.py:1231] [51850] train/loss = 4.533430099487305
+I1201 16:20:15.793796 137274321021824 utils.py:1231] [51850] l2_grads = 1.325538992881775
+I1201 16:20:15.793896 137274321021824 utils.py:1231] [51850] lr = 0.0006427007399241815
+I1201 16:20:15.793957 137274321021824 utils.py:1231] [51850] uptime = 325805.156317845
+I1201 16:20:15.794013 137274321021824 utils.py:1231] [51850] examples_seen = 53094400.0
+I1201 16:20:15.794064 137274321021824 utils.py:1231] [51850] progress = 0.46046730548919657
+I1201 16:20:15.794114 137274321021824 utils.py:1231] [51850] epoch = 41.44221635430822
+I1201 16:20:15.794172 137274321021824 utils.py:1231] [51850] img/sec/core = 164.2505229398958
+I1201 16:20:15.794233 137274321021824 utils.py:1231] [51850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.46711470815582
+I1201 16:20:15.794292 137274321021824 utils.py:1231] [51850] core_hours = 90.46711470815582
+I1201 16:20:15.794355 137274321021824 train.py:125] NOTE: Steps:51850/112603 [46.0%]
+Walltime:3d18h30m (0s eval)
+ETA:4d10h0m
+Total train time:8d4h28m
+I1201 16:25:27.584479 137274321021824 utils.py:1231] [51900] l2_params = 308.9572374605244
+I1201 16:25:27.584710 137274321021824 utils.py:1231] [51900] train/loss = 2.864687740802765
+I1201 16:25:27.584805 137274321021824 utils.py:1231] [51900] l2_grads = 1.3698960542678833
+I1201 16:25:27.584866 137274321021824 utils.py:1231] [51900] lr = 0.0006419669375851794
+I1201 16:25:27.584932 137274321021824 utils.py:1231] [51900] uptime = 326116.947293589
+I1201 16:25:27.584985 137274321021824 utils.py:1231] [51900] examples_seen = 53145600.0
+I1201 16:25:27.585033 137274321021824 utils.py:1231] [51900] progress = 0.46091134339227197
+I1201 16:25:27.585084 137274321021824 utils.py:1231] [51900] epoch = 41.48217991877718
+I1201 16:25:27.585135 137274321021824 utils.py:1231] [51900] img/sec/core = 164.21257824355683
+I1201 16:25:27.585196 137274321021824 utils.py:1231] [51900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.55372331252917
+I1201 16:25:27.585247 137274321021824 utils.py:1231] [51900] core_hours = 90.55372331252917
+I1201 16:25:27.585307 137274321021824 train.py:125] NOTE: Steps:51900/112603 [46.1%]
+Walltime:3d18h35m (0s eval)
+ETA:4d9h55m
+Total train time:8d4h28m
+I1201 16:30:39.363935 137274321021824 utils.py:1231] [51950] l2_params = 308.9027868843742
+I1201 16:30:39.364150 137274321021824 utils.py:1231] [51950] train/loss = 3.1428568363189697
+I1201 16:30:39.364250 137274321021824 utils.py:1231] [51950] l2_grads = 1.3634668588638306
+I1201 16:30:39.364332 137274321021824 utils.py:1231] [51950] lr = 0.0006412328025048349
+I1201 16:30:39.364420 137274321021824 utils.py:1231] [51950] uptime = 326428.726779382
+I1201 16:30:39.364483 137274321021824 utils.py:1231] [51950] examples_seen = 53196800.0
+I1201 16:30:39.364540 137274321021824 utils.py:1231] [51950] progress = 0.46135538129534737
+I1201 16:30:39.364597 137274321021824 utils.py:1231] [51950] epoch = 41.52214348324613
+I1201 16:30:39.364658 137274321021824 utils.py:1231] [51950] img/sec/core = 164.21862993897645
+I1201 16:30:39.364727 137274321021824 utils.py:1231] [51950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.64032872524943
+I1201 16:30:39.364800 137274321021824 utils.py:1231] [51950] core_hours = 90.64032872524943
+I1201 16:30:39.364869 137274321021824 train.py:125] NOTE: Steps:51950/112603 [46.1%]
+Walltime:3d18h40m (0s eval)
+ETA:4d9h49m
+Total train time:8d4h28m
+I1201 16:35:51.142934 137274321021824 utils.py:1231] [52000] l2_params = 308.8051124118363
+I1201 16:35:51.143173 137274321021824 utils.py:1231] [52000] train/loss = 2.6488751769065857
+I1201 16:35:51.143312 137274321021824 utils.py:1231] [52000] l2_grads = 1.4606684446334839
+I1201 16:35:51.143399 137274321021824 utils.py:1231] [52000] lr = 0.0006404983364038094
+I1201 16:35:51.143468 137274321021824 utils.py:1231] [52000] uptime = 326740.505829598
+I1201 16:35:51.143530 137274321021824 utils.py:1231] [52000] examples_seen = 53248000.0
+I1201 16:35:51.143588 137274321021824 utils.py:1231] [52000] progress = 0.46179941919842277
+I1201 16:35:51.143647 137274321021824 utils.py:1231] [52000] epoch = 41.56210704771509
+I1201 16:35:51.143709 137274321021824 utils.py:1231] [52000] img/sec/core = 164.21885936378777
+I1201 16:35:51.143773 137274321021824 utils.py:1231] [52000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.72693401697612
+I1201 16:35:51.143834 137274321021824 utils.py:1231] [52000] core_hours = 90.72693401697612
+I1201 16:35:51.143914 137274321021824 train.py:125] NOTE: Steps:52000/112603 [46.2%]
+Walltime:3d18h45m (0s eval)
+ETA:4d9h44m
+Total train time:8d4h28m
+I1201 16:41:03.207138 137274321021824 utils.py:1231] [52050] l2_params = 308.7448059412516
+I1201 16:41:03.207359 137274321021824 utils.py:1231] [52050] train/loss = 4.317943751811981
+I1201 16:41:03.207459 137274321021824 utils.py:1231] [52050] l2_grads = 1.4181278944015503
+I1201 16:41:03.207528 137274321021824 utils.py:1231] [52050] lr = 0.0006397635410035409
+I1201 16:41:03.207587 137274321021824 utils.py:1231] [52050] uptime = 327052.569948535
+I1201 16:41:03.207648 137274321021824 utils.py:1231] [52050] examples_seen = 53299200.0
+I1201 16:41:03.207706 137274321021824 utils.py:1231] [52050] progress = 0.46224345710149817
+I1201 16:41:03.207763 137274321021824 utils.py:1231] [52050] epoch = 41.60207061218405
+I1201 16:41:03.207817 137274321021824 utils.py:1231] [52050] img/sec/core = 164.06884641016728
+I1201 16:41:03.207878 137274321021824 utils.py:1231] [52050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.81361849445861
+I1201 16:41:03.207941 137274321021824 utils.py:1231] [52050] core_hours = 90.81361849445861
+I1201 16:41:03.208008 137274321021824 train.py:125] NOTE: Steps:52050/112603 [46.2%]
+Walltime:3d18h50m (0s eval)
+ETA:4d9h39m
+Total train time:8d4h28m
+I1201 16:46:14.936952 137274321021824 utils.py:1231] [52100] l2_params = 308.64960243636165
+I1201 16:46:14.937171 137274321021824 utils.py:1231] [52100] train/loss = 4.4862595200538635
+I1201 16:46:14.937273 137274321021824 utils.py:1231] [52100] l2_grads = 1.319220781326294
+I1201 16:46:14.937344 137274321021824 utils.py:1231] [52100] lr = 0.0006390284180262384
+I1201 16:46:14.937406 137274321021824 utils.py:1231] [52100] uptime = 327364.29976765596
+I1201 16:46:14.937467 137274321021824 utils.py:1231] [52100] examples_seen = 53350400.0
+I1201 16:46:14.937524 137274321021824 utils.py:1231] [52100] progress = 0.46268749500457357
+I1201 16:46:14.937580 137274321021824 utils.py:1231] [52100] epoch = 41.642034176653006
+I1201 16:46:14.937638 137274321021824 utils.py:1231] [52100] img/sec/core = 164.24479424001996
+I1201 16:46:14.937701 137274321021824 utils.py:1231] [52100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.90021011088109
+I1201 16:46:14.937757 137274321021824 utils.py:1231] [52100] core_hours = 90.90021011088109
+I1201 16:46:14.937824 137274321021824 train.py:125] NOTE: Steps:52100/112603 [46.3%]
+Walltime:3d18h56m (0s eval)
+ETA:4d9h33m
+Total train time:8d4h28m
+I1201 16:51:26.667551 137274321021824 utils.py:1231] [52150] l2_params = 308.5705351000656
+I1201 16:51:26.667770 137274321021824 utils.py:1231] [52150] train/loss = 4.328365445137024
+I1201 16:51:26.667870 137274321021824 utils.py:1231] [52150] l2_grads = 1.383265733718872
+I1201 16:51:26.667940 137274321021824 utils.py:1231] [52150] lr = 0.0006382929691948795
+I1201 16:51:26.667991 137274321021824 utils.py:1231] [52150] uptime = 327676.03035325196
+I1201 16:51:26.668045 137274321021824 utils.py:1231] [52150] examples_seen = 53401600.0
+I1201 16:51:26.668097 137274321021824 utils.py:1231] [52150] progress = 0.463131532907649
+I1201 16:51:26.668145 137274321021824 utils.py:1231] [52150] epoch = 41.68199774112196
+I1201 16:51:26.668196 137274321021824 utils.py:1231] [52150] img/sec/core = 164.24439039919773
+I1201 16:51:26.668252 137274321021824 utils.py:1231] [52150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 90.98680194021333
+I1201 16:51:26.668302 137274321021824 utils.py:1231] [52150] core_hours = 90.98680194021333
+I1201 16:51:26.668365 137274321021824 train.py:125] NOTE: Steps:52150/112603 [46.3%]
+Walltime:3d19h1m (0s eval)
+ETA:4d9h28m
+Total train time:8d4h28m
+I1201 16:56:38.450837 137274321021824 utils.py:1231] [52200] l2_params = 308.50750457169465
+I1201 16:56:38.451106 137274321021824 utils.py:1231] [52200] train/loss = 2.2995269298553467
+I1201 16:56:38.451239 137274321021824 utils.py:1231] [52200] l2_grads = 1.527561068534851
+I1201 16:56:38.451337 137274321021824 utils.py:1231] [52200] lr = 0.000637557196233206
+I1201 16:56:38.451414 137274321021824 utils.py:1231] [52200] uptime = 327987.81377447397
+I1201 16:56:38.451482 137274321021824 utils.py:1231] [52200] examples_seen = 53452800.0
+I1201 16:56:38.451540 137274321021824 utils.py:1231] [52200] progress = 0.4635755708107244
+I1201 16:56:38.451610 137274321021824 utils.py:1231] [52200] epoch = 41.721961305590916
+I1201 16:56:38.451668 137274321021824 utils.py:1231] [52200] img/sec/core = 164.21655711944751
+I1201 16:56:38.451729 137274321021824 utils.py:1231] [52200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.07340844610832
+I1201 16:56:38.451786 137274321021824 utils.py:1231] [52200] core_hours = 91.07340844610832
+I1201 16:56:38.451867 137274321021824 train.py:125] NOTE: Steps:52200/112603 [46.4%]
+Walltime:3d19h6m (0s eval)
+ETA:4d9h23m
+Total train time:8d4h27m
+I1201 17:01:50.470929 137274321021824 utils.py:1231] [52250] l2_params = 308.431828091598
+I1201 17:01:50.471154 137274321021824 utils.py:1231] [52250] train/loss = 2.3434599936008453
+I1201 17:01:50.471311 137274321021824 utils.py:1231] [52250] l2_grads = 1.4936819076538086
+I1201 17:01:50.471399 137274321021824 utils.py:1231] [52250] lr = 0.0006368211008657166
+I1201 17:01:50.471461 137274321021824 utils.py:1231] [52250] uptime = 328299.833822136
+I1201 17:01:50.471527 137274321021824 utils.py:1231] [52250] examples_seen = 53504000.0
+I1201 17:01:50.471588 137274321021824 utils.py:1231] [52250] progress = 0.4640196087137998
+I1201 17:01:50.471675 137274321021824 utils.py:1231] [52250] epoch = 41.76192487005987
+I1201 17:01:50.471777 137274321021824 utils.py:1231] [52250] img/sec/core = 164.09202031614112
+I1201 17:01:50.471861 137274321021824 utils.py:1231] [52250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.16008068157
+I1201 17:01:50.471950 137274321021824 utils.py:1231] [52250] core_hours = 91.16008068157
+I1201 17:01:50.472044 137274321021824 train.py:125] NOTE: Steps:52250/112603 [46.4%]
+Walltime:3d19h11m (0s eval)
+ETA:4d9h18m
+Total train time:8d4h27m
+I1201 17:07:02.227509 137274321021824 utils.py:1231] [52300] l2_params = 308.36885271190323
+I1201 17:07:02.227724 137274321021824 utils.py:1231] [52300] train/loss = 4.558140456676483
+I1201 17:07:02.227825 137274321021824 utils.py:1231] [52300] l2_grads = 1.4299107789993286
+I1201 17:07:02.227900 137274321021824 utils.py:1231] [52300] lr = 0.0006360846848176691
+I1201 17:07:02.227964 137274321021824 utils.py:1231] [52300] uptime = 328611.590325068
+I1201 17:07:02.228024 137274321021824 utils.py:1231] [52300] examples_seen = 53555200.0
+I1201 17:07:02.228083 137274321021824 utils.py:1231] [52300] progress = 0.4644636466168752
+I1201 17:07:02.228140 137274321021824 utils.py:1231] [52300] epoch = 41.801888434528834
+I1201 17:07:02.228197 137274321021824 utils.py:1231] [52300] img/sec/core = 164.23073622675446
+I1201 17:07:02.228262 137274321021824 utils.py:1231] [52300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.24667971016221
+I1201 17:07:02.228318 137274321021824 utils.py:1231] [52300] core_hours = 91.24667971016221
+I1201 17:07:02.228390 137274321021824 train.py:125] NOTE: Steps:52300/112603 [46.4%]
+Walltime:3d19h16m (0s eval)
+ETA:4d9h12m
+Total train time:8d4h27m
+I1201 17:12:13.987780 137274321021824 utils.py:1231] [52350] l2_params = 308.28564801044405
+I1201 17:12:13.988027 137274321021824 utils.py:1231] [52350] train/loss = 3.957196891307831
+I1201 17:12:13.988161 137274321021824 utils.py:1231] [52350] l2_grads = 1.3464319705963135
+I1201 17:12:13.988225 137274321021824 utils.py:1231] [52350] lr = 0.0006353479498150711
+I1201 17:12:13.988276 137274321021824 utils.py:1231] [52350] uptime = 328923.350638367
+I1201 17:12:13.988332 137274321021824 utils.py:1231] [52350] examples_seen = 53606400.0
+I1201 17:12:13.988384 137274321021824 utils.py:1231] [52350] progress = 0.4649076845199506
+I1201 17:12:13.988431 137274321021824 utils.py:1231] [52350] epoch = 41.84185199899779
+I1201 17:12:13.988502 137274321021824 utils.py:1231] [52350] img/sec/core = 164.22872898158948
+I1201 17:12:13.988556 137274321021824 utils.py:1231] [52350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.33327979718972
+I1201 17:12:13.988604 137274321021824 utils.py:1231] [52350] core_hours = 91.33327979718972
+I1201 17:12:13.988668 137274321021824 train.py:125] NOTE: Steps:52350/112603 [46.5%]
+Walltime:3d19h22m (0s eval)
+ETA:4d9h7m
+Total train time:8d4h27m
+I1201 17:17:25.751694 137274321021824 utils.py:1231] [52400] l2_params = 308.20990726449315
+I1201 17:17:25.751943 137274321021824 utils.py:1231] [52400] train/loss = 2.7119638323783875
+I1201 17:17:25.752076 137274321021824 utils.py:1231] [52400] l2_grads = 1.4282488822937012
+I1201 17:17:25.752151 137274321021824 utils.py:1231] [52400] lr = 0.0006346108975846786
+I1201 17:17:25.752211 137274321021824 utils.py:1231] [52400] uptime = 329235.114572932
+I1201 17:17:25.752281 137274321021824 utils.py:1231] [52400] examples_seen = 53657600.0
+I1201 17:17:25.752337 137274321021824 utils.py:1231] [52400] progress = 0.465351722423026
+I1201 17:17:25.752391 137274321021824 utils.py:1231] [52400] epoch = 41.881815563466745
+I1201 17:17:25.752449 137274321021824 utils.py:1231] [52400] img/sec/core = 164.22682139753303
+I1201 17:17:25.752513 137274321021824 utils.py:1231] [52400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.41988089012445
+I1201 17:17:25.752570 137274321021824 utils.py:1231] [52400] core_hours = 91.41988089012445
+I1201 17:17:25.752640 137274321021824 train.py:125] NOTE: Steps:52400/112603 [46.5%]
+Walltime:3d19h27m (0s eval)
+ETA:4d9h2m
+Total train time:8d4h27m
+I1201 17:22:37.515621 137274321021824 utils.py:1231] [52450] l2_params = 308.13251264845826
+I1201 17:22:37.515871 137274321021824 utils.py:1231] [52450] train/loss = 2.5793131291866302
+I1201 17:22:37.516000 137274321021824 utils.py:1231] [52450] l2_grads = 1.5244053602218628
+I1201 17:22:37.516081 137274321021824 utils.py:1231] [52450] lr = 0.00063387352985399
+I1201 17:22:37.516162 137274321021824 utils.py:1231] [52450] uptime = 329546.878523252
+I1201 17:22:37.516224 137274321021824 utils.py:1231] [52450] examples_seen = 53708800.0
+I1201 17:22:37.516287 137274321021824 utils.py:1231] [52450] progress = 0.4657957603261014
+I1201 17:22:37.516366 137274321021824 utils.py:1231] [52450] epoch = 41.9217791279357
+I1201 17:22:37.516428 137274321021824 utils.py:1231] [52450] img/sec/core = 164.22681309835173
+I1201 17:22:37.516483 137274321021824 utils.py:1231] [52450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.50648198743555
+I1201 17:22:37.516537 137274321021824 utils.py:1231] [52450] core_hours = 91.50648198743555
+I1201 17:22:37.516600 137274321021824 train.py:125] NOTE: Steps:52450/112603 [46.6%]
+Walltime:3d19h32m (0s eval)
+ETA:4d8h56m
+Total train time:8d4h27m
+I1201 17:27:49.296303 137274321021824 utils.py:1231] [52500] l2_params = 308.05927166240747
+I1201 17:27:49.296585 137274321021824 utils.py:1231] [52500] train/loss = 2.5625177025794983
+I1201 17:27:49.296763 137274321021824 utils.py:1231] [52500] l2_grads = 1.5037915706634521
+I1201 17:27:49.296865 137274321021824 utils.py:1231] [52500] lr = 0.000633135848351244
+I1201 17:27:49.296933 137274321021824 utils.py:1231] [52500] uptime = 329858.659294094
+I1201 17:27:49.296995 137274321021824 utils.py:1231] [52500] examples_seen = 53760000.0
+I1201 17:27:49.297055 137274321021824 utils.py:1231] [52500] progress = 0.4662397982291768
+I1201 17:27:49.297114 137274321021824 utils.py:1231] [52500] epoch = 41.96174269240466
+I1201 17:27:49.297180 137274321021824 utils.py:1231] [52500] img/sec/core = 164.2179530883872
+I1201 17:27:49.297248 137274321021824 utils.py:1231] [52500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.59308775711389
+I1201 17:27:49.297313 137274321021824 utils.py:1231] [52500] core_hours = 91.59308775711389
+I1201 17:27:49.297399 137274321021824 train.py:125] NOTE: Steps:52500/112603 [46.6%]
+Walltime:3d19h37m (0s eval)
+ETA:4d8h51m
+Total train time:8d4h27m
+I1201 17:27:49.297528 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:52500/112603 [46.6%]
+Walltime:3d19h37m (0s eval)
+ETA:4d8h51m
+Total train time:8d4h27m
+I1201 17:29:27.137302 137274321021824 utils.py:1231] [52500] val/acc@1 = 0.6491350446428571
+I1201 17:29:27.137575 137274321021824 utils.py:1231] [52500] val/loss = 1.4428963691604382
+I1201 17:29:27.137773 137274321021824 utils.py:1231] [52500] z/secs/eval/val = 97.84013297402998
+I1201 17:29:27.137850 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.84013297402998
+I1201 17:34:38.189457 137274321021824 utils.py:1231] [52550] l2_params = 307.979731290814
+I1201 17:34:38.189718 137274321021824 utils.py:1231] [52550] train/loss = 4.049999415874481
+I1201 17:34:38.189856 137274321021824 utils.py:1231] [52550] l2_grads = 1.2709826231002808
+I1201 17:34:38.189939 137274321021824 utils.py:1231] [52550] lr = 0.0006323978548054149
+I1201 17:34:38.189999 137274321021824 utils.py:1231] [52550] uptime = 330267.552360858
+I1201 17:34:38.190056 137274321021824 utils.py:1231] [52550] examples_seen = 53811200.0
+I1201 17:34:38.190109 137274321021824 utils.py:1231] [52550] progress = 0.4666838361322522
+I1201 17:34:38.190162 137274321021824 utils.py:1231] [52550] epoch = 42.00170625687362
+I1201 17:34:38.190215 137274321021824 utils.py:1231] [52550] img/sec/core = 125.21611189252988
+I1201 17:34:38.190273 137274321021824 utils.py:1231] [52550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.70666916454834
+I1201 17:34:38.190325 137274321021824 utils.py:1231] [52550] core_hours = 91.70666916454834
+I1201 17:34:38.190386 137274321021824 train.py:125] NOTE: Steps:52550/112603 [46.7%]
+Walltime:3d19h44m (0s eval)
+ETA:4d8h48m
+Total train time:8d4h30m
+I1201 17:39:49.971137 137274321021824 utils.py:1231] [52600] l2_params = 307.9055005611969
+I1201 17:39:49.971394 137274321021824 utils.py:1231] [52600] train/loss = 3.4231128692626953
+I1201 17:39:49.971536 137274321021824 utils.py:1231] [52600] l2_grads = 1.3519624471664429
+I1201 17:39:49.971639 137274321021824 utils.py:1231] [52600] lr = 0.0006316595509462076
+I1201 17:39:49.971728 137274321021824 utils.py:1231] [52600] uptime = 330579.334085334
+I1201 17:39:49.971820 137274321021824 utils.py:1231] [52600] examples_seen = 53862400.0
+I1201 17:39:49.971908 137274321021824 utils.py:1231] [52600] progress = 0.46712787403532763
+I1201 17:39:49.971992 137274321021824 utils.py:1231] [52600] epoch = 42.04166982134257
+I1201 17:39:49.972077 137274321021824 utils.py:1231] [52600] img/sec/core = 164.2174508016972
+I1201 17:39:49.972140 137274321021824 utils.py:1231] [52600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.79327519912499
+I1201 17:39:49.972205 137274321021824 utils.py:1231] [52600] core_hours = 91.79327519912499
+I1201 17:39:49.972279 137274321021824 train.py:125] NOTE: Steps:52600/112603 [46.7%]
+Walltime:3d19h49m (0s eval)
+ETA:4d8h42m
+Total train time:8d4h30m
+I1201 17:45:01.771258 137274321021824 utils.py:1231] [52650] l2_params = 307.8346795017009
+I1201 17:45:01.771547 137274321021824 utils.py:1231] [52650] train/loss = 2.2865826338529587
+I1201 17:45:01.771736 137274321021824 utils.py:1231] [52650] l2_grads = 1.4842581748962402
+I1201 17:45:01.771839 137274321021824 utils.py:1231] [52650] lr = 0.0006309209385040544
+I1201 17:45:01.771908 137274321021824 utils.py:1231] [52650] uptime = 330891.134269138
+I1201 17:45:01.771987 137274321021824 utils.py:1231] [52650] examples_seen = 53913600.0
+I1201 17:45:01.772046 137274321021824 utils.py:1231] [52650] progress = 0.4675719119384031
+I1201 17:45:01.772103 137274321021824 utils.py:1231] [52650] epoch = 42.08163338581153
+I1201 17:45:01.772159 137274321021824 utils.py:1231] [52650] img/sec/core = 164.20772872983142
+I1201 17:45:01.772216 137274321021824 utils.py:1231] [52650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.87988636129278
+I1201 17:45:01.772264 137274321021824 utils.py:1231] [52650] core_hours = 91.87988636129278
+I1201 17:45:01.772342 137274321021824 train.py:125] NOTE: Steps:52650/112603 [46.8%]
+Walltime:3d19h54m (0s eval)
+ETA:4d8h37m
+Total train time:8d4h30m
+I1201 17:50:13.561088 137274321021824 utils.py:1231] [52700] l2_params = 307.7747254614976
+I1201 17:50:13.561269 137274321021824 utils.py:1231] [52700] train/loss = 4.888374209403992
+I1201 17:50:13.561353 137274321021824 utils.py:1231] [52700] l2_grads = 1.3593775033950806
+I1201 17:50:13.561407 137274321021824 utils.py:1231] [52700] lr = 0.0006301820192101114
+I1201 17:50:13.561459 137274321021824 utils.py:1231] [52700] uptime = 331202.923821805
+I1201 17:50:13.561513 137274321021824 utils.py:1231] [52700] examples_seen = 53964800.0
+I1201 17:50:13.561557 137274321021824 utils.py:1231] [52700] progress = 0.4680159498414785
+I1201 17:50:13.561601 137274321021824 utils.py:1231] [52700] epoch = 42.121596950280484
+I1201 17:50:13.561647 137274321021824 utils.py:1231] [52700] img/sec/core = 164.21332774636758
+I1201 17:50:13.561698 137274321021824 utils.py:1231] [52700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 91.96649457036693
+I1201 17:50:13.561742 137274321021824 utils.py:1231] [52700] core_hours = 91.96649457036693
+I1201 17:50:13.561796 137274321021824 train.py:125] NOTE: Steps:52700/112603 [46.8%]
+Walltime:3d20h0m (0s eval)
+ETA:4d8h32m
+Total train time:8d4h30m
+I1201 17:55:25.346024 137274321021824 utils.py:1231] [52750] l2_params = 307.68882412289446
+I1201 17:55:25.346232 137274321021824 utils.py:1231] [52750] train/loss = 2.3311917334795
+I1201 17:55:25.346331 137274321021824 utils.py:1231] [52750] l2_grads = 1.5647380352020264
+I1201 17:55:25.346388 137274321021824 utils.py:1231] [52750] lr = 0.0006294427947962542
+I1201 17:55:25.346437 137274321021824 utils.py:1231] [52750] uptime = 331514.708799565
+I1201 17:55:25.346488 137274321021824 utils.py:1231] [52750] examples_seen = 54016000.0
+I1201 17:55:25.346554 137274321021824 utils.py:1231] [52750] progress = 0.4684599877445539
+I1201 17:55:25.346604 137274321021824 utils.py:1231] [52750] epoch = 42.161560514749446
+I1201 17:55:25.346652 137274321021824 utils.py:1231] [52750] img/sec/core = 164.21573729383846
+I1201 17:55:25.346706 137274321021824 utils.py:1231] [52750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.0531015086336
+I1201 17:55:25.346754 137274321021824 utils.py:1231] [52750] core_hours = 92.0531015086336
+I1201 17:55:25.346812 137274321021824 train.py:125] NOTE: Steps:52750/112603 [46.8%]
+Walltime:3d20h5m (0s eval)
+ETA:4d8h27m
+Total train time:8d4h30m
+I1201 18:00:37.144536 137274321021824 utils.py:1231] [52800] l2_params = 307.5883390953902
+I1201 18:00:37.144756 137274321021824 utils.py:1231] [52800] train/loss = 3.2207144796848297
+I1201 18:00:37.144874 137274321021824 utils.py:1231] [52800] l2_grads = 1.4912317991256714
+I1201 18:00:37.144965 137274321021824 utils.py:1231] [52800] lr = 0.0006287032669950728
+I1201 18:00:37.145034 137274321021824 utils.py:1231] [52800] uptime = 331826.507393712
+I1201 18:00:37.145101 137274321021824 utils.py:1231] [52800] examples_seen = 54067200.0
+I1201 18:00:37.145171 137274321021824 utils.py:1231] [52800] progress = 0.4689040256476293
+I1201 18:00:37.145234 137274321021824 utils.py:1231] [52800] epoch = 42.2015240792184
+I1201 18:00:37.145298 137274321021824 utils.py:1231] [52800] img/sec/core = 164.20856591757823
+I1201 18:00:37.145370 137274321021824 utils.py:1231] [52800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.13971222923
+I1201 18:00:37.145435 137274321021824 utils.py:1231] [52800] core_hours = 92.13971222923
+I1201 18:00:37.145507 137274321021824 train.py:125] NOTE: Steps:52800/112603 [46.9%]
+Walltime:3d20h10m (0s eval)
+ETA:4d8h21m
+Total train time:8d4h30m
+I1201 18:05:48.933623 137274321021824 utils.py:1231] [52850] l2_params = 307.48951632870507
+I1201 18:05:48.933850 137274321021824 utils.py:1231] [52850] train/loss = 4.824789464473724
+I1201 18:05:48.933966 137274321021824 utils.py:1231] [52850] l2_grads = 1.4334748983383179
+I1201 18:05:48.934039 137274321021824 utils.py:1231] [52850] lr = 0.0006279634375398679
+I1201 18:05:48.934099 137274321021824 utils.py:1231] [52850] uptime = 332138.296460444
+I1201 18:05:48.934159 137274321021824 utils.py:1231] [52850] examples_seen = 54118400.0
+I1201 18:05:48.934214 137274321021824 utils.py:1231] [52850] progress = 0.4693480635507047
+I1201 18:05:48.934277 137274321021824 utils.py:1231] [52850] epoch = 42.24148764368736
+I1201 18:05:48.934332 137274321021824 utils.py:1231] [52850] img/sec/core = 164.21358367903008
+I1201 18:05:48.934396 137274321021824 utils.py:1231] [52850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.22632030332221
+I1201 18:05:48.934453 137274321021824 utils.py:1231] [52850] core_hours = 92.22632030332221
+I1201 18:05:48.934521 137274321021824 train.py:125] NOTE: Steps:52850/112603 [46.9%]
+Walltime:3d20h15m (0s eval)
+ETA:4d8h16m
+Total train time:8d4h30m
+I1201 18:11:00.727729 137274321021824 utils.py:1231] [52900] l2_params = 307.407253552356
+I1201 18:11:00.727967 137274321021824 utils.py:1231] [52900] train/loss = 2.3583752512931824
+I1201 18:11:00.728074 137274321021824 utils.py:1231] [52900] l2_grads = 1.4750393629074097
+I1201 18:11:00.728146 137274321021824 utils.py:1231] [52900] lr = 0.0006272233081646479
+I1201 18:11:00.728213 137274321021824 utils.py:1231] [52900] uptime = 332450.090574182
+I1201 18:11:00.728275 137274321021824 utils.py:1231] [52900] examples_seen = 54169600.0
+I1201 18:11:00.728332 137274321021824 utils.py:1231] [52900] progress = 0.4697921014537801
+I1201 18:11:00.728389 137274321021824 utils.py:1231] [52900] epoch = 42.28145120815631
+I1201 18:11:00.728455 137274321021824 utils.py:1231] [52900] img/sec/core = 164.2109255565479
+I1201 18:11:00.728515 137274321021824 utils.py:1231] [52900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.31292977936054
+I1201 18:11:00.728571 137274321021824 utils.py:1231] [52900] core_hours = 92.31292977936054
+I1201 18:11:00.728637 137274321021824 train.py:125] NOTE: Steps:52900/112603 [47.0%]
+Walltime:3d20h20m (0s eval)
+ETA:4d8h11m
+Total train time:8d4h30m
+I1201 18:16:12.510215 137274321021824 utils.py:1231] [52950] l2_params = 307.3304595706348
+I1201 18:16:12.510434 137274321021824 utils.py:1231] [52950] train/loss = 2.369665175676346
+I1201 18:16:12.510543 137274321021824 utils.py:1231] [52950] l2_grads = 1.5468322038650513
+I1201 18:16:12.510616 137274321021824 utils.py:1231] [52950] lr = 0.0006264828806041245
+I1201 18:16:12.510678 137274321021824 utils.py:1231] [52950] uptime = 332761.87303917797
+I1201 18:16:12.510746 137274321021824 utils.py:1231] [52950] examples_seen = 54220800.0
+I1201 18:16:12.510810 137274321021824 utils.py:1231] [52950] progress = 0.4702361393568555
+I1201 18:16:12.510919 137274321021824 utils.py:1231] [52950] epoch = 42.32141477262527
+I1201 18:16:12.511000 137274321021824 utils.py:1231] [52950] img/sec/core = 164.21706076594177
+I1201 18:16:12.511125 137274321021824 utils.py:1231] [52950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.39953601963721
+I1201 18:16:12.511222 137274321021824 utils.py:1231] [52950] core_hours = 92.39953601963721
+I1201 18:16:12.511314 137274321021824 train.py:125] NOTE: Steps:52950/112603 [47.0%]
+Walltime:3d20h26m (0s eval)
+ETA:4d8h6m
+Total train time:8d4h30m
+I1201 18:21:24.298000 137274321021824 utils.py:1231] [53000] l2_params = 307.2649777053273
+I1201 18:21:24.298236 137274321021824 utils.py:1231] [53000] train/loss = 2.287712901830673
+I1201 18:21:24.298372 137274321021824 utils.py:1231] [53000] l2_grads = 1.6215616464614868
+I1201 18:21:24.298456 137274321021824 utils.py:1231] [53000] lr = 0.0006257421565937073
+I1201 18:21:24.298519 137274321021824 utils.py:1231] [53000] uptime = 333073.66088144
+I1201 18:21:24.298575 137274321021824 utils.py:1231] [53000] examples_seen = 54272000.0
+I1201 18:21:24.298627 137274321021824 utils.py:1231] [53000] progress = 0.4706801772599309
+I1201 18:21:24.298680 137274321021824 utils.py:1231] [53000] epoch = 42.36137833709423
+I1201 18:21:24.298741 137274321021824 utils.py:1231] [53000] img/sec/core = 164.21422858744387
+I1201 18:21:24.298819 137274321021824 utils.py:1231] [53000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.48614375359888
+I1201 18:21:24.298872 137274321021824 utils.py:1231] [53000] core_hours = 92.48614375359888
+I1201 18:21:24.298950 137274321021824 train.py:125] NOTE: Steps:53000/112603 [47.1%]
+Walltime:3d20h31m (0s eval)
+ETA:4d8h0m
+Total train time:8d4h30m
+I1201 18:26:36.435192 137274321021824 utils.py:1231] [53050] l2_params = 307.19400492821313
+I1201 18:26:36.435422 137274321021824 utils.py:1231] [53050] train/loss = 4.980439484119415
+I1201 18:26:36.435531 137274321021824 utils.py:1231] [53050] l2_grads = 1.467261791229248
+I1201 18:26:36.435615 137274321021824 utils.py:1231] [53050] lr = 0.0006250011378695022
+I1201 18:26:36.435678 137274321021824 utils.py:1231] [53050] uptime = 333385.79803904396
+I1201 18:26:36.435739 137274321021824 utils.py:1231] [53050] examples_seen = 54323200.0
+I1201 18:26:36.435797 137274321021824 utils.py:1231] [53050] progress = 0.4711242151630063
+I1201 18:26:36.435853 137274321021824 utils.py:1231] [53050] epoch = 42.401341901563185
+I1201 18:26:36.435932 137274321021824 utils.py:1231] [53050] img/sec/core = 164.03045505065765
+I1201 18:26:36.435993 137274321021824 utils.py:1231] [53050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.57284851959999
+I1201 18:26:36.436050 137274321021824 utils.py:1231] [53050] core_hours = 92.57284851959999
+I1201 18:26:36.436115 137274321021824 train.py:125] NOTE: Steps:53050/112603 [47.1%]
+Walltime:3d20h36m (0s eval)
+ETA:4d7h55m
+Total train time:8d4h30m
+I1201 18:31:48.214628 137274321021824 utils.py:1231] [53100] l2_params = 307.1098096247169
+I1201 18:31:48.214932 137274321021824 utils.py:1231] [53100] train/loss = 2.789331704378128
+I1201 18:31:48.215109 137274321021824 utils.py:1231] [53100] l2_grads = 1.5959755182266235
+I1201 18:31:48.215194 137274321021824 utils.py:1231] [53100] lr = 0.0006242598261683039
+I1201 18:31:48.215257 137274321021824 utils.py:1231] [53100] uptime = 333697.57761942997
+I1201 18:31:48.215312 137274321021824 utils.py:1231] [53100] examples_seen = 54374400.0
+I1201 18:31:48.215362 137274321021824 utils.py:1231] [53100] progress = 0.47156825306608174
+I1201 18:31:48.215411 137274321021824 utils.py:1231] [53100] epoch = 42.44130546603214
+I1201 18:31:48.215468 137274321021824 utils.py:1231] [53100] img/sec/core = 164.2185801155117
+I1201 18:31:48.215539 137274321021824 utils.py:1231] [53100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.65945395859609
+I1201 18:31:48.215596 137274321021824 utils.py:1231] [53100] core_hours = 92.65945395859609
+I1201 18:31:48.215660 137274321021824 train.py:125] NOTE: Steps:53100/112603 [47.2%]
+Walltime:3d20h41m (0s eval)
+ETA:4d7h50m
+Total train time:8d4h29m
+I1201 18:37:00.004649 137274321021824 utils.py:1231] [53150] l2_params = 307.0240124551686
+I1201 18:37:00.004891 137274321021824 utils.py:1231] [53150] train/loss = 4.667475938796997
+I1201 18:37:00.005038 137274321021824 utils.py:1231] [53150] l2_grads = 1.3033556938171387
+I1201 18:37:00.005117 137274321021824 utils.py:1231] [53150] lr = 0.0006235182232275955
+I1201 18:37:00.005185 137274321021824 utils.py:1231] [53150] uptime = 334009.367541719
+I1201 18:37:00.005306 137274321021824 utils.py:1231] [53150] examples_seen = 54425600.0
+I1201 18:37:00.005391 137274321021824 utils.py:1231] [53150] progress = 0.47201229096915714
+I1201 18:37:00.005465 137274321021824 utils.py:1231] [53150] epoch = 42.481269030501096
+I1201 18:37:00.005535 137274321021824 utils.py:1231] [53150] img/sec/core = 164.2131330740641
+I1201 18:37:00.005609 137274321021824 utils.py:1231] [53150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.74606227034306
+I1201 18:37:00.005686 137274321021824 utils.py:1231] [53150] core_hours = 92.74606227034306
+I1201 18:37:00.005760 137274321021824 train.py:125] NOTE: Steps:53150/112603 [47.2%]
+Walltime:3d20h46m (0s eval)
+ETA:4d7h44m
+Total train time:8d4h29m
+I1201 18:42:11.797323 137274321021824 utils.py:1231] [53200] l2_params = 306.93512645255004
+I1201 18:42:11.797585 137274321021824 utils.py:1231] [53200] train/loss = 3.5836695432662964
+I1201 18:42:11.797698 137274321021824 utils.py:1231] [53200] l2_grads = 1.3747713565826416
+I1201 18:42:11.797797 137274321021824 utils.py:1231] [53200] lr = 0.0006227763307855409
+I1201 18:42:11.797878 137274321021824 utils.py:1231] [53200] uptime = 334321.16023899
+I1201 18:42:11.797951 137274321021824 utils.py:1231] [53200] examples_seen = 54476800.0
+I1201 18:42:11.798010 137274321021824 utils.py:1231] [53200] progress = 0.47245632887223254
+I1201 18:42:11.798069 137274321021824 utils.py:1231] [53200] epoch = 42.52123259497005
+I1201 18:42:11.798129 137274321021824 utils.py:1231] [53200] img/sec/core = 164.21167156298546
+I1201 18:42:11.798192 137274321021824 utils.py:1231] [53200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.83267135291833
+I1201 18:42:11.798251 137274321021824 utils.py:1231] [53200] core_hours = 92.83267135291833
+I1201 18:42:11.798320 137274321021824 train.py:125] NOTE: Steps:53200/112603 [47.2%]
+Walltime:3d20h52m (0s eval)
+ETA:4d7h39m
+Total train time:8d4h29m
+I1201 18:47:23.589049 137274321021824 utils.py:1231] [53250] l2_params = 306.8592761647694
+I1201 18:47:23.589244 137274321021824 utils.py:1231] [53250] train/loss = 3.2035230696201324
+I1201 18:47:23.589344 137274321021824 utils.py:1231] [53250] l2_grads = 1.4104169607162476
+I1201 18:47:23.589408 137274321021824 utils.py:1231] [53250] lr = 0.0006220341505809844
+I1201 18:47:23.589461 137274321021824 utils.py:1231] [53250] uptime = 334632.951823173
+I1201 18:47:23.589515 137274321021824 utils.py:1231] [53250] examples_seen = 54528000.0
+I1201 18:47:23.589565 137274321021824 utils.py:1231] [53250] progress = 0.47290036677530795
+I1201 18:47:23.589620 137274321021824 utils.py:1231] [53250] epoch = 42.561196159439014
+I1201 18:47:23.589672 137274321021824 utils.py:1231] [53250] img/sec/core = 164.21225779445595
+I1201 18:47:23.589729 137274321021824 utils.py:1231] [53250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 92.9192801263025
+I1201 18:47:23.589780 137274321021824 utils.py:1231] [53250] core_hours = 92.9192801263025
+I1201 18:47:23.589847 137274321021824 train.py:125] NOTE: Steps:53250/112603 [47.3%]
+Walltime:3d20h57m (0s eval)
+ETA:4d7h34m
+Total train time:8d4h29m
+I1201 18:52:35.393281 137274321021824 utils.py:1231] [53300] l2_params = 306.77740201395176
+I1201 18:52:35.393495 137274321021824 utils.py:1231] [53300] train/loss = 3.8410814702510834
+I1201 18:52:35.393613 137274321021824 utils.py:1231] [53300] l2_grads = 1.379684329032898
+I1201 18:52:35.393698 137274321021824 utils.py:1231] [53300] lr = 0.0006212916843534444
+I1201 18:52:35.393771 137274321021824 utils.py:1231] [53300] uptime = 334944.75612987
+I1201 18:52:35.393845 137274321021824 utils.py:1231] [53300] examples_seen = 54579200.0
+I1201 18:52:35.393918 137274321021824 utils.py:1231] [53300] progress = 0.47334440467838335
+I1201 18:52:35.393985 137274321021824 utils.py:1231] [53300] epoch = 42.60115972390797
+I1201 18:52:35.394055 137274321021824 utils.py:1231] [53300] img/sec/core = 164.20555746122955
+I1201 18:52:35.394121 137274321021824 utils.py:1231] [53300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.00589243371834
+I1201 18:52:35.394186 137274321021824 utils.py:1231] [53300] core_hours = 93.00589243371834
+I1201 18:52:35.394282 137274321021824 train.py:125] NOTE: Steps:53300/112603 [47.3%]
+Walltime:3d21h2m (0s eval)
+ETA:4d7h29m
+Total train time:8d4h29m
+I1201 18:57:47.193110 137274321021824 utils.py:1231] [53350] l2_params = 306.69151668627495
+I1201 18:57:47.193323 137274321021824 utils.py:1231] [53350] train/loss = 4.65791791677475
+I1201 18:57:47.193443 137274321021824 utils.py:1231] [53350] l2_grads = 1.3491531610488892
+I1201 18:57:47.193513 137274321021824 utils.py:1231] [53350] lr = 0.0006205489338431082
+I1201 18:57:47.193572 137274321021824 utils.py:1231] [53350] uptime = 335256.555933057
+I1201 18:57:47.193627 137274321021824 utils.py:1231] [53350] examples_seen = 54630400.0
+I1201 18:57:47.193675 137274321021824 utils.py:1231] [53350] progress = 0.47378844258145875
+I1201 18:57:47.193728 137274321021824 utils.py:1231] [53350] epoch = 42.641123288376924
+I1201 18:57:47.193798 137274321021824 utils.py:1231] [53350] img/sec/core = 164.20792917979782
+I1201 18:57:47.193858 137274321021824 utils.py:1231] [53350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.09250349015915
+I1201 18:57:47.193911 137274321021824 utils.py:1231] [53350] core_hours = 93.09250349015915
+I1201 18:57:47.193968 137274321021824 train.py:125] NOTE: Steps:53350/112603 [47.4%]
+Walltime:3d21h7m (0s eval)
+ETA:4d7h23m
+Total train time:8d4h29m
+I1201 19:02:59.000747 137274321021824 utils.py:1231] [53400] l2_params = 306.61666324640737
+I1201 19:02:59.249306 137274321021824 utils.py:1231] [53400] train/loss = 2.2675550431013107
+I1201 19:02:59.249679 137274321021824 utils.py:1231] [53400] l2_grads = 1.5650731325149536
+I1201 19:02:59.249799 137274321021824 utils.py:1231] [53400] lr = 0.0006198059007908309
+I1201 19:02:59.249876 137274321021824 utils.py:1231] [53400] uptime = 335568.612226928
+I1201 19:02:59.249964 137274321021824 utils.py:1231] [53400] examples_seen = 54681600.0
+I1201 19:02:59.250031 137274321021824 utils.py:1231] [53400] progress = 0.47423248048453415
+I1201 19:02:59.250090 137274321021824 utils.py:1231] [53400] epoch = 42.68108685284588
+I1201 19:02:59.250152 137274321021824 utils.py:1231] [53400] img/sec/core = 164.07296057025525
+I1201 19:02:59.250218 137274321021824 utils.py:1231] [53400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.17918579401223
+I1201 19:02:59.250278 137274321021824 utils.py:1231] [53400] core_hours = 93.17918579401223
+I1201 19:02:59.250350 137274321021824 train.py:125] NOTE: Steps:53400/112603 [47.4%]
+Walltime:3d21h12m (0s eval)
+ETA:4d7h18m
+Total train time:8d4h29m
+I1201 19:08:11.060253 137274321021824 utils.py:1231] [53450] l2_params = 306.50092332278507
+I1201 19:08:11.060547 137274321021824 utils.py:1231] [53450] train/loss = 2.4512759149074554
+I1201 19:08:11.060713 137274321021824 utils.py:1231] [53450] l2_grads = 1.6378511190414429
+I1201 19:08:11.060799 137274321021824 utils.py:1231] [53450] lr = 0.0006190625869381293
+I1201 19:08:11.060864 137274321021824 utils.py:1231] [53450] uptime = 335880.42322552396
+I1201 19:08:11.060960 137274321021824 utils.py:1231] [53450] examples_seen = 54732800.0
+I1201 19:08:11.061020 137274321021824 utils.py:1231] [53450] progress = 0.47467651838760955
+I1201 19:08:11.061080 137274321021824 utils.py:1231] [53450] epoch = 42.72105041731484
+I1201 19:08:11.061136 137274321021824 utils.py:1231] [53450] img/sec/core = 164.20203338094623
+I1201 19:08:11.061197 137274321021824 utils.py:1231] [53450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.26579996028887
+I1201 19:08:11.061253 137274321021824 utils.py:1231] [53450] core_hours = 93.26579996028887
+I1201 19:08:11.061325 137274321021824 train.py:125] NOTE: Steps:53450/112603 [47.5%]
+Walltime:3d21h18m (0s eval)
+ETA:4d7h13m
+Total train time:8d4h29m
+I1201 19:13:22.861616 137274321021824 utils.py:1231] [53500] l2_params = 306.4231406518292
+I1201 19:13:22.861901 137274321021824 utils.py:1231] [53500] train/loss = 2.401927947998047
+I1201 19:13:22.862081 137274321021824 utils.py:1231] [53500] l2_grads = 1.5223932266235352
+I1201 19:13:22.862170 137274321021824 utils.py:1231] [53500] lr = 0.0006183189940271785
+I1201 19:13:22.862238 137274321021824 utils.py:1231] [53500] uptime = 336192.224599189
+I1201 19:13:22.862311 137274321021824 utils.py:1231] [53500] examples_seen = 54784000.0
+I1201 19:13:22.862380 137274321021824 utils.py:1231] [53500] progress = 0.47512055629068495
+I1201 19:13:22.862446 137274321021824 utils.py:1231] [53500] epoch = 42.7610139817838
+I1201 19:13:22.862513 137274321021824 utils.py:1231] [53500] img/sec/core = 164.2071020989207
+I1201 19:13:22.862581 137274321021824 utils.py:1231] [53500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.35241145297361
+I1201 19:13:22.862637 137274321021824 utils.py:1231] [53500] core_hours = 93.35241145297361
+I1201 19:13:22.862710 137274321021824 train.py:125] NOTE: Steps:53500/112603 [47.5%]
+Walltime:3d21h23m (0s eval)
+ETA:4d7h7m
+Total train time:8d4h29m
+I1201 19:18:34.648143 137274321021824 utils.py:1231] [53550] l2_params = 306.3362403838556
+I1201 19:18:34.648334 137274321021824 utils.py:1231] [53550] train/loss = 4.726800918579102
+I1201 19:18:34.648431 137274321021824 utils.py:1231] [53550] l2_grads = 1.3111612796783447
+I1201 19:18:34.648492 137274321021824 utils.py:1231] [53550] lr = 0.0006175751238008073
+I1201 19:18:34.648544 137274321021824 utils.py:1231] [53550] uptime = 336504.01090561796
+I1201 19:18:34.648596 137274321021824 utils.py:1231] [53550] examples_seen = 54835200.0
+I1201 19:18:34.648644 137274321021824 utils.py:1231] [53550] progress = 0.4755645941937604
+I1201 19:18:34.648692 137274321021824 utils.py:1231] [53550] epoch = 42.80097754625275
+I1201 19:18:34.648744 137274321021824 utils.py:1231] [53550] img/sec/core = 164.21503749288783
+I1201 19:18:34.648799 137274321021824 utils.py:1231] [53550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.439018760315
+I1201 19:18:34.648850 137274321021824 utils.py:1231] [53550] core_hours = 93.439018760315
+I1201 19:18:34.648917 137274321021824 train.py:125] NOTE: Steps:53550/112603 [47.6%]
+Walltime:3d21h28m (0s eval)
+ETA:4d7h2m
+Total train time:8d4h29m
+I1201 19:23:46.437173 137274321021824 utils.py:1231] [53600] l2_params = 306.26413361866594
+I1201 19:23:46.437387 137274321021824 utils.py:1231] [53600] train/loss = 4.132614254951477
+I1201 19:23:46.437486 137274321021824 utils.py:1231] [53600] l2_grads = 1.3320612907409668
+I1201 19:23:46.437547 137274321021824 utils.py:1231] [53600] lr = 0.000616830978002495
+I1201 19:23:46.437602 137274321021824 utils.py:1231] [53600] uptime = 336815.799963706
+I1201 19:23:46.437655 137274321021824 utils.py:1231] [53600] examples_seen = 54886400.0
+I1201 19:23:46.437705 137274321021824 utils.py:1231] [53600] progress = 0.4760086320968358
+I1201 19:23:46.437756 137274321021824 utils.py:1231] [53600] epoch = 42.84094111072171
+I1201 19:23:46.437807 137274321021824 utils.py:1231] [53600] img/sec/core = 164.2135882316423
+I1201 19:23:46.437865 137274321021824 utils.py:1231] [53600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.52562683200611
+I1201 19:23:46.437922 137274321021824 utils.py:1231] [53600] core_hours = 93.52562683200611
+I1201 19:23:46.437984 137274321021824 train.py:125] NOTE: Steps:53600/112603 [47.6%]
+Walltime:3d21h33m (0s eval)
+ETA:4d6h57m
+Total train time:8d4h29m
+I1201 19:28:58.160151 137274321021824 utils.py:1231] [53650] l2_params = 306.18193959920023
+I1201 19:28:58.160368 137274321021824 utils.py:1231] [53650] train/loss = 2.434811234474182
+I1201 19:28:58.160459 137274321021824 utils.py:1231] [53650] l2_grads = 1.4659698009490967
+I1201 19:28:58.160518 137274321021824 utils.py:1231] [53650] lr = 0.0006160865583763666
+I1201 19:28:58.160569 137274321021824 utils.py:1231] [53650] uptime = 337127.522931106
+I1201 19:28:58.160621 137274321021824 utils.py:1231] [53650] examples_seen = 54937600.0
+I1201 19:28:58.160674 137274321021824 utils.py:1231] [53650] progress = 0.4764526699999112
+I1201 19:28:58.160721 137274321021824 utils.py:1231] [53650] epoch = 42.88090467519066
+I1201 19:28:58.160771 137274321021824 utils.py:1231] [53650] img/sec/core = 164.24840436702078
+I1201 19:28:58.160824 137274321021824 utils.py:1231] [53650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.61221654517277
+I1201 19:28:58.160873 137274321021824 utils.py:1231] [53650] core_hours = 93.61221654517277
+I1201 19:28:58.160942 137274321021824 train.py:125] NOTE: Steps:53650/112603 [47.6%]
+Walltime:3d21h38m (0s eval)
+ETA:4d6h52m
+Total train time:8d4h29m
+I1201 19:34:09.955897 137274321021824 utils.py:1231] [53700] l2_params = 306.1121333723635
+I1201 19:34:09.956229 137274321021824 utils.py:1231] [53700] train/loss = 4.716337978839874
+I1201 19:34:09.956459 137274321021824 utils.py:1231] [53700] l2_grads = 1.3942536115646362
+I1201 19:34:09.956572 137274321021824 utils.py:1231] [53700] lr = 0.0006153418666671892
+I1201 19:34:09.956671 137274321021824 utils.py:1231] [53700] uptime = 337439.31901821
+I1201 19:34:09.956770 137274321021824 utils.py:1231] [53700] examples_seen = 54988800.0
+I1201 19:34:09.956870 137274321021824 utils.py:1231] [53700] progress = 0.4768967079029866
+I1201 19:34:09.956966 137274321021824 utils.py:1231] [53700] epoch = 42.920868239659626
+I1201 19:34:09.957048 137274321021824 utils.py:1231] [53700] img/sec/core = 164.20988626108112
+I1201 19:34:09.957150 137274321021824 utils.py:1231] [53700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.69882656936832
+I1201 19:34:09.957246 137274321021824 utils.py:1231] [53700] core_hours = 93.69882656936832
+I1201 19:34:09.957363 137274321021824 train.py:125] NOTE: Steps:53700/112603 [47.7%]
+Walltime:3d21h43m (0s eval)
+ETA:4d6h46m
+Total train time:8d4h29m
+I1201 19:39:21.737833 137274321021824 utils.py:1231] [53750] l2_params = 306.0122760791052
+I1201 19:39:21.738063 137274321021824 utils.py:1231] [53750] train/loss = 3.0591704547405243
+I1201 19:39:21.738200 137274321021824 utils.py:1231] [53750] l2_grads = 1.3769510984420776
+I1201 19:39:21.738279 137274321021824 utils.py:1231] [53750] lr = 0.0006145969046203664
+I1201 19:39:21.738339 137274321021824 utils.py:1231] [53750] uptime = 337751.10070112
+I1201 19:39:21.738406 137274321021824 utils.py:1231] [53750] examples_seen = 55040000.0
+I1201 19:39:21.738471 137274321021824 utils.py:1231] [53750] progress = 0.477340745806062
+I1201 19:39:21.738536 137274321021824 utils.py:1231] [53750] epoch = 42.96083180412858
+I1201 19:39:21.738593 137274321021824 utils.py:1231] [53750] img/sec/core = 164.2174726947676
+I1201 19:39:21.738674 137274321021824 utils.py:1231] [53750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.78543259239888
+I1201 19:39:21.738729 137274321021824 utils.py:1231] [53750] core_hours = 93.78543259239888
+I1201 19:39:21.738793 137274321021824 train.py:125] NOTE: Steps:53750/112603 [47.7%]
+Walltime:3d21h49m (0s eval)
+ETA:4d6h41m
+Total train time:8d4h28m
+I1201 19:44:33.544853 137274321021824 utils.py:1231] [53800] l2_params = 305.94861548047845
+I1201 19:44:33.545066 137274321021824 utils.py:1231] [53800] train/loss = 2.2763459384441376
+I1201 19:44:33.545163 137274321021824 utils.py:1231] [53800] l2_grads = 1.6593190431594849
+I1201 19:44:33.545226 137274321021824 utils.py:1231] [53800] lr = 0.0006138516739819367
+I1201 19:44:33.545282 137274321021824 utils.py:1231] [53800] uptime = 338062.907643974
+I1201 19:44:33.545336 137274321021824 utils.py:1231] [53800] examples_seen = 55091200.0
+I1201 19:44:33.545385 137274321021824 utils.py:1231] [53800] progress = 0.4777847837091374
+I1201 19:44:33.545435 137274321021824 utils.py:1231] [53800] epoch = 43.00079536859754
+I1201 19:44:33.545489 137274321021824 utils.py:1231] [53800] img/sec/core = 164.2041691931497
+I1201 19:44:33.545552 137274321021824 utils.py:1231] [53800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.87204563208054
+I1201 19:44:33.545603 137274321021824 utils.py:1231] [53800] core_hours = 93.87204563208054
+I1201 19:44:33.545664 137274321021824 train.py:125] NOTE: Steps:53800/112603 [47.8%]
+Walltime:3d21h54m (0s eval)
+ETA:4d6h36m
+Total train time:8d4h28m
+I1201 19:49:45.333748 137274321021824 utils.py:1231] [53850] l2_params = 305.85830255186426
+I1201 19:49:45.333970 137274321021824 utils.py:1231] [53850] train/loss = 3.8222494423389435
+I1201 19:49:45.334067 137274321021824 utils.py:1231] [53850] l2_grads = 1.4621572494506836
+I1201 19:49:45.334126 137274321021824 utils.py:1231] [53850] lr = 0.0006131061764985669
+I1201 19:49:45.334181 137274321021824 utils.py:1231] [53850] uptime = 338374.696542216
+I1201 19:49:45.334237 137274321021824 utils.py:1231] [53850] examples_seen = 55142400.0
+I1201 19:49:45.334288 137274321021824 utils.py:1231] [53850] progress = 0.4782288216122128
+I1201 19:49:45.334336 137274321021824 utils.py:1231] [53850] epoch = 43.04075893306649
+I1201 19:49:45.334388 137274321021824 utils.py:1231] [53850] img/sec/core = 164.2136724196663
+I1201 19:49:45.334444 137274321021824 utils.py:1231] [53850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 93.95865365936999
+I1201 19:49:45.334507 137274321021824 utils.py:1231] [53850] core_hours = 93.95865365936999
+I1201 19:49:45.334568 137274321021824 train.py:125] NOTE: Steps:53850/112603 [47.8%]
+Walltime:3d21h59m (0s eval)
+ETA:4d6h31m
+Total train time:8d4h28m
+I1201 19:54:57.131721 137274321021824 utils.py:1231] [53900] l2_params = 305.77390320762237
+I1201 19:54:57.131999 137274321021824 utils.py:1231] [53900] train/loss = 4.025709837675095
+I1201 19:54:57.132130 137274321021824 utils.py:1231] [53900] l2_grads = 1.2878776788711548
+I1201 19:54:57.132201 137274321021824 utils.py:1231] [53900] lr = 0.000612360413917551
+I1201 19:54:57.132264 137274321021824 utils.py:1231] [53900] uptime = 338686.494625451
+I1201 19:54:57.132321 137274321021824 utils.py:1231] [53900] examples_seen = 55193600.0
+I1201 19:54:57.132374 137274321021824 utils.py:1231] [53900] progress = 0.4786728595152882
+I1201 19:54:57.132431 137274321021824 utils.py:1231] [53900] epoch = 43.08072249753545
+I1201 19:54:57.132496 137274321021824 utils.py:1231] [53900] img/sec/core = 164.20883498957642
+I1201 19:54:57.132565 137274321021824 utils.py:1231] [53900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.04526423804639
+I1201 19:54:57.132634 137274321021824 utils.py:1231] [53900] core_hours = 94.04526423804639
+I1201 19:54:57.132699 137274321021824 train.py:125] NOTE: Steps:53900/112603 [47.9%]
+Walltime:3d22h4m (0s eval)
+ETA:4d6h25m
+Total train time:8d4h28m
+I1201 20:00:08.920654 137274321021824 utils.py:1231] [53950] l2_params = 305.7095389444196
+I1201 20:00:08.920866 137274321021824 utils.py:1231] [53950] train/loss = 2.3651121854782104
+I1201 20:00:08.920982 137274321021824 utils.py:1231] [53950] l2_grads = 1.672836422920227
+I1201 20:00:08.921051 137274321021824 utils.py:1231] [53950] lr = 0.0006116143879868021
+I1201 20:00:08.921109 137274321021824 utils.py:1231] [53950] uptime = 338998.283470886
+I1201 20:00:08.921190 137274321021824 utils.py:1231] [53950] examples_seen = 55244800.0
+I1201 20:00:08.921251 137274321021824 utils.py:1231] [53950] progress = 0.4791168974183636
+I1201 20:00:08.921308 137274321021824 utils.py:1231] [53950] epoch = 43.12068606200441
+I1201 20:00:08.921365 137274321021824 utils.py:1231] [53950] img/sec/core = 164.21370023216878
+I1201 20:00:08.921426 137274321021824 utils.py:1231] [53950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.13187225066721
+I1201 20:00:08.921482 137274321021824 utils.py:1231] [53950] core_hours = 94.13187225066721
+I1201 20:00:08.921550 137274321021824 train.py:125] NOTE: Steps:53950/112603 [47.9%]
+Walltime:3d22h9m (0s eval)
+ETA:4d6h20m
+Total train time:8d4h28m
+I1201 20:05:20.699065 137274321021824 utils.py:1231] [54000] l2_params = 305.6126147535896
+I1201 20:05:20.699312 137274321021824 utils.py:1231] [54000] train/loss = 2.260664850473404
+I1201 20:05:20.699429 137274321021824 utils.py:1231] [54000] l2_grads = 1.6435129642486572
+I1201 20:05:20.699512 137274321021824 utils.py:1231] [54000] lr = 0.0006108681004548532
+I1201 20:05:20.699580 137274321021824 utils.py:1231] [54000] uptime = 339310.061941874
+I1201 20:05:20.699655 137274321021824 utils.py:1231] [54000] examples_seen = 55296000.0
+I1201 20:05:20.699713 137274321021824 utils.py:1231] [54000] progress = 0.47956093532143906
+I1201 20:05:20.699767 137274321021824 utils.py:1231] [54000] epoch = 43.160649626473365
+I1201 20:05:20.699822 137274321021824 utils.py:1231] [54000] img/sec/core = 164.21916445274456
+I1201 20:05:20.699887 137274321021824 utils.py:1231] [54000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.21847738149721
+I1201 20:05:20.699944 137274321021824 utils.py:1231] [54000] core_hours = 94.21847738149721
+I1201 20:05:20.700009 137274321021824 train.py:125] NOTE: Steps:54000/112603 [48.0%]
+Walltime:3d22h15m (0s eval)
+ETA:4d6h15m
+Total train time:8d4h28m
+I1201 20:10:32.845180 137274321021824 utils.py:1231] [54050] l2_params = 305.5328603392664
+I1201 20:10:32.845456 137274321021824 utils.py:1231] [54050] train/loss = 2.417278215289116
+I1201 20:10:32.845585 137274321021824 utils.py:1231] [54050] l2_grads = 1.6669762134552002
+I1201 20:10:32.845677 137274321021824 utils.py:1231] [54050] lr = 0.0006101215530708487
+I1201 20:10:32.845748 137274321021824 utils.py:1231] [54050] uptime = 339622.208108894
+I1201 20:10:32.845811 137274321021824 utils.py:1231] [54050] examples_seen = 55347200.0
+I1201 20:10:32.845871 137274321021824 utils.py:1231] [54050] progress = 0.48000497322451446
+I1201 20:10:32.845939 137274321021824 utils.py:1231] [54050] epoch = 43.20061319094232
+I1201 20:10:32.845998 137274321021824 utils.py:1231] [54050] img/sec/core = 164.02572067052103
+I1201 20:10:32.846062 137274321021824 utils.py:1231] [54050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.30518465011389
+I1201 20:10:32.846119 137274321021824 utils.py:1231] [54050] core_hours = 94.30518465011389
+I1201 20:10:32.846188 137274321021824 train.py:125] NOTE: Steps:54050/112603 [48.0%]
+Walltime:3d22h20m (0s eval)
+ETA:4d6h9m
+Total train time:8d4h28m
+I1201 20:15:44.582625 137274321021824 utils.py:1231] [54100] l2_params = 305.4568186102502
+I1201 20:15:44.582878 137274321021824 utils.py:1231] [54100] train/loss = 3.531987875699997
+I1201 20:15:44.582990 137274321021824 utils.py:1231] [54100] l2_grads = 1.4266691207885742
+I1201 20:15:44.583067 137274321021824 utils.py:1231] [54100] lr = 0.0006093747475845419
+I1201 20:15:44.583127 137274321021824 utils.py:1231] [54100] uptime = 339933.94548841997
+I1201 20:15:44.583181 137274321021824 utils.py:1231] [54100] examples_seen = 55398400.0
+I1201 20:15:44.583231 137274321021824 utils.py:1231] [54100] progress = 0.48044901112758986
+I1201 20:15:44.583282 137274321021824 utils.py:1231] [54100] epoch = 43.240576755411276
+I1201 20:15:44.583337 137274321021824 utils.py:1231] [54100] img/sec/core = 164.24081089619816
+I1201 20:15:44.583394 137274321021824 utils.py:1231] [54100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.39177836664888
+I1201 20:15:44.583446 137274321021824 utils.py:1231] [54100] core_hours = 94.39177836664888
+I1201 20:15:44.583508 137274321021824 train.py:125] NOTE: Steps:54100/112603 [48.0%]
+Walltime:3d22h25m (0s eval)
+ETA:4d6h4m
+Total train time:8d4h28m
+I1201 20:20:56.364311 137274321021824 utils.py:1231] [54150] l2_params = 305.40044721888853
+I1201 20:20:56.364538 137274321021824 utils.py:1231] [54150] train/loss = 2.7966738045215607
+I1201 20:20:56.364646 137274321021824 utils.py:1231] [54150] l2_grads = 1.5000711679458618
+I1201 20:20:56.364706 137274321021824 utils.py:1231] [54150] lr = 0.0006086276857462914
+I1201 20:20:56.364758 137274321021824 utils.py:1231] [54150] uptime = 340245.727120174
+I1201 20:20:56.364812 137274321021824 utils.py:1231] [54150] examples_seen = 55449600.0
+I1201 20:20:56.364864 137274321021824 utils.py:1231] [54150] progress = 0.48089304903066526
+I1201 20:20:56.364918 137274321021824 utils.py:1231] [54150] epoch = 43.28054031988023
+I1201 20:20:56.364969 137274321021824 utils.py:1231] [54150] img/sec/core = 164.21749963896335
+I1201 20:20:56.365026 137274321021824 utils.py:1231] [54150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.47838437546943
+I1201 20:20:56.365077 137274321021824 utils.py:1231] [54150] core_hours = 94.47838437546943
+I1201 20:20:56.365142 137274321021824 train.py:125] NOTE: Steps:54150/112603 [48.1%]
+Walltime:3d22h30m (0s eval)
+ETA:4d5h59m
+Total train time:8d4h28m
+I1201 20:26:08.149274 137274321021824 utils.py:1231] [54200] l2_params = 305.3208310882333
+I1201 20:26:08.149509 137274321021824 utils.py:1231] [54200] train/loss = 2.482012093067169
+I1201 20:26:08.149634 137274321021824 utils.py:1231] [54200] l2_grads = 1.468253493309021
+I1201 20:26:08.149710 137274321021824 utils.py:1231] [54200] lr = 0.0006078803693070563
+I1201 20:26:08.149782 137274321021824 utils.py:1231] [54200] uptime = 340557.51214372896
+I1201 20:26:08.149851 137274321021824 utils.py:1231] [54200] examples_seen = 55500800.0
+I1201 20:26:08.149932 137274321021824 utils.py:1231] [54200] progress = 0.48133708693374067
+I1201 20:26:08.150001 137274321021824 utils.py:1231] [54200] epoch = 43.32050388434919
+I1201 20:26:08.150076 137274321021824 utils.py:1231] [54200] img/sec/core = 164.21571317382603
+I1201 20:26:08.150137 137274321021824 utils.py:1231] [54200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.56499132645693
+I1201 20:26:08.150189 137274321021824 utils.py:1231] [54200] core_hours = 94.56499132645693
+I1201 20:26:08.150253 137274321021824 train.py:125] NOTE: Steps:54200/112603 [48.1%]
+Walltime:3d22h35m (0s eval)
+ETA:4d5h54m
+Total train time:8d4h28m
+I1201 20:31:19.922712 137274321021824 utils.py:1231] [54250] l2_params = 305.23506016426137
+I1201 20:31:19.922930 137274321021824 utils.py:1231] [54250] train/loss = 2.256314367055893
+I1201 20:31:19.923039 137274321021824 utils.py:1231] [54250] l2_grads = 1.7330271005630493
+I1201 20:31:19.923114 137274321021824 utils.py:1231] [54250] lr = 0.0006071328000183936
+I1201 20:31:19.923183 137274321021824 utils.py:1231] [54250] uptime = 340869.28554375697
+I1201 20:31:19.923246 137274321021824 utils.py:1231] [54250] examples_seen = 55552000.0
+I1201 20:31:19.923305 137274321021824 utils.py:1231] [54250] progress = 0.48178112483681607
+I1201 20:31:19.923363 137274321021824 utils.py:1231] [54250] epoch = 43.36046744881815
+I1201 20:31:19.923425 137274321021824 utils.py:1231] [54250] img/sec/core = 164.22183545934718
+I1201 20:31:19.923490 137274321021824 utils.py:1231] [54250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.65159504868693
+I1201 20:31:19.923550 137274321021824 utils.py:1231] [54250] core_hours = 94.65159504868693
+I1201 20:31:19.923619 137274321021824 train.py:125] NOTE: Steps:54250/112603 [48.2%]
+Walltime:3d22h41m (0s eval)
+ETA:4d5h48m
+Total train time:8d4h28m
+I1201 20:36:31.697185 137274321021824 utils.py:1231] [54300] l2_params = 305.1623636607978
+I1201 20:36:31.697440 137274321021824 utils.py:1231] [54300] train/loss = 4.719946205615997
+I1201 20:36:31.697546 137274321021824 utils.py:1231] [54300] l2_grads = 1.3559119701385498
+I1201 20:36:31.697618 137274321021824 utils.py:1231] [54300] lr = 0.000606384979632452
+I1201 20:36:31.697687 137274321021824 utils.py:1231] [54300] uptime = 341181.060047083
+I1201 20:36:31.697758 137274321021824 utils.py:1231] [54300] examples_seen = 55603200.0
+I1201 20:36:31.697839 137274321021824 utils.py:1231] [54300] progress = 0.48222516273989147
+I1201 20:36:31.697927 137274321021824 utils.py:1231] [54300] epoch = 43.400431013287104
+I1201 20:36:31.697995 137274321021824 utils.py:1231] [54300] img/sec/core = 164.22125431615788
+I1201 20:36:31.698073 137274321021824 utils.py:1231] [54300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.73819907738861
+I1201 20:36:31.698159 137274321021824 utils.py:1231] [54300] core_hours = 94.73819907738861
+I1201 20:36:31.698236 137274321021824 train.py:125] NOTE: Steps:54300/112603 [48.2%]
+Walltime:3d22h46m (0s eval)
+ETA:4d5h43m
+Total train time:8d4h28m
+I1201 20:41:43.478512 137274321021824 utils.py:1231] [54350] l2_params = 305.06847016768984
+I1201 20:41:43.478723 137274321021824 utils.py:1231] [54350] train/loss = 2.568399339914322
+I1201 20:41:43.478829 137274321021824 utils.py:1231] [54350] l2_grads = 1.6813240051269531
+I1201 20:41:43.478898 137274321021824 utils.py:1231] [54350] lr = 0.000605636909901968
+I1201 20:41:43.478952 137274321021824 utils.py:1231] [54350] uptime = 341492.841313553
+I1201 20:41:43.479005 137274321021824 utils.py:1231] [54350] examples_seen = 55654400.0
+I1201 20:41:43.479056 137274321021824 utils.py:1231] [54350] progress = 0.48266920064296687
+I1201 20:41:43.479105 137274321021824 utils.py:1231] [54350] epoch = 43.44039457775606
+I1201 20:41:43.479158 137274321021824 utils.py:1231] [54350] img/sec/core = 164.21769203676897
+I1201 20:41:43.479217 137274321021824 utils.py:1231] [54350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.82480498474138
+I1201 20:41:43.479282 137274321021824 utils.py:1231] [54350] core_hours = 94.82480498474138
+I1201 20:41:43.479349 137274321021824 train.py:125] NOTE: Steps:54350/112603 [48.3%]
+Walltime:3d22h51m (0s eval)
+ETA:4d5h38m
+Total train time:8d4h27m
+I1201 20:46:55.249323 137274321021824 utils.py:1231] [54400] l2_params = 304.9765784729267
+I1201 20:46:55.249519 137274321021824 utils.py:1231] [54400] train/loss = 2.8406608402729034
+I1201 20:46:55.249613 137274321021824 utils.py:1231] [54400] l2_grads = 1.5940665006637573
+I1201 20:46:55.249686 137274321021824 utils.py:1231] [54400] lr = 0.0006048885925802653
+I1201 20:46:55.249737 137274321021824 utils.py:1231] [54400] uptime = 341804.61209952796
+I1201 20:46:55.249786 137274321021824 utils.py:1231] [54400] examples_seen = 55705600.0
+I1201 20:46:55.249833 137274321021824 utils.py:1231] [54400] progress = 0.48311323854604227
+I1201 20:46:55.249879 137274321021824 utils.py:1231] [54400] epoch = 43.48035814222502
+I1201 20:46:55.249935 137274321021824 utils.py:1231] [54400] img/sec/core = 164.22321238305196
+I1201 20:46:55.249989 137274321021824 utils.py:1231] [54400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.91140798084554
+I1201 20:46:55.250037 137274321021824 utils.py:1231] [54400] core_hours = 94.91140798084554
+I1201 20:46:55.250093 137274321021824 train.py:125] NOTE: Steps:54400/112603 [48.3%]
+Walltime:3d22h56m (0s eval)
+ETA:4d5h33m
+Total train time:8d4h27m
+I1201 20:52:07.023245 137274321021824 utils.py:1231] [54450] l2_params = 304.8919107467401
+I1201 20:52:07.023486 137274321021824 utils.py:1231] [54450] train/loss = 2.747270107269287
+I1201 20:52:07.023618 137274321021824 utils.py:1231] [54450] l2_grads = 1.4372570514678955
+I1201 20:52:07.023709 137274321021824 utils.py:1231] [54450] lr = 0.0006041400294212444
+I1201 20:52:07.023786 137274321021824 utils.py:1231] [54450] uptime = 342116.386143266
+I1201 20:52:07.023862 137274321021824 utils.py:1231] [54450] examples_seen = 55756800.0
+I1201 20:52:07.023946 137274321021824 utils.py:1231] [54450] progress = 0.4835572764491177
+I1201 20:52:07.024007 137274321021824 utils.py:1231] [54450] epoch = 43.52032170669398
+I1201 20:52:07.024074 137274321021824 utils.py:1231] [54450] img/sec/core = 164.22149639570887
+I1201 20:52:07.024143 137274321021824 utils.py:1231] [54450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 94.99801188188388
+I1201 20:52:07.024216 137274321021824 utils.py:1231] [54450] core_hours = 94.99801188188388
+I1201 20:52:07.024287 137274321021824 train.py:125] NOTE: Steps:54450/112603 [48.4%]
+Walltime:3d23h1m (0s eval)
+ETA:4d5h27m
+Total train time:8d4h27m
+I1201 20:57:18.804860 137274321021824 utils.py:1231] [54500] l2_params = 304.79817025504934
+I1201 20:57:18.805064 137274321021824 utils.py:1231] [54500] train/loss = 2.2291915267705917
+I1201 20:57:18.805164 137274321021824 utils.py:1231] [54500] l2_grads = 1.5406979322433472
+I1201 20:57:18.805225 137274321021824 utils.py:1231] [54500] lr = 0.0006033912221793843
+I1201 20:57:18.805276 137274321021824 utils.py:1231] [54500] uptime = 342428.167638043
+I1201 20:57:18.805329 137274321021824 utils.py:1231] [54500] examples_seen = 55808000.0
+I1201 20:57:18.805381 137274321021824 utils.py:1231] [54500] progress = 0.4840013143521931
+I1201 20:57:18.805429 137274321021824 utils.py:1231] [54500] epoch = 43.56028527116293
+I1201 20:57:18.805480 137274321021824 utils.py:1231] [54500] img/sec/core = 164.2175717857116
+I1201 20:57:18.805535 137274321021824 utils.py:1231] [54500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.08461785265528
+I1201 20:57:18.805587 137274321021824 utils.py:1231] [54500] core_hours = 95.08461785265528
+I1201 20:57:18.805656 137274321021824 train.py:125] NOTE: Steps:54500/112603 [48.4%]
+Walltime:3d23h7m (0s eval)
+ETA:4d5h22m
+Total train time:8d4h27m
+I1201 21:02:30.581851 137274321021824 utils.py:1231] [54550] l2_params = 304.7211369900305
+I1201 21:02:30.582067 137274321021824 utils.py:1231] [54550] train/loss = 2.518129140138626
+I1201 21:02:30.582180 137274321021824 utils.py:1231] [54550] l2_grads = 1.4609625339508057
+I1201 21:02:30.582256 137274321021824 utils.py:1231] [54550] lr = 0.000602642172609735
+I1201 21:02:30.582329 137274321021824 utils.py:1231] [54550] uptime = 342739.944689166
+I1201 21:02:30.582395 137274321021824 utils.py:1231] [54550] examples_seen = 55859200.0
+I1201 21:02:30.582460 137274321021824 utils.py:1231] [54550] progress = 0.4844453522552685
+I1201 21:02:30.582527 137274321021824 utils.py:1231] [54550] epoch = 43.60024883563189
+I1201 21:02:30.582607 137274321021824 utils.py:1231] [54550] img/sec/core = 164.21991232383863
+I1201 21:02:30.582678 137274321021824 utils.py:1231] [54550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.17122258907833
+I1201 21:02:30.582740 137274321021824 utils.py:1231] [54550] core_hours = 95.17122258907833
+I1201 21:02:30.582811 137274321021824 train.py:125] NOTE: Steps:54550/112603 [48.4%]
+Walltime:3d23h12m (0s eval)
+ETA:4d5h17m
+Total train time:8d4h27m
+I1201 21:07:42.492294 137274321021824 utils.py:1231] [54600] l2_params = 304.6455308816857
+I1201 21:07:42.492520 137274321021824 utils.py:1231] [54600] train/loss = 3.9465316832065582
+I1201 21:07:42.492619 137274321021824 utils.py:1231] [54600] l2_grads = 1.2788000106811523
+I1201 21:07:42.492686 137274321021824 utils.py:1231] [54600] lr = 0.0006018928824679155
+I1201 21:07:42.492757 137274321021824 utils.py:1231] [54600] uptime = 343051.855104031
+I1201 21:07:42.492813 137274321021824 utils.py:1231] [54600] examples_seen = 55910400.0
+I1201 21:07:42.492868 137274321021824 utils.py:1231] [54600] progress = 0.4848893901583439
+I1201 21:07:42.492933 137274321021824 utils.py:1231] [54600] epoch = 43.64021240010084
+I1201 21:07:42.492990 137274321021824 utils.py:1231] [54600] img/sec/core = 164.1496967074952
+I1201 21:07:42.493048 137274321021824 utils.py:1231] [54600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.25786437098527
+I1201 21:07:42.493101 137274321021824 utils.py:1231] [54600] core_hours = 95.25786437098527
+I1201 21:07:42.493165 137274321021824 train.py:125] NOTE: Steps:54600/112603 [48.5%]
+Walltime:3d23h17m (0s eval)
+ETA:4d5h11m
+Total train time:8d4h27m
+I1201 21:12:54.267411 137274321021824 utils.py:1231] [54650] l2_params = 304.56414627999806
+I1201 21:12:54.267611 137274321021824 utils.py:1231] [54650] train/loss = 3.477863311767578
+I1201 21:12:54.267706 137274321021824 utils.py:1231] [54650] l2_grads = 1.316322922706604
+I1201 21:12:54.267765 137274321021824 utils.py:1231] [54650] lr = 0.0006011433535101074
+I1201 21:12:54.267815 137274321021824 utils.py:1231] [54650] uptime = 343363.630177814
+I1201 21:12:54.267868 137274321021824 utils.py:1231] [54650] examples_seen = 55961600.0
+I1201 21:12:54.267920 137274321021824 utils.py:1231] [54650] progress = 0.4853334280614193
+I1201 21:12:54.267967 137274321021824 utils.py:1231] [54650] epoch = 43.680175964569806
+I1201 21:12:54.268015 137274321021824 utils.py:1231] [54650] img/sec/core = 164.22095383941718
+I1201 21:12:54.268070 137274321021824 utils.py:1231] [54650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.34446855814721
+I1201 21:12:54.268117 137274321021824 utils.py:1231] [54650] core_hours = 95.34446855814721
+I1201 21:12:54.268176 137274321021824 train.py:125] NOTE: Steps:54650/112603 [48.5%]
+Walltime:3d23h22m (0s eval)
+ETA:4d5h6m
+Total train time:8d4h27m
+I1201 21:18:06.041650 137274321021824 utils.py:1231] [54700] l2_params = 304.4898481213331
+I1201 21:18:06.041932 137274321021824 utils.py:1231] [54700] train/loss = 2.3922298550605774
+I1201 21:18:06.042121 137274321021824 utils.py:1231] [54700] l2_grads = 1.6270874738693237
+I1201 21:18:06.042211 137274321021824 utils.py:1231] [54700] lr = 0.0006003935874930524
+I1201 21:18:06.042273 137274321021824 utils.py:1231] [54700] uptime = 343675.404634615
+I1201 21:18:06.042333 137274321021824 utils.py:1231] [54700] examples_seen = 56012800.0
+I1201 21:18:06.042383 137274321021824 utils.py:1231] [54700] progress = 0.4857774659644947
+I1201 21:18:06.042437 137274321021824 utils.py:1231] [54700] epoch = 43.72013952903876
+I1201 21:18:06.042523 137274321021824 utils.py:1231] [54700] img/sec/core = 164.2212788223331
+I1201 21:18:06.042579 137274321021824 utils.py:1231] [54700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.43107257392528
+I1201 21:18:06.042633 137274321021824 utils.py:1231] [54700] core_hours = 95.43107257392528
+I1201 21:18:06.042693 137274321021824 train.py:125] NOTE: Steps:54700/112603 [48.6%]
+Walltime:3d23h27m (0s eval)
+ETA:4d5h1m
+Total train time:8d4h27m
+I1201 21:23:17.722731 137274321021824 utils.py:1231] [54750] l2_params = 304.42563352216223
+I1201 21:23:17.722953 137274321021824 utils.py:1231] [54750] train/loss = 3.516896665096283
+I1201 21:23:17.723048 137274321021824 utils.py:1231] [54750] l2_grads = 1.406084418296814
+I1201 21:23:17.723107 137274321021824 utils.py:1231] [54750] lr = 0.0005996435861740487
+I1201 21:23:17.723165 137274321021824 utils.py:1231] [54750] uptime = 343987.08552702
+I1201 21:23:17.723217 137274321021824 utils.py:1231] [54750] examples_seen = 56064000.0
+I1201 21:23:17.723267 137274321021824 utils.py:1231] [54750] progress = 0.4862215038675701
+I1201 21:23:17.723315 137274321021824 utils.py:1231] [54750] epoch = 43.760103093507716
+I1201 21:23:17.723366 137274321021824 utils.py:1231] [54750] img/sec/core = 164.2705768869305
+I1201 21:23:17.723423 137274321021824 utils.py:1231] [54750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.51765059959332
+I1201 21:23:17.723474 137274321021824 utils.py:1231] [54750] core_hours = 95.51765059959332
+I1201 21:23:17.723540 137274321021824 train.py:125] NOTE: Steps:54750/112603 [48.6%]
+Walltime:3d23h33m (0s eval)
+ETA:4d4h56m
+Total train time:8d4h27m
+I1201 21:28:29.465455 137274321021824 utils.py:1231] [54800] l2_params = 304.3360607266926
+I1201 21:28:29.465698 137274321021824 utils.py:1231] [54800] train/loss = 4.14691287279129
+I1201 21:28:29.465826 137274321021824 utils.py:1231] [54800] l2_grads = 1.334874153137207
+I1201 21:28:29.465919 137274321021824 utils.py:1231] [54800] lr = 0.0005988933513109446
+I1201 21:28:29.465981 137274321021824 utils.py:1231] [54800] uptime = 344298.82834217796
+I1201 21:28:29.466041 137274321021824 utils.py:1231] [54800] examples_seen = 56115200.0
+I1201 21:28:29.466099 137274321021824 utils.py:1231] [54800] progress = 0.4866655417706455
+I1201 21:28:29.466155 137274321021824 utils.py:1231] [54800] epoch = 43.80006665797667
+I1201 21:28:29.466214 137274321021824 utils.py:1231] [54800] img/sec/core = 164.23794714900993
+I1201 21:28:29.466272 137274321021824 utils.py:1231] [54800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.6042458260261
+I1201 21:28:29.466328 137274321021824 utils.py:1231] [54800] core_hours = 95.6042458260261
+I1201 21:28:29.466394 137274321021824 train.py:125] NOTE: Steps:54800/112603 [48.7%]
+Walltime:3d23h38m (0s eval)
+ETA:4d4h50m
+Total train time:8d4h27m
+I1201 21:33:41.239495 137274321021824 utils.py:1231] [54850] l2_params = 304.25284199074946
+I1201 21:33:41.239727 137274321021824 utils.py:1231] [54850] train/loss = 3.4178451597690582
+I1201 21:33:41.239838 137274321021824 utils.py:1231] [54850] l2_grads = 1.4786405563354492
+I1201 21:33:41.239938 137274321021824 utils.py:1231] [54850] lr = 0.0005981428846621377
+I1201 21:33:41.240023 137274321021824 utils.py:1231] [54850] uptime = 344610.602381602
+I1201 21:33:41.240093 137274321021824 utils.py:1231] [54850] examples_seen = 56166400.0
+I1201 21:33:41.240157 137274321021824 utils.py:1231] [54850] progress = 0.4871095796737209
+I1201 21:33:41.240219 137274321021824 utils.py:1231] [54850] epoch = 43.84003022244563
+I1201 21:33:41.240282 137274321021824 utils.py:1231] [54850] img/sec/core = 164.22149866803392
+I1201 21:33:41.240354 137274321021824 utils.py:1231] [54850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.69084972586612
+I1201 21:33:41.240414 137274321021824 utils.py:1231] [54850] core_hours = 95.69084972586612
+I1201 21:33:41.240485 137274321021824 train.py:125] NOTE: Steps:54850/112603 [48.7%]
+Walltime:3d23h43m (0s eval)
+ETA:4d4h45m
+Total train time:8d4h27m
+I1201 21:38:53.012774 137274321021824 utils.py:1231] [54900] l2_params = 304.13802184630487
+I1201 21:38:53.013044 137274321021824 utils.py:1231] [54900] train/loss = 2.5034250617027283
+I1201 21:38:53.013230 137274321021824 utils.py:1231] [54900] l2_grads = 1.5497018098831177
+I1201 21:38:53.013321 137274321021824 utils.py:1231] [54900] lr = 0.0005973921879865662
+I1201 21:38:53.013409 137274321021824 utils.py:1231] [54900] uptime = 344922.375769738
+I1201 21:38:53.013491 137274321021824 utils.py:1231] [54900] examples_seen = 56217600.0
+I1201 21:38:53.013574 137274321021824 utils.py:1231] [54900] progress = 0.4875536175767963
+I1201 21:38:53.013662 137274321021824 utils.py:1231] [54900] epoch = 43.87999378691459
+I1201 21:38:53.013758 137274321021824 utils.py:1231] [54900] img/sec/core = 164.22184172327508
+I1201 21:38:53.013857 137274321021824 utils.py:1231] [54900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.77745344479278
+I1201 21:38:53.013920 137274321021824 utils.py:1231] [54900] core_hours = 95.77745344479278
+I1201 21:38:53.013989 137274321021824 train.py:125] NOTE: Steps:54900/112603 [48.8%]
+Walltime:3d23h48m (0s eval)
+ETA:4d4h40m
+Total train time:8d4h27m
+I1201 21:44:04.794695 137274321021824 utils.py:1231] [54950] l2_params = 304.0608903180687
+I1201 21:44:04.794924 137274321021824 utils.py:1231] [54950] train/loss = 2.244384750723839
+I1201 21:44:04.795093 137274321021824 utils.py:1231] [54950] l2_grads = 1.6363493204116821
+I1201 21:44:04.795203 137274321021824 utils.py:1231] [54950] lr = 0.0005966412630437104
+I1201 21:44:04.795280 137274321021824 utils.py:1231] [54950] uptime = 345234.15763821197
+I1201 21:44:04.795351 137274321021824 utils.py:1231] [54950] examples_seen = 56268800.0
+I1201 21:44:04.795426 137274321021824 utils.py:1231] [54950] progress = 0.4879976554798718
+I1201 21:44:04.795495 137274321021824 utils.py:1231] [54950] epoch = 43.919957351383545
+I1201 21:44:04.795574 137274321021824 utils.py:1231] [54950] img/sec/core = 164.21737495704824
+I1201 21:44:04.795641 137274321021824 utils.py:1231] [54950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.86405951936887
+I1201 21:44:04.795698 137274321021824 utils.py:1231] [54950] core_hours = 95.86405951936887
+I1201 21:44:04.795765 137274321021824 train.py:125] NOTE: Steps:54950/112603 [48.8%]
+Walltime:3d23h53m (0s eval)
+ETA:4d4h34m
+Total train time:8d4h27m
+I1201 21:49:16.638403 137274321021824 utils.py:1231] [55000] l2_params = 303.9932258083141
+I1201 21:49:16.638714 137274321021824 utils.py:1231] [55000] train/loss = 2.2191111594438553
+I1201 21:49:16.638920 137274321021824 utils.py:1231] [55000] l2_grads = 1.605878472328186
+I1201 21:49:16.639019 137274321021824 utils.py:1231] [55000] lr = 0.0005958901115935834
+I1201 21:49:16.639091 137274321021824 utils.py:1231] [55000] uptime = 345546.001450262
+I1201 21:49:16.639173 137274321021824 utils.py:1231] [55000] examples_seen = 56320000.0
+I1201 21:49:16.639233 137274321021824 utils.py:1231] [55000] progress = 0.4884416933829472
+I1201 21:49:16.639291 137274321021824 utils.py:1231] [55000] epoch = 43.9599209158525
+I1201 21:49:16.639345 137274321021824 utils.py:1231] [55000] img/sec/core = 164.18475538578153
+I1201 21:49:16.639414 137274321021824 utils.py:1231] [55000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 95.95068280049388
+I1201 21:49:16.639475 137274321021824 utils.py:1231] [55000] core_hours = 95.95068280049388
+I1201 21:49:16.639542 137274321021824 train.py:125] NOTE: Steps:55000/112603 [48.8%]
+Walltime:3d23h59m (0s eval)
+ETA:4d4h29m
+Total train time:8d4h26m
+I1201 21:49:16.999446 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:55000/112603 [48.8%]
+Walltime:3d23h59m (0s eval)
+ETA:4d4h29m
+Total train time:8d4h26m
+I1201 21:50:53.106897 137274321021824 utils.py:1231] [55000] val/acc@1 = 0.657844387755102
+I1201 21:50:53.107186 137274321021824 utils.py:1231] [55000] val/loss = 1.3818465670456692
+I1201 21:50:53.107376 137274321021824 utils.py:1231] [55000] z/secs/eval/val = 96.10765215498395
+I1201 21:50:53.107459 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 96.10765215498395
+I1201 21:56:03.359359 137274321021824 utils.py:1231] [55050] l2_params = 303.8961148390605
+I1201 21:56:03.359586 137274321021824 utils.py:1231] [55050] train/loss = 2.163019582629204
+I1201 21:56:03.359694 137274321021824 utils.py:1231] [55050] l2_grads = 1.542027235031128
+I1201 21:56:03.359797 137274321021824 utils.py:1231] [55050] lr = 0.0005951387353967295
+I1201 21:56:03.359861 137274321021824 utils.py:1231] [55050] uptime = 345952.722222453
+I1201 21:56:03.359929 137274321021824 utils.py:1231] [55050] examples_seen = 56371200.0
+I1201 21:56:03.359988 137274321021824 utils.py:1231] [55050] progress = 0.4888857312860226
+I1201 21:56:03.360046 137274321021824 utils.py:1231] [55050] epoch = 43.999884480321455
+I1201 21:56:03.360108 137274321021824 utils.py:1231] [55050] img/sec/core = 125.88489081633345
+I1201 21:56:03.360168 137274321021824 utils.py:1231] [55050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.06366079276917
+I1201 21:56:03.360226 137274321021824 utils.py:1231] [55050] core_hours = 96.06366079276917
+I1201 21:56:03.360288 137274321021824 train.py:125] NOTE: Steps:55050/112603 [48.9%]
+Walltime:4d0h5m (0s eval)
+ETA:4d4h26m
+Total train time:8d4h30m
+I1201 22:01:15.142112 137274321021824 utils.py:1231] [55100] l2_params = 303.801758729127
+I1201 22:01:15.142304 137274321021824 utils.py:1231] [55100] train/loss = 4.784853100776672
+I1201 22:01:15.142394 137274321021824 utils.py:1231] [55100] l2_grads = 1.721821904182434
+I1201 22:01:15.142458 137274321021824 utils.py:1231] [55100] lr = 0.0005943871362142208
+I1201 22:01:15.142508 137274321021824 utils.py:1231] [55100] uptime = 346264.504870222
+I1201 22:01:15.142559 137274321021824 utils.py:1231] [55100] examples_seen = 56422400.0
+I1201 22:01:15.142608 137274321021824 utils.py:1231] [55100] progress = 0.489329769189098
+I1201 22:01:15.142655 137274321021824 utils.py:1231] [55100] epoch = 44.03984804479041
+I1201 22:01:15.142705 137274321021824 utils.py:1231] [55100] img/sec/core = 164.21696449871067
+I1201 22:01:15.142759 137274321021824 utils.py:1231] [55100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.1502670838161
+I1201 22:01:15.142809 137274321021824 utils.py:1231] [55100] core_hours = 96.1502670838161
+I1201 22:01:15.142868 137274321021824 train.py:125] NOTE: Steps:55100/112603 [48.9%]
+Walltime:4d0h11m (0s eval)
+ETA:4d4h20m
+Total train time:8d4h30m
+I1201 22:06:26.923190 137274321021824 utils.py:1231] [55150] l2_params = 303.71398031387224
+I1201 22:06:26.923439 137274321021824 utils.py:1231] [55150] train/loss = 2.1470144391059875
+I1201 22:06:26.923536 137274321021824 utils.py:1231] [55150] l2_grads = 1.5199477672576904
+I1201 22:06:26.923596 137274321021824 utils.py:1231] [55150] lr = 0.000593635315807651
+I1201 22:06:26.923647 137274321021824 utils.py:1231] [55150] uptime = 346576.286009282
+I1201 22:06:26.923700 137274321021824 utils.py:1231] [55150] examples_seen = 56473600.0
+I1201 22:06:26.923749 137274321021824 utils.py:1231] [55150] progress = 0.4897738070921734
+I1201 22:06:26.923799 137274321021824 utils.py:1231] [55150] epoch = 44.07981160925937
+I1201 22:06:26.923848 137274321021824 utils.py:1231] [55150] img/sec/core = 164.21775914466053
+I1201 22:06:26.923910 137274321021824 utils.py:1231] [55150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.23687295577722
+I1201 22:06:26.923963 137274321021824 utils.py:1231] [55150] core_hours = 96.23687295577722
+I1201 22:06:26.924043 137274321021824 train.py:125] NOTE: Steps:55150/112603 [49.0%]
+Walltime:4d0h16m (0s eval)
+ETA:4d4h15m
+Total train time:8d4h29m
+I1201 22:11:38.683094 137274321021824 utils.py:1231] [55200] l2_params = 303.64291561685684
+I1201 22:11:38.683340 137274321021824 utils.py:1231] [55200] train/loss = 3.1302143335342407
+I1201 22:11:38.683514 137274321021824 utils.py:1231] [55200] l2_grads = 1.430015206336975
+I1201 22:11:38.683616 137274321021824 utils.py:1231] [55200] lr = 0.0005928832759391329
+I1201 22:11:38.683702 137274321021824 utils.py:1231] [55200] uptime = 346888.046059248
+I1201 22:11:38.683786 137274321021824 utils.py:1231] [55200] examples_seen = 56524800.0
+I1201 22:11:38.683870 137274321021824 utils.py:1231] [55200] progress = 0.4902178449952488
+I1201 22:11:38.683945 137274321021824 utils.py:1231] [55200] epoch = 44.11977517372833
+I1201 22:11:38.684014 137274321021824 utils.py:1231] [55200] img/sec/core = 164.22886769996842
+I1201 22:11:38.684097 137274321021824 utils.py:1231] [55200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.32347296965666
+I1201 22:11:38.684166 137274321021824 utils.py:1231] [55200] core_hours = 96.32347296965666
+I1201 22:11:38.684246 137274321021824 train.py:125] NOTE: Steps:55200/112603 [49.0%]
+Walltime:4d0h21m (0s eval)
+ETA:4d4h10m
+Total train time:8d4h29m
+I1201 22:16:50.462193 137274321021824 utils.py:1231] [55250] l2_params = 303.57367927686425
+I1201 22:16:50.462458 137274321021824 utils.py:1231] [55250] train/loss = 2.374663919210434
+I1201 22:16:50.462693 137274321021824 utils.py:1231] [55250] l2_grads = 1.6235363483428955
+I1201 22:16:50.462831 137274321021824 utils.py:1231] [55250] lr = 0.0005921310183712936
+I1201 22:16:50.462973 137274321021824 utils.py:1231] [55250] uptime = 347199.82532081497
+I1201 22:16:50.463085 137274321021824 utils.py:1231] [55250] examples_seen = 56576000.0
+I1201 22:16:50.463205 137274321021824 utils.py:1231] [55250] progress = 0.4906618828983242
+I1201 22:16:50.463296 137274321021824 utils.py:1231] [55250] epoch = 44.159738738197284
+I1201 22:16:50.463393 137274321021824 utils.py:1231] [55250] img/sec/core = 164.2187480420312
+I1201 22:16:50.463504 137274321021824 utils.py:1231] [55250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.41007832009194
+I1201 22:16:50.463582 137274321021824 utils.py:1231] [55250] core_hours = 96.41007832009194
+I1201 22:16:50.463683 137274321021824 train.py:125] NOTE: Steps:55250/112603 [49.1%]
+Walltime:4d0h26m (0s eval)
+ETA:4d4h5m
+Total train time:8d4h29m
+I1201 22:22:02.240924 137274321021824 utils.py:1231] [55300] l2_params = 303.48914075364945
+I1201 22:22:02.241156 137274321021824 utils.py:1231] [55300] train/loss = 2.5181650519371033
+I1201 22:22:02.241261 137274321021824 utils.py:1231] [55300] l2_grads = 1.4671295881271362
+I1201 22:22:02.241342 137274321021824 utils.py:1231] [55300] lr = 0.0005913785448672703
+I1201 22:22:02.241433 137274321021824 utils.py:1231] [55300] uptime = 347511.60378857
+I1201 22:22:02.241503 137274321021824 utils.py:1231] [55300] examples_seen = 56627200.0
+I1201 22:22:02.241565 137274321021824 utils.py:1231] [55300] progress = 0.4911059208013996
+I1201 22:22:02.241623 137274321021824 utils.py:1231] [55300] epoch = 44.19970230266624
+I1201 22:22:02.241683 137274321021824 utils.py:1231] [55300] img/sec/core = 164.21916615560627
+I1201 22:22:02.241748 137274321021824 utils.py:1231] [55300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.49668345002388
+I1201 22:22:02.241808 137274321021824 utils.py:1231] [55300] core_hours = 96.49668345002388
+I1201 22:22:02.241876 137274321021824 train.py:125] NOTE: Steps:55300/112603 [49.1%]
+Walltime:4d0h31m (0s eval)
+ETA:4d3h59m
+Total train time:8d4h29m
+I1201 22:27:14.023591 137274321021824 utils.py:1231] [55350] l2_params = 303.40414733818204
+I1201 22:27:14.023846 137274321021824 utils.py:1231] [55350] train/loss = 4.341785252094269
+I1201 22:27:14.023972 137274321021824 utils.py:1231] [55350] l2_grads = 1.3696850538253784
+I1201 22:27:14.024046 137274321021824 utils.py:1231] [55350] lr = 0.0005906258571907061
+I1201 22:27:14.024109 137274321021824 utils.py:1231] [55350] uptime = 347823.38646598597
+I1201 22:27:14.024168 137274321021824 utils.py:1231] [55350] examples_seen = 56678400.0
+I1201 22:27:14.024218 137274321021824 utils.py:1231] [55350] progress = 0.491549958704475
+I1201 22:27:14.024266 137274321021824 utils.py:1231] [55350] epoch = 44.239665867135194
+I1201 22:27:14.024316 137274321021824 utils.py:1231] [55350] img/sec/core = 164.21694888355583
+I1201 22:27:14.024374 137274321021824 utils.py:1231] [55350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.58328974930609
+I1201 22:27:14.024429 137274321021824 utils.py:1231] [55350] core_hours = 96.58328974930609
+I1201 22:27:14.024502 137274321021824 train.py:125] NOTE: Steps:55350/112603 [49.2%]
+Walltime:4d0h37m (0s eval)
+ETA:4d3h54m
+Total train time:8d4h29m
+I1201 22:32:25.803279 137274321021824 utils.py:1231] [55400] l2_params = 303.3060152750754
+I1201 22:32:25.803485 137274321021824 utils.py:1231] [55400] train/loss = 3.991381734609604
+I1201 22:32:25.803584 137274321021824 utils.py:1231] [55400] l2_grads = 1.4515963792800903
+I1201 22:32:25.803645 137274321021824 utils.py:1231] [55400] lr = 0.000589872957105747
+I1201 22:32:25.803695 137274321021824 utils.py:1231] [55400] uptime = 348135.166057256
+I1201 22:32:25.803744 137274321021824 utils.py:1231] [55400] examples_seen = 56729600.0
+I1201 22:32:25.803790 137274321021824 utils.py:1231] [55400] progress = 0.49199399660755044
+I1201 22:32:25.803839 137274321021824 utils.py:1231] [55400] epoch = 44.27962943160416
+I1201 22:32:25.803894 137274321021824 utils.py:1231] [55400] img/sec/core = 164.21857438275057
+I1201 22:32:25.803949 137274321021824 utils.py:1231] [55400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.66989519132555
+I1201 22:32:25.803998 137274321021824 utils.py:1231] [55400] core_hours = 96.66989519132555
+I1201 22:32:25.804057 137274321021824 train.py:125] NOTE: Steps:55400/112603 [49.2%]
+Walltime:4d0h42m (0s eval)
+ETA:4d3h49m
+Total train time:8d4h29m
+I1201 22:37:37.590510 137274321021824 utils.py:1231] [55450] l2_params = 303.22067697308427
+I1201 22:37:37.590764 137274321021824 utils.py:1231] [55450] train/loss = 3.9480142295360565
+I1201 22:37:37.591001 137274321021824 utils.py:1231] [55450] l2_grads = 1.360283374786377
+I1201 22:37:37.591137 137274321021824 utils.py:1231] [55450] lr = 0.0005891198463770357
+I1201 22:37:37.591247 137274321021824 utils.py:1231] [55450] uptime = 348446.953603915
+I1201 22:37:37.591353 137274321021824 utils.py:1231] [55450] examples_seen = 56780800.0
+I1201 22:37:37.591469 137274321021824 utils.py:1231] [55450] progress = 0.49243803451062584
+I1201 22:37:37.591567 137274321021824 utils.py:1231] [55450] epoch = 44.31959299607311
+I1201 22:37:37.591657 137274321021824 utils.py:1231] [55450] img/sec/core = 164.2143842774975
+I1201 22:37:37.591754 137274321021824 utils.py:1231] [55450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.75650284317527
+I1201 22:37:37.591836 137274321021824 utils.py:1231] [55450] core_hours = 96.75650284317527
+I1201 22:37:37.591946 137274321021824 train.py:125] NOTE: Steps:55450/112603 [49.2%]
+Walltime:4d0h47m (0s eval)
+ETA:4d3h43m
+Total train time:8d4h29m
+I1201 22:42:49.368516 137274321021824 utils.py:1231] [55500] l2_params = 303.13905880632916
+I1201 22:42:49.368720 137274321021824 utils.py:1231] [55500] train/loss = 3.8072835505008698
+I1201 22:42:49.368817 137274321021824 utils.py:1231] [55500] l2_grads = 1.5013636350631714
+I1201 22:42:49.368907 137274321021824 utils.py:1231] [55500] lr = 0.0005883665267697093
+I1201 22:42:49.368969 137274321021824 utils.py:1231] [55500] uptime = 348758.731330743
+I1201 22:42:49.369038 137274321021824 utils.py:1231] [55500] examples_seen = 56832000.0
+I1201 22:42:49.369093 137274321021824 utils.py:1231] [55500] progress = 0.49288207241370124
+I1201 22:42:49.369154 137274321021824 utils.py:1231] [55500] epoch = 44.35955656054207
+I1201 22:42:49.369225 137274321021824 utils.py:1231] [55500] img/sec/core = 164.2195564157316
+I1201 22:42:49.369283 137274321021824 utils.py:1231] [55500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.84310776729417
+I1201 22:42:49.369338 137274321021824 utils.py:1231] [55500] core_hours = 96.84310776729417
+I1201 22:42:49.369431 137274321021824 train.py:125] NOTE: Steps:55500/112603 [49.3%]
+Walltime:4d0h52m (0s eval)
+ETA:4d3h38m
+Total train time:8d4h29m
+I1201 22:48:01.127345 137274321021824 utils.py:1231] [55550] l2_params = 303.0567985964244
+I1201 22:48:01.127596 137274321021824 utils.py:1231] [55550] train/loss = 2.6129426062107086
+I1201 22:48:01.127760 137274321021824 utils.py:1231] [55550] l2_grads = 1.6084694862365723
+I1201 22:48:01.127872 137274321021824 utils.py:1231] [55550] lr = 0.0005876130000493937
+I1201 22:48:01.127968 137274321021824 utils.py:1231] [55550] uptime = 349070.490324753
+I1201 22:48:01.128051 137274321021824 utils.py:1231] [55550] examples_seen = 56883200.0
+I1201 22:48:01.128143 137274321021824 utils.py:1231] [55550] progress = 0.49332611031677664
+I1201 22:48:01.128222 137274321021824 utils.py:1231] [55550] epoch = 44.39952012501102
+I1201 22:48:01.128297 137274321021824 utils.py:1231] [55550] img/sec/core = 164.22942395804625
+I1201 22:48:01.128372 137274321021824 utils.py:1231] [55550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 96.9297074878525
+I1201 22:48:01.128449 137274321021824 utils.py:1231] [55550] core_hours = 96.9297074878525
+I1201 22:48:01.128529 137274321021824 train.py:125] NOTE: Steps:55550/112603 [49.3%]
+Walltime:4d0h57m (0s eval)
+ETA:4d3h33m
+Total train time:8d4h29m
+I1201 22:53:12.906107 137274321021824 utils.py:1231] [55600] l2_params = 302.9830308174878
+I1201 22:53:12.906341 137274321021824 utils.py:1231] [55600] train/loss = 2.283434748649597
+I1201 22:53:12.906440 137274321021824 utils.py:1231] [55600] l2_grads = 1.487984538078308
+I1201 22:53:12.906523 137274321021824 utils.py:1231] [55600] lr = 0.000586859267982201
+I1201 22:53:12.906585 137274321021824 utils.py:1231] [55600] uptime = 349382.268946975
+I1201 22:53:12.906641 137274321021824 utils.py:1231] [55600] examples_seen = 56934400.0
+I1201 22:53:12.906695 137274321021824 utils.py:1231] [55600] progress = 0.49377014821985205
+I1201 22:53:12.906763 137274321021824 utils.py:1231] [55600] epoch = 44.439483689479985
+I1201 22:53:12.906822 137274321021824 utils.py:1231] [55600] img/sec/core = 164.21908479517444
+I1201 22:53:12.906889 137274321021824 utils.py:1231] [55600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.01631266069194
+I1201 22:53:12.906945 137274321021824 utils.py:1231] [55600] core_hours = 97.01631266069194
+I1201 22:53:12.907008 137274321021824 train.py:125] NOTE: Steps:55600/112603 [49.4%]
+Walltime:4d1h3m (0s eval)
+ETA:4d3h28m
+Total train time:8d4h29m
+I1201 22:58:24.681522 137274321021824 utils.py:1231] [55650] l2_params = 302.90886623573493
+I1201 22:58:24.681723 137274321021824 utils.py:1231] [55650] train/loss = 2.226188689470291
+I1201 22:58:24.681815 137274321021824 utils.py:1231] [55650] l2_grads = 1.6550178527832031
+I1201 22:58:24.681872 137274321021824 utils.py:1231] [55650] lr = 0.0005861053323347246
+I1201 22:58:24.681926 137274321021824 utils.py:1231] [55650] uptime = 349694.044288041
+I1201 22:58:24.681976 137274321021824 utils.py:1231] [55650] examples_seen = 56985600.0
+I1201 22:58:24.682024 137274321021824 utils.py:1231] [55650] progress = 0.49421418612292745
+I1201 22:58:24.682070 137274321021824 utils.py:1231] [55650] epoch = 44.47944725394894
+I1201 22:58:24.682118 137274321021824 utils.py:1231] [55650] img/sec/core = 164.2208130538461
+I1201 22:58:24.682171 137274321021824 utils.py:1231] [55650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.10291692209917
+I1201 22:58:24.682221 137274321021824 utils.py:1231] [55650] core_hours = 97.10291692209917
+I1201 22:58:24.682278 137274321021824 train.py:125] NOTE: Steps:55650/112603 [49.4%]
+Walltime:4d1h8m (0s eval)
+ETA:4d3h22m
+Total train time:8d4h29m
+I1201 23:03:36.448098 137274321021824 utils.py:1231] [55700] l2_params = 302.835326278263
+I1201 23:03:36.448317 137274321021824 utils.py:1231] [55700] train/loss = 2.370321750640869
+I1201 23:03:36.448437 137274321021824 utils.py:1231] [55700] l2_grads = 1.571252465248108
+I1201 23:03:36.448509 137274321021824 utils.py:1231] [55700] lr = 0.0005853511948740343
+I1201 23:03:36.448571 137274321021824 utils.py:1231] [55700] uptime = 350005.81093239697
+I1201 23:03:36.448634 137274321021824 utils.py:1231] [55700] examples_seen = 57036800.0
+I1201 23:03:36.448699 137274321021824 utils.py:1231] [55700] progress = 0.49465822402600285
+I1201 23:03:36.448761 137274321021824 utils.py:1231] [55700] epoch = 44.519410818417896
+I1201 23:03:36.448822 137274321021824 utils.py:1231] [55700] img/sec/core = 164.22539398263132
+I1201 23:03:36.448902 137274321021824 utils.py:1231] [55700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.1895187677536
+I1201 23:03:36.448961 137274321021824 utils.py:1231] [55700] core_hours = 97.1895187677536
+I1201 23:03:36.449028 137274321021824 train.py:125] NOTE: Steps:55700/112603 [49.5%]
+Walltime:4d1h13m (0s eval)
+ETA:4d3h17m
+Total train time:8d4h29m
+I1201 23:08:48.228776 137274321021824 utils.py:1231] [55750] l2_params = 302.7439606142541
+I1201 23:08:48.229022 137274321021824 utils.py:1231] [55750] train/loss = 2.4416153579950333
+I1201 23:08:48.229156 137274321021824 utils.py:1231] [55750] l2_grads = 1.6210803985595703
+I1201 23:08:48.229227 137274321021824 utils.py:1231] [55750] lr = 0.0005845968573676734
+I1201 23:08:48.229283 137274321021824 utils.py:1231] [55750] uptime = 350317.591645373
+I1201 23:08:48.229339 137274321021824 utils.py:1231] [55750] examples_seen = 57088000.0
+I1201 23:08:48.229387 137274321021824 utils.py:1231] [55750] progress = 0.49510226192907825
+I1201 23:08:48.229433 137274321021824 utils.py:1231] [55750] epoch = 44.55937438288685
+I1201 23:08:48.229485 137274321021824 utils.py:1231] [55750] img/sec/core = 164.2179835669969
+I1201 23:08:48.229540 137274321021824 utils.py:1231] [55750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.27612452135804
+I1201 23:08:48.229589 137274321021824 utils.py:1231] [55750] core_hours = 97.27612452135804
+I1201 23:08:48.229648 137274321021824 train.py:125] NOTE: Steps:55750/112603 [49.5%]
+Walltime:4d1h18m (0s eval)
+ETA:4d3h12m
+Total train time:8d4h29m
+I1201 23:14:00.011362 137274321021824 utils.py:1231] [55800] l2_params = 302.6575989798997
+I1201 23:14:00.011589 137274321021824 utils.py:1231] [55800] train/loss = 2.5589789152145386
+I1201 23:14:00.011699 137274321021824 utils.py:1231] [55800] l2_grads = 1.616214394569397
+I1201 23:14:00.011770 137274321021824 utils.py:1231] [55800] lr = 0.0005838423215836547
+I1201 23:14:00.011830 137274321021824 utils.py:1231] [55800] uptime = 350629.37419103197
+I1201 23:14:00.011896 137274321021824 utils.py:1231] [55800] examples_seen = 57139200.0
+I1201 23:14:00.011954 137274321021824 utils.py:1231] [55800] progress = 0.49554629983215365
+I1201 23:14:00.012012 137274321021824 utils.py:1231] [55800] epoch = 44.59933794735581
+I1201 23:14:00.012071 137274321021824 utils.py:1231] [55800] img/sec/core = 164.21701828042427
+I1201 23:14:00.012130 137274321021824 utils.py:1231] [55800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.3627307840411
+I1201 23:14:00.012195 137274321021824 utils.py:1231] [55800] core_hours = 97.3627307840411
+I1201 23:14:00.012289 137274321021824 train.py:125] NOTE: Steps:55800/112603 [49.6%]
+Walltime:4d1h23m (0s eval)
+ETA:4d3h6m
+Total train time:8d4h28m
+I1201 23:19:11.786094 137274321021824 utils.py:1231] [55850] l2_params = 302.562073192914
+I1201 23:19:11.786350 137274321021824 utils.py:1231] [55850] train/loss = 2.2438489198684692
+I1201 23:19:11.786482 137274321021824 utils.py:1231] [55850] l2_grads = 1.5901175737380981
+I1201 23:19:11.786569 137274321021824 utils.py:1231] [55850] lr = 0.0005830875892904548
+I1201 23:19:11.786640 137274321021824 utils.py:1231] [55850] uptime = 350941.149001327
+I1201 23:19:11.786708 137274321021824 utils.py:1231] [55850] examples_seen = 57190400.0
+I1201 23:19:11.786769 137274321021824 utils.py:1231] [55850] progress = 0.4959903377352291
+I1201 23:19:11.786825 137274321021824 utils.py:1231] [55850] epoch = 44.63930151182477
+I1201 23:19:11.786888 137274321021824 utils.py:1231] [55850] img/sec/core = 164.22109262627777
+I1201 23:19:11.786952 137274321021824 utils.py:1231] [55850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.44933489801194
+I1201 23:19:11.787003 137274321021824 utils.py:1231] [55850] core_hours = 97.44933489801194
+I1201 23:19:11.787068 137274321021824 train.py:125] NOTE: Steps:55850/112603 [49.6%]
+Walltime:4d1h29m (0s eval)
+ETA:4d3h1m
+Total train time:8d4h28m
+I1201 23:24:23.562351 137274321021824 utils.py:1231] [55900] l2_params = 302.45994657876133
+I1201 23:24:23.562609 137274321021824 utils.py:1231] [55900] train/loss = 4.174044370651245
+I1201 23:24:23.562748 137274321021824 utils.py:1231] [55900] l2_grads = 1.3492389917373657
+I1201 23:24:23.562843 137274321021824 utils.py:1231] [55900] lr = 0.000582332662257011
+I1201 23:24:23.562911 137274321021824 utils.py:1231] [55900] uptime = 351252.925271245
+I1201 23:24:23.562972 137274321021824 utils.py:1231] [55900] examples_seen = 57241600.0
+I1201 23:24:23.563029 137274321021824 utils.py:1231] [55900] progress = 0.4964343756383045
+I1201 23:24:23.563085 137274321021824 utils.py:1231] [55900] epoch = 44.679265076293724
+I1201 23:24:23.563143 137274321021824 utils.py:1231] [55900] img/sec/core = 164.2203238029095
+I1201 23:24:23.563205 137274321021824 utils.py:1231] [55900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.53593941743361
+I1201 23:24:23.563262 137274321021824 utils.py:1231] [55900] core_hours = 97.53593941743361
+I1201 23:24:23.563332 137274321021824 train.py:125] NOTE: Steps:55900/112603 [49.6%]
+Walltime:4d1h34m (0s eval)
+ETA:4d2h56m
+Total train time:8d4h28m
+I1201 23:29:35.347137 137274321021824 utils.py:1231] [55950] l2_params = 302.37776024277275
+I1201 23:29:35.347389 137274321021824 utils.py:1231] [55950] train/loss = 2.243025481700897
+I1201 23:29:35.347507 137274321021824 utils.py:1231] [55950] l2_grads = 1.5305291414260864
+I1201 23:29:35.347589 137274321021824 utils.py:1231] [55950] lr = 0.0005815775422527176
+I1201 23:29:35.347652 137274321021824 utils.py:1231] [55950] uptime = 351564.710013878
+I1201 23:29:35.347713 137274321021824 utils.py:1231] [55950] examples_seen = 57292800.0
+I1201 23:29:35.347767 137274321021824 utils.py:1231] [55950] progress = 0.4968784135413799
+I1201 23:29:35.347815 137274321021824 utils.py:1231] [55950] epoch = 44.71922864076268
+I1201 23:29:35.347901 137274321021824 utils.py:1231] [55950] img/sec/core = 164.21586113425593
+I1201 23:29:35.347963 137274321021824 utils.py:1231] [55950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.62254629038722
+I1201 23:29:35.348016 137274321021824 utils.py:1231] [55950] core_hours = 97.62254629038722
+I1201 23:29:35.348076 137274321021824 train.py:125] NOTE: Steps:55950/112603 [49.7%]
+Walltime:4d1h39m (0s eval)
+ETA:4d2h51m
+Total train time:8d4h28m
+I1201 23:34:47.120965 137274321021824 utils.py:1231] [56000] l2_params = 302.2954848583797
+I1201 23:34:47.121248 137274321021824 utils.py:1231] [56000] train/loss = 2.280812829732895
+I1201 23:34:47.121403 137274321021824 utils.py:1231] [56000] l2_grads = 1.6139827966690063
+I1201 23:34:47.121495 137274321021824 utils.py:1231] [56000] lr = 0.0005808222310474211
+I1201 23:34:47.121590 137274321021824 utils.py:1231] [56000] uptime = 351876.483949486
+I1201 23:34:47.121674 137274321021824 utils.py:1231] [56000] examples_seen = 57344000.0
+I1201 23:34:47.121726 137274321021824 utils.py:1231] [56000] progress = 0.4973224514444553
+I1201 23:34:47.121782 137274321021824 utils.py:1231] [56000] epoch = 44.759192205231635
+I1201 23:34:47.121843 137274321021824 utils.py:1231] [56000] img/sec/core = 164.221553351328
+I1201 23:34:47.121912 137274321021824 utils.py:1231] [56000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.70915016138945
+I1201 23:34:47.121976 137274321021824 utils.py:1231] [56000] core_hours = 97.70915016138945
+I1201 23:34:47.122045 137274321021824 train.py:125] NOTE: Steps:56000/112603 [49.7%]
+Walltime:4d1h44m (0s eval)
+ETA:4d2h45m
+Total train time:8d4h28m
+I1201 23:39:59.227082 137274321021824 utils.py:1231] [56050] l2_params = 302.20601431868374
+I1201 23:39:59.227327 137274321021824 utils.py:1231] [56050] train/loss = 3.4517483711242676
+I1201 23:39:59.227461 137274321021824 utils.py:1231] [56050] l2_grads = 1.4151031970977783
+I1201 23:39:59.227552 137274321021824 utils.py:1231] [56050] lr = 0.0005800667304114149
+I1201 23:39:59.227624 137274321021824 utils.py:1231] [56050] uptime = 352188.589980721
+I1201 23:39:59.227701 137274321021824 utils.py:1231] [56050] examples_seen = 57395200.0
+I1201 23:39:59.227780 137274321021824 utils.py:1231] [56050] progress = 0.4977664893475307
+I1201 23:39:59.227846 137274321021824 utils.py:1231] [56050] epoch = 44.79915576970059
+I1201 23:39:59.227923 137274321021824 utils.py:1231] [56050] img/sec/core = 164.04681382605614
+I1201 23:39:59.227996 137274321021824 utils.py:1231] [56050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.79584628117694
+I1201 23:39:59.228075 137274321021824 utils.py:1231] [56050] core_hours = 97.79584628117694
+I1201 23:39:59.228153 137274321021824 train.py:125] NOTE: Steps:56050/112603 [49.8%]
+Walltime:4d1h49m (0s eval)
+ETA:4d2h40m
+Total train time:8d4h28m
+I1201 23:45:10.911535 137274321021824 utils.py:1231] [56100] l2_params = 302.12159533213844
+I1201 23:45:10.911800 137274321021824 utils.py:1231] [56100] train/loss = 2.9972204864025116
+I1201 23:45:10.911929 137274321021824 utils.py:1231] [56100] l2_grads = 1.4209836721420288
+I1201 23:45:10.912023 137274321021824 utils.py:1231] [56100] lr = 0.0005793110421154371
+I1201 23:45:10.912097 137274321021824 utils.py:1231] [56100] uptime = 352500.27445471
+I1201 23:45:10.912175 137274321021824 utils.py:1231] [56100] examples_seen = 57446400.0
+I1201 23:45:10.912239 137274321021824 utils.py:1231] [56100] progress = 0.4982105272506061
+I1201 23:45:10.912304 137274321021824 utils.py:1231] [56100] epoch = 44.83911933416955
+I1201 23:45:10.912363 137274321021824 utils.py:1231] [56100] img/sec/core = 164.2686892443854
+I1201 23:45:10.912424 137274321021824 utils.py:1231] [56100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.88242530172944
+I1201 23:45:10.912480 137274321021824 utils.py:1231] [56100] core_hours = 97.88242530172944
+I1201 23:45:10.912544 137274321021824 train.py:125] NOTE: Steps:56100/112603 [49.8%]
+Walltime:4d1h55m (0s eval)
+ETA:4d2h35m
+Total train time:8d4h28m
+I1201 23:50:22.506361 137274321021824 utils.py:1231] [56150] l2_params = 302.0349702612869
+I1201 23:50:22.506559 137274321021824 utils.py:1231] [56150] train/loss = 2.3821824193000793
+I1201 23:50:22.506661 137274321021824 utils.py:1231] [56150] l2_grads = 1.6426819562911987
+I1201 23:50:22.506727 137274321021824 utils.py:1231] [56150] lr = 0.0005785551679306671
+I1201 23:50:22.506793 137274321021824 utils.py:1231] [56150] uptime = 352811.869154851
+I1201 23:50:22.506854 137274321021824 utils.py:1231] [56150] examples_seen = 57497600.0
+I1201 23:50:22.506928 137274321021824 utils.py:1231] [56150] progress = 0.4986545651536815
+I1201 23:50:22.506983 137274321021824 utils.py:1231] [56150] epoch = 44.87908289863851
+I1201 23:50:22.507038 137274321021824 utils.py:1231] [56150] img/sec/core = 164.31601685404877
+I1201 23:50:22.507098 137274321021824 utils.py:1231] [56150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 97.96897938510195
+I1201 23:50:22.507148 137274321021824 utils.py:1231] [56150] core_hours = 97.96897938510195
+I1201 23:50:22.507208 137274321021824 train.py:125] NOTE: Steps:56150/112603 [49.9%]
+Walltime:4d2h0m (0s eval)
+ETA:4d2h30m
+Total train time:8d4h28m
+I1201 23:55:34.284946 137274321021824 utils.py:1231] [56200] l2_params = 301.94945008670703
+I1201 23:55:34.285131 137274321021824 utils.py:1231] [56200] train/loss = 2.2989799231290817
+I1201 23:55:34.285222 137274321021824 utils.py:1231] [56200] l2_grads = 1.5718014240264893
+I1201 23:55:34.285279 137274321021824 utils.py:1231] [56200] lr = 0.0005777991096287176
+I1201 23:55:34.285329 137274321021824 utils.py:1231] [56200] uptime = 353123.64769177197
+I1201 23:55:34.285379 137274321021824 utils.py:1231] [56200] examples_seen = 57548800.0
+I1201 23:55:34.285425 137274321021824 utils.py:1231] [56200] progress = 0.4990986030567569
+I1201 23:55:34.285474 137274321021824 utils.py:1231] [56200] epoch = 44.91904646310746
+I1201 23:55:34.285523 137274321021824 utils.py:1231] [56200] img/sec/core = 164.21912972469926
+I1201 23:55:34.285575 137274321021824 utils.py:1231] [56200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.05558453424666
+I1201 23:55:34.285622 137274321021824 utils.py:1231] [56200] core_hours = 98.05558453424666
+I1201 23:55:34.285688 137274321021824 train.py:125] NOTE: Steps:56200/112603 [49.9%]
+Walltime:4d2h5m (0s eval)
+ETA:4d2h24m
+Total train time:8d4h28m
+I1202 00:00:46.054542 137274321021824 utils.py:1231] [56250] l2_params = 301.87398414983875
+I1202 00:00:46.054783 137274321021824 utils.py:1231] [56250] train/loss = 2.3606571555137634
+I1202 00:00:46.054912 137274321021824 utils.py:1231] [56250] l2_grads = 1.6079496145248413
+I1202 00:00:46.054972 137274321021824 utils.py:1231] [56250] lr = 0.0005770428689816351
+I1202 00:00:46.055022 137274321021824 utils.py:1231] [56250] uptime = 353435.417384113
+I1202 00:00:46.055072 137274321021824 utils.py:1231] [56250] examples_seen = 57600000.0
+I1202 00:00:46.055125 137274321021824 utils.py:1231] [56250] progress = 0.4995426409598323
+I1202 00:00:46.055170 137274321021824 utils.py:1231] [56250] epoch = 44.95901002757642
+I1202 00:00:46.055224 137274321021824 utils.py:1231] [56250] img/sec/core = 164.22378844956188
+I1202 00:00:46.055285 137274321021824 utils.py:1231] [56250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.14218722656362
+I1202 00:00:46.055340 137274321021824 utils.py:1231] [56250] core_hours = 98.14218722656362
+I1202 00:00:46.055405 137274321021824 train.py:125] NOTE: Steps:56250/112603 [50.0%]
+Walltime:4d2h10m (0s eval)
+ETA:4d2h19m
+Total train time:8d4h28m
+I1202 00:05:57.828617 137274321021824 utils.py:1231] [56300] l2_params = 301.80072706273916
+I1202 00:05:57.828860 137274321021824 utils.py:1231] [56300] train/loss = 4.625911235809326
+I1202 00:05:57.828961 137274321021824 utils.py:1231] [56300] l2_grads = 1.5323108434677124
+I1202 00:05:57.829041 137274321021824 utils.py:1231] [56300] lr = 0.0005762864477618915
+I1202 00:05:57.829091 137274321021824 utils.py:1231] [56300] uptime = 353747.191453445
+I1202 00:05:57.829142 137274321021824 utils.py:1231] [56300] examples_seen = 57651200.0
+I1202 00:05:57.829188 137274321021824 utils.py:1231] [56300] progress = 0.49998667886290776
+I1202 00:05:57.829234 137274321021824 utils.py:1231] [56300] epoch = 44.998973592045374
+I1202 00:05:57.829284 137274321021824 utils.py:1231] [56300] img/sec/core = 164.22148291453797
+I1202 00:05:57.829337 137274321021824 utils.py:1231] [56300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.22879113471139
+I1202 00:05:57.829386 137274321021824 utils.py:1231] [56300] core_hours = 98.22879113471139
+I1202 00:05:57.829443 137274321021824 train.py:125] NOTE: Steps:56300/112603 [50.0%]
+Walltime:4d2h15m (0s eval)
+ETA:4d2h14m
+Total train time:8d4h28m
+I1202 00:11:09.600522 137274321021824 utils.py:1231] [56350] l2_params = 301.7111979595818
+I1202 00:11:09.600790 137274321021824 utils.py:1231] [56350] train/loss = 3.142241448163986
+I1202 00:11:09.600924 137274321021824 utils.py:1231] [56350] l2_grads = 1.3786324262619019
+I1202 00:11:09.601004 137274321021824 utils.py:1231] [56350] lr = 0.0005755298477423831
+I1202 00:11:09.601065 137274321021824 utils.py:1231] [56350] uptime = 354058.963426767
+I1202 00:11:09.601130 137274321021824 utils.py:1231] [56350] examples_seen = 57702400.0
+I1202 00:11:09.601186 137274321021824 utils.py:1231] [56350] progress = 0.5004307167659832
+I1202 00:11:09.601241 137274321021824 utils.py:1231] [56350] epoch = 45.038937156514336
+I1202 00:11:09.601299 137274321021824 utils.py:1231] [56350] img/sec/core = 164.22258695820963
+I1202 00:11:09.601361 137274321021824 utils.py:1231] [56350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.31539446063417
+I1202 00:11:09.601418 137274321021824 utils.py:1231] [56350] core_hours = 98.31539446063417
+I1202 00:11:09.601484 137274321021824 train.py:125] NOTE: Steps:56350/112603 [50.0%]
+Walltime:4d2h20m (0s eval)
+ETA:4d2h8m
+Total train time:8d4h28m
+I1202 00:16:21.366824 137274321021824 utils.py:1231] [56400] l2_params = 301.62960515931263
+I1202 00:16:21.367035 137274321021824 utils.py:1231] [56400] train/loss = 3.156880885362625
+I1202 00:16:21.367133 137274321021824 utils.py:1231] [56400] l2_grads = 1.453433632850647
+I1202 00:16:21.367204 137274321021824 utils.py:1231] [56400] lr = 0.000574773070696424
+I1202 00:16:21.367262 137274321021824 utils.py:1231] [56400] uptime = 354370.729623512
+I1202 00:16:21.367327 137274321021824 utils.py:1231] [56400] examples_seen = 57753600.0
+I1202 00:16:21.367390 137274321021824 utils.py:1231] [56400] progress = 0.5008747546690585
+I1202 00:16:21.367448 137274321021824 utils.py:1231] [56400] epoch = 45.07890072098329
+I1202 00:16:21.367506 137274321021824 utils.py:1231] [56400] img/sec/core = 164.22562976534275
+I1202 00:16:21.367569 137274321021824 utils.py:1231] [56400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.40199618195223
+I1202 00:16:21.367629 137274321021824 utils.py:1231] [56400] core_hours = 98.40199618195223
+I1202 00:16:21.367697 137274321021824 train.py:125] NOTE: Steps:56400/112603 [50.1%]
+Walltime:4d2h26m (0s eval)
+ETA:4d2h3m
+Total train time:8d4h28m
+I1202 00:21:33.138227 137274321021824 utils.py:1231] [56450] l2_params = 301.527714666183
+I1202 00:21:33.138486 137274321021824 utils.py:1231] [56450] train/loss = 3.056733787059784
+I1202 00:21:33.138613 137274321021824 utils.py:1231] [56450] l2_grads = 1.4670677185058594
+I1202 00:21:33.138715 137274321021824 utils.py:1231] [56450] lr = 0.0005740161183977454
+I1202 00:21:33.138776 137274321021824 utils.py:1231] [56450] uptime = 354682.501138238
+I1202 00:21:33.138844 137274321021824 utils.py:1231] [56450] examples_seen = 57804800.0
+I1202 00:21:33.138923 137274321021824 utils.py:1231] [56450] progress = 0.501318792572134
+I1202 00:21:33.138979 137274321021824 utils.py:1231] [56450] epoch = 45.11886428545225
+I1202 00:21:33.139073 137274321021824 utils.py:1231] [56450] img/sec/core = 164.22282851914505
+I1202 00:21:33.139137 137274321021824 utils.py:1231] [56450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.48859938048722
+I1202 00:21:33.139189 137274321021824 utils.py:1231] [56450] core_hours = 98.48859938048722
+I1202 00:21:33.139256 137274321021824 train.py:125] NOTE: Steps:56450/112603 [50.1%]
+Walltime:4d2h31m (0s eval)
+ETA:4d1h58m
+Total train time:8d4h27m
+I1202 00:26:44.917973 137274321021824 utils.py:1231] [56500] l2_params = 301.4521414166784
+I1202 00:26:44.918179 137274321021824 utils.py:1231] [56500] train/loss = 3.199169158935547
+I1202 00:26:44.918268 137274321021824 utils.py:1231] [56500] l2_grads = 1.3866008520126343
+I1202 00:26:44.918325 137274321021824 utils.py:1231] [56500] lr = 0.0005732589926204876
+I1202 00:26:44.918375 137274321021824 utils.py:1231] [56500] uptime = 354994.280737817
+I1202 00:26:44.918427 137274321021824 utils.py:1231] [56500] examples_seen = 57856000.0
+I1202 00:26:44.918475 137274321021824 utils.py:1231] [56500] progress = 0.5017628304752093
+I1202 00:26:44.918523 137274321021824 utils.py:1231] [56500] epoch = 45.1588278499212
+I1202 00:26:44.918572 137274321021824 utils.py:1231] [56500] img/sec/core = 164.2185700063031
+I1202 00:26:44.918626 137274321021824 utils.py:1231] [56500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.57520482481472
+I1202 00:26:44.918674 137274321021824 utils.py:1231] [56500] core_hours = 98.57520482481472
+I1202 00:26:44.918732 137274321021824 train.py:125] NOTE: Steps:56500/112603 [50.2%]
+Walltime:4d2h36m (0s eval)
+ETA:4d1h53m
+Total train time:8d4h27m
+I1202 00:31:56.682275 137274321021824 utils.py:1231] [56550] l2_params = 301.37580959513133
+I1202 00:31:56.682482 137274321021824 utils.py:1231] [56550] train/loss = 2.3039152324199677
+I1202 00:31:56.682577 137274321021824 utils.py:1231] [56550] l2_grads = 1.6021252870559692
+I1202 00:31:56.682643 137274321021824 utils.py:1231] [56550] lr = 0.000572501695139198
+I1202 00:31:56.682695 137274321021824 utils.py:1231] [56550] uptime = 355306.04505730397
+I1202 00:31:56.682748 137274321021824 utils.py:1231] [56550] examples_seen = 57907200.0
+I1202 00:31:56.682798 137274321021824 utils.py:1231] [56550] progress = 0.5022068683782848
+I1202 00:31:56.682846 137274321021824 utils.py:1231] [56550] epoch = 45.198791414390165
+I1202 00:31:56.682902 137274321021824 utils.py:1231] [56550] img/sec/core = 164.2266186337567
+I1202 00:31:56.682959 137274321021824 utils.py:1231] [56550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.6618060246722
+I1202 00:31:56.683010 137274321021824 utils.py:1231] [56550] core_hours = 98.6618060246722
+I1202 00:31:56.683069 137274321021824 train.py:125] NOTE: Steps:56550/112603 [50.2%]
+Walltime:4d2h41m (0s eval)
+ETA:4d1h47m
+Total train time:8d4h27m
+I1202 00:37:08.458535 137274321021824 utils.py:1231] [56600] l2_params = 301.30595791435854
+I1202 00:37:08.458750 137274321021824 utils.py:1231] [56600] train/loss = 2.628742963075638
+I1202 00:37:08.458861 137274321021824 utils.py:1231] [56600] l2_grads = 1.504475474357605
+I1202 00:37:08.458944 137274321021824 utils.py:1231] [56600] lr = 0.0005717442277288265
+I1202 00:37:08.459007 137274321021824 utils.py:1231] [56600] uptime = 355617.821367552
+I1202 00:37:08.459069 137274321021824 utils.py:1231] [56600] examples_seen = 57958400.0
+I1202 00:37:08.459128 137274321021824 utils.py:1231] [56600] progress = 0.5026509062813602
+I1202 00:37:08.459184 137274321021824 utils.py:1231] [56600] epoch = 45.23875497885912
+I1202 00:37:08.459244 137274321021824 utils.py:1231] [56600] img/sec/core = 164.22030256010095
+I1202 00:37:08.459308 137274321021824 utils.py:1231] [56600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.74841055529666
+I1202 00:37:08.459371 137274321021824 utils.py:1231] [56600] core_hours = 98.74841055529666
+I1202 00:37:08.459438 137274321021824 train.py:125] NOTE: Steps:56600/112603 [50.3%]
+Walltime:4d2h46m (0s eval)
+ETA:4d1h42m
+Total train time:8d4h27m
+I1202 00:42:20.214859 137274321021824 utils.py:1231] [56650] l2_params = 301.2246988449939
+I1202 00:42:20.215131 137274321021824 utils.py:1231] [56650] train/loss = 2.1645597219467163
+I1202 00:42:20.215259 137274321021824 utils.py:1231] [56650] l2_grads = 1.525899887084961
+I1202 00:42:20.215346 137274321021824 utils.py:1231] [56650] lr = 0.0005709865921647213
+I1202 00:42:20.215403 137274321021824 utils.py:1231] [56650] uptime = 355929.577765351
+I1202 00:42:20.215461 137274321021824 utils.py:1231] [56650] examples_seen = 58009600.0
+I1202 00:42:20.215516 137274321021824 utils.py:1231] [56650] progress = 0.5030949441844356
+I1202 00:42:20.215577 137274321021824 utils.py:1231] [56650] epoch = 45.278718543328075
+I1202 00:42:20.215632 137274321021824 utils.py:1231] [56650] img/sec/core = 164.23079160995854
+I1202 00:42:20.215691 137274321021824 utils.py:1231] [56650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.83500955468527
+I1202 00:42:20.215746 137274321021824 utils.py:1231] [56650] core_hours = 98.83500955468527
+I1202 00:42:20.215807 137274321021824 train.py:125] NOTE: Steps:56650/112603 [50.3%]
+Walltime:4d2h52m (0s eval)
+ETA:4d1h37m
+Total train time:8d4h27m
+I1202 00:47:31.989748 137274321021824 utils.py:1231] [56700] l2_params = 301.1325637981631
+I1202 00:47:31.990008 137274321021824 utils.py:1231] [56700] train/loss = 4.632145583629608
+I1202 00:47:31.990121 137274321021824 utils.py:1231] [56700] l2_grads = 1.4456840753555298
+I1202 00:47:31.990209 137274321021824 utils.py:1231] [56700] lr = 0.0005702287902226237
+I1202 00:47:31.990325 137274321021824 utils.py:1231] [56700] uptime = 356241.352678393
+I1202 00:47:31.990429 137274321021824 utils.py:1231] [56700] examples_seen = 58060800.0
+I1202 00:47:31.990519 137274321021824 utils.py:1231] [56700] progress = 0.503538982087511
+I1202 00:47:31.990639 137274321021824 utils.py:1231] [56700] epoch = 45.31868210779703
+I1202 00:47:31.990747 137274321021824 utils.py:1231] [56700] img/sec/core = 164.2210385063864
+I1202 00:47:31.990856 137274321021824 utils.py:1231] [56700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 98.92161369719695
+I1202 00:47:31.990957 137274321021824 utils.py:1231] [56700] core_hours = 98.92161369719695
+I1202 00:47:31.991050 137274321021824 train.py:125] NOTE: Steps:56700/112603 [50.4%]
+Walltime:4d2h57m (0s eval)
+ETA:4d1h32m
+Total train time:8d4h27m
+I1202 00:52:43.739816 137274321021824 utils.py:1231] [56750] l2_params = 301.0500791550133
+I1202 00:52:43.740007 137274321021824 utils.py:1231] [56750] train/loss = 3.6827027797698975
+I1202 00:52:43.740098 137274321021824 utils.py:1231] [56750] l2_grads = 1.4617795944213867
+I1202 00:52:43.740157 137274321021824 utils.py:1231] [56750] lr = 0.0005694708236786665
+I1202 00:52:43.740212 137274321021824 utils.py:1231] [56750] uptime = 356553.10257432
+I1202 00:52:43.740266 137274321021824 utils.py:1231] [56750] examples_seen = 58112000.0
+I1202 00:52:43.740314 137274321021824 utils.py:1231] [56750] progress = 0.5039830199905864
+I1202 00:52:43.740362 137274321021824 utils.py:1231] [56750] epoch = 45.358645672265986
+I1202 00:52:43.740412 137274321021824 utils.py:1231] [56750] img/sec/core = 164.2342168158851
+I1202 00:52:43.740469 137274321021824 utils.py:1231] [56750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.00821089051
+I1202 00:52:43.740520 137274321021824 utils.py:1231] [56750] core_hours = 99.00821089051
+I1202 00:52:43.740580 137274321021824 train.py:125] NOTE: Steps:56750/112603 [50.4%]
+Walltime:4d3h2m (0s eval)
+ETA:4d1h26m
+Total train time:8d4h27m
+I1202 00:57:55.511730 137274321021824 utils.py:1231] [56800] l2_params = 300.95646225655577
+I1202 00:57:55.511942 137274321021824 utils.py:1231] [56800] train/loss = 2.8409015834331512
+I1202 00:57:55.512070 137274321021824 utils.py:1231] [56800] l2_grads = 1.490911602973938
+I1202 00:57:55.512153 137274321021824 utils.py:1231] [56800] lr = 0.0005687126943093671
+I1202 00:57:55.512225 137274321021824 utils.py:1231] [56800] uptime = 356864.874583249
+I1202 00:57:55.512307 137274321021824 utils.py:1231] [56800] examples_seen = 58163200.0
+I1202 00:57:55.512372 137274321021824 utils.py:1231] [56800] progress = 0.5044270578936618
+I1202 00:57:55.512434 137274321021824 utils.py:1231] [56800] epoch = 45.39860923673495
+I1202 00:57:55.512505 137274321021824 utils.py:1231] [56800] img/sec/core = 164.2225682025892
+I1202 00:57:55.512569 137274321021824 utils.py:1231] [56800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.0948142263236
+I1202 00:57:55.512628 137274321021824 utils.py:1231] [56800] core_hours = 99.0948142263236
+I1202 00:57:55.512700 137274321021824 train.py:125] NOTE: Steps:56800/112603 [50.4%]
+Walltime:4d3h7m (0s eval)
+ETA:4d1h21m
+Total train time:8d4h27m
+I1202 01:03:07.293827 137274321021824 utils.py:1231] [56850] l2_params = 300.8655198219824
+I1202 01:03:07.294028 137274321021824 utils.py:1231] [56850] train/loss = 2.2487616539001465
+I1202 01:03:07.294137 137274321021824 utils.py:1231] [56850] l2_grads = 1.6551636457443237
+I1202 01:03:07.294206 137274321021824 utils.py:1231] [56850] lr = 0.0005679544038916256
+I1202 01:03:07.294269 137274321021824 utils.py:1231] [56850] uptime = 357176.656630229
+I1202 01:03:07.294343 137274321021824 utils.py:1231] [56850] examples_seen = 58214400.0
+I1202 01:03:07.294398 137274321021824 utils.py:1231] [56850] progress = 0.5048710957967372
+I1202 01:03:07.294452 137274321021824 utils.py:1231] [56850] epoch = 45.438572801203904
+I1202 01:03:07.294511 137274321021824 utils.py:1231] [56850] img/sec/core = 164.21728093691587
+I1202 01:03:07.294570 137274321021824 utils.py:1231] [56850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.18142035048471
+I1202 01:03:07.294624 137274321021824 utils.py:1231] [56850] core_hours = 99.18142035048471
+I1202 01:03:07.294688 137274321021824 train.py:125] NOTE: Steps:56850/112603 [50.5%]
+Walltime:4d3h12m (0s eval)
+ETA:4d1h16m
+Total train time:8d4h27m
+I1202 01:08:19.062775 137274321021824 utils.py:1231] [56900] l2_params = 300.77235294093884
+I1202 01:08:19.063003 137274321021824 utils.py:1231] [56900] train/loss = 2.4284113943576813
+I1202 01:08:19.063138 137274321021824 utils.py:1231] [56900] l2_grads = 1.4786595106124878
+I1202 01:08:19.063235 137274321021824 utils.py:1231] [56900] lr = 0.0005671959542027191
+I1202 01:08:19.063320 137274321021824 utils.py:1231] [56900] uptime = 357488.42567733297
+I1202 01:08:19.063401 137274321021824 utils.py:1231] [56900] examples_seen = 58265600.0
+I1202 01:08:19.063474 137274321021824 utils.py:1231] [56900] progress = 0.5053151336998126
+I1202 01:08:19.063555 137274321021824 utils.py:1231] [56900] epoch = 45.47853636567286
+I1202 01:08:19.063623 137274321021824 utils.py:1231] [56900] img/sec/core = 164.22412832702153
+I1202 01:08:19.063695 137274321021824 utils.py:1231] [56900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.26802286356914
+I1202 01:08:19.063764 137274321021824 utils.py:1231] [56900] core_hours = 99.26802286356914
+I1202 01:08:19.063838 137274321021824 train.py:125] NOTE: Steps:56900/112603 [50.5%]
+Walltime:4d3h18m (0s eval)
+ETA:4d1h10m
+Total train time:8d4h27m
+I1202 01:13:30.842445 137274321021824 utils.py:1231] [56950] l2_params = 300.6697308361254
+I1202 01:13:30.842692 137274321021824 utils.py:1231] [56950] train/loss = 4.476750493049622
+I1202 01:13:30.842829 137274321021824 utils.py:1231] [56950] l2_grads = 1.32370924949646
+I1202 01:13:30.842932 137274321021824 utils.py:1231] [56950] lr = 0.0005664373470202969
+I1202 01:13:30.843025 137274321021824 utils.py:1231] [56950] uptime = 357800.205386103
+I1202 01:13:30.843092 137274321021824 utils.py:1231] [56950] examples_seen = 58316800.0
+I1202 01:13:30.843152 137274321021824 utils.py:1231] [56950] progress = 0.505759171602888
+I1202 01:13:30.843210 137274321021824 utils.py:1231] [56950] epoch = 45.518499930141815
+I1202 01:13:30.843268 137274321021824 utils.py:1231] [56950] img/sec/core = 164.21851249390568
+I1202 01:13:30.843331 137274321021824 utils.py:1231] [56950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.3546283382275
+I1202 01:13:30.843391 137274321021824 utils.py:1231] [56950] core_hours = 99.3546283382275
+I1202 01:13:30.843461 137274321021824 train.py:125] NOTE: Steps:56950/112603 [50.6%]
+Walltime:4d3h23m (0s eval)
+ETA:4d1h5m
+Total train time:8d4h27m
+I1202 01:18:42.628515 137274321021824 utils.py:1231] [57000] l2_params = 300.59866614356315
+I1202 01:18:42.628762 137274321021824 utils.py:1231] [57000] train/loss = 3.434200644493103
+I1202 01:18:42.628889 137274321021824 utils.py:1231] [57000] l2_grads = 1.4356627464294434
+I1202 01:18:42.628979 137274321021824 utils.py:1231] [57000] lr = 0.0005656785841223787
+I1202 01:18:42.629059 137274321021824 utils.py:1231] [57000] uptime = 358111.99141676497
+I1202 01:18:42.629143 137274321021824 utils.py:1231] [57000] examples_seen = 58368000.0
+I1202 01:18:42.629209 137274321021824 utils.py:1231] [57000] progress = 0.5062032095059634
+I1202 01:18:42.629271 137274321021824 utils.py:1231] [57000] epoch = 45.55846349461077
+I1202 01:18:42.629335 137274321021824 utils.py:1231] [57000] img/sec/core = 164.21518273701017
+I1202 01:18:42.629406 137274321021824 utils.py:1231] [57000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.44123556896693
+I1202 01:18:42.629465 137274321021824 utils.py:1231] [57000] core_hours = 99.44123556896693
+I1202 01:18:42.629543 137274321021824 train.py:125] NOTE: Steps:57000/112603 [50.6%]
+Walltime:4d3h28m (0s eval)
+ETA:4d1h0m
+Total train time:8d4h27m
+I1202 01:23:53.869754 137274321021824 utils.py:1231] [57050] l2_params = 300.52788548630514
+I1202 01:23:53.869959 137274321021824 utils.py:1231] [57050] train/loss = 2.2564177215099335
+I1202 01:23:53.870059 137274321021824 utils.py:1231] [57050] l2_grads = 1.616241216659546
+I1202 01:23:53.870123 137274321021824 utils.py:1231] [57050] lr = 0.0005649196672873495
+I1202 01:23:53.870181 137274321021824 utils.py:1231] [57050] uptime = 358423.23254231596
+I1202 01:23:53.870235 137274321021824 utils.py:1231] [57050] examples_seen = 58419200.0
+I1202 01:23:53.870291 137274321021824 utils.py:1231] [57050] progress = 0.5066472474090389
+I1202 01:23:53.870339 137274321021824 utils.py:1231] [57050] epoch = 45.59842705907973
+I1202 01:23:53.870389 137274321021824 utils.py:1231] [57050] img/sec/core = 164.50268231539124
+I1202 01:23:53.870443 137274321021824 utils.py:1231] [57050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.52769143717553
+I1202 01:23:53.870492 137274321021824 utils.py:1231] [57050] core_hours = 99.52769143717553
+I1202 01:23:53.870552 137274321021824 train.py:125] NOTE: Steps:57050/112603 [50.7%]
+Walltime:4d3h33m (0s eval)
+ETA:4d0h55m
+Total train time:8d4h27m
+I1202 01:29:05.514536 137274321021824 utils.py:1231] [57100] l2_params = 300.43661582000084
+I1202 01:29:05.514782 137274321021824 utils.py:1231] [57100] train/loss = 2.8672560453414917
+I1202 01:29:05.514914 137274321021824 utils.py:1231] [57100] l2_grads = 1.5015615224838257
+I1202 01:29:05.515013 137274321021824 utils.py:1231] [57100] lr = 0.0005641605982939548
+I1202 01:29:05.515073 137274321021824 utils.py:1231] [57100] uptime = 358734.87743502297
+I1202 01:29:05.515133 137274321021824 utils.py:1231] [57100] examples_seen = 58470400.0
+I1202 01:29:05.515181 137274321021824 utils.py:1231] [57100] progress = 0.5070912853121142
+I1202 01:29:05.515233 137274321021824 utils.py:1231] [57100] epoch = 45.63839062354869
+I1202 01:29:05.515282 137274321021824 utils.py:1231] [57100] img/sec/core = 164.28955262275582
+I1202 01:29:05.515342 137274321021824 utils.py:1231] [57100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.61425946292749
+I1202 01:29:05.515394 137274321021824 utils.py:1231] [57100] core_hours = 99.61425946292749
+I1202 01:29:05.515452 137274321021824 train.py:125] NOTE: Steps:57100/112603 [50.7%]
+Walltime:4d3h38m (0s eval)
+ETA:4d0h49m
+Total train time:8d4h26m
+I1202 01:34:17.288071 137274321021824 utils.py:1231] [57150] l2_params = 300.36523086091535
+I1202 01:34:17.288295 137274321021824 utils.py:1231] [57150] train/loss = 2.3518910706043243
+I1202 01:34:17.288447 137274321021824 utils.py:1231] [57150] l2_grads = 1.6479319334030151
+I1202 01:34:17.288533 137274321021824 utils.py:1231] [57150] lr = 0.0005634013789212947
+I1202 01:34:17.288594 137274321021824 utils.py:1231] [57150] uptime = 359046.650955577
+I1202 01:34:17.288650 137274321021824 utils.py:1231] [57150] examples_seen = 58521600.0
+I1202 01:34:17.288701 137274321021824 utils.py:1231] [57150] progress = 0.5075353232151897
+I1202 01:34:17.288749 137274321021824 utils.py:1231] [57150] epoch = 45.67835418801764
+I1202 01:34:17.288802 137274321021824 utils.py:1231] [57150] img/sec/core = 164.22177197415533
+I1202 01:34:17.288858 137274321021824 utils.py:1231] [57150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.70086321863693
+I1202 01:34:17.288914 137274321021824 utils.py:1231] [57150] core_hours = 99.70086321863693
+I1202 01:34:17.288974 137274321021824 train.py:125] NOTE: Steps:57150/112603 [50.8%]
+Walltime:4d3h44m (0s eval)
+ETA:4d0h44m
+Total train time:8d4h26m
+I1202 01:39:28.932662 137274321021824 utils.py:1231] [57200] l2_params = 300.275433712484
+I1202 01:39:28.932889 137274321021824 utils.py:1231] [57200] train/loss = 2.305054545402527
+I1202 01:39:28.932990 137274321021824 utils.py:1231] [57200] l2_grads = 1.6830003261566162
+I1202 01:39:28.933055 137274321021824 utils.py:1231] [57200] lr = 0.0005626420109488253
+I1202 01:39:28.933119 137274321021824 utils.py:1231] [57200] uptime = 359358.295480332
+I1202 01:39:28.933179 137274321021824 utils.py:1231] [57200] examples_seen = 58572800.0
+I1202 01:39:28.933229 137274321021824 utils.py:1231] [57200] progress = 0.507979361118265
+I1202 01:39:28.933278 137274321021824 utils.py:1231] [57200] epoch = 45.7183177524866
+I1202 01:39:28.933328 137274321021824 utils.py:1231] [57200] img/sec/core = 164.2897465958967
+I1202 01:39:28.933386 137274321021824 utils.py:1231] [57200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.78743114217998
+I1202 01:39:28.933439 137274321021824 utils.py:1231] [57200] core_hours = 99.78743114217998
+I1202 01:39:28.933501 137274321021824 train.py:125] NOTE: Steps:57200/112603 [50.8%]
+Walltime:4d3h49m (0s eval)
+ETA:4d0h39m
+Total train time:8d4h26m
+I1202 01:44:40.715576 137274321021824 utils.py:1231] [57250] l2_params = 300.1962371512117
+I1202 01:44:40.715780 137274321021824 utils.py:1231] [57250] train/loss = 4.683926343917847
+I1202 01:44:40.715878 137274321021824 utils.py:1231] [57250] l2_grads = 1.4746253490447998
+I1202 01:44:40.715953 137274321021824 utils.py:1231] [57250] lr = 0.0005618824961563485
+I1202 01:44:40.716009 137274321021824 utils.py:1231] [57250] uptime = 359670.078370579
+I1202 01:44:40.716076 137274321021824 utils.py:1231] [57250] examples_seen = 58624000.0
+I1202 01:44:40.716130 137274321021824 utils.py:1231] [57250] progress = 0.5084233990213405
+I1202 01:44:40.716182 137274321021824 utils.py:1231] [57250] epoch = 45.758281316955554
+I1202 01:44:40.716243 137274321021824 utils.py:1231] [57250] img/sec/core = 164.21683678484533
+I1202 01:44:40.716299 137274321021824 utils.py:1231] [57250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.87403750058193
+I1202 01:44:40.716355 137274321021824 utils.py:1231] [57250] core_hours = 99.87403750058193
+I1202 01:44:40.716424 137274321021824 train.py:125] NOTE: Steps:57250/112603 [50.8%]
+Walltime:4d3h54m (0s eval)
+ETA:4d0h34m
+Total train time:8d4h26m
+I1202 01:49:52.350371 137274321021824 utils.py:1231] [57300] l2_params = 300.1006820131918
+I1202 01:49:52.350617 137274321021824 utils.py:1231] [57300] train/loss = 2.9835900962352753
+I1202 01:49:52.350719 137274321021824 utils.py:1231] [57300] l2_grads = 1.4188154935836792
+I1202 01:49:52.350802 137274321021824 utils.py:1231] [57300] lr = 0.0005611228363240106
+I1202 01:49:52.350865 137274321021824 utils.py:1231] [57300] uptime = 359981.713226498
+I1202 01:49:52.350947 137274321021824 utils.py:1231] [57300] examples_seen = 58675200.0
+I1202 01:49:52.351003 137274321021824 utils.py:1231] [57300] progress = 0.5088674369244158
+I1202 01:49:52.351059 137274321021824 utils.py:1231] [57300] epoch = 45.798244881424516
+I1202 01:49:52.351115 137274321021824 utils.py:1231] [57300] img/sec/core = 164.2948438775055
+I1202 01:49:52.351174 137274321021824 utils.py:1231] [57300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 99.96060273833722
+I1202 01:49:52.351229 137274321021824 utils.py:1231] [57300] core_hours = 99.96060273833722
+I1202 01:49:52.351294 137274321021824 train.py:125] NOTE: Steps:57300/112603 [50.9%]
+Walltime:4d3h59m (0s eval)
+ETA:4d0h28m
+Total train time:8d4h26m
+I1202 01:55:04.128179 137274321021824 utils.py:1231] [57350] l2_params = 300.0218229721967
+I1202 01:55:04.128412 137274321021824 utils.py:1231] [57350] train/loss = 2.818775564432144
+I1202 01:55:04.128535 137274321021824 utils.py:1231] [57350] l2_grads = 1.4322344064712524
+I1202 01:55:04.128607 137274321021824 utils.py:1231] [57350] lr = 0.0005603630332322988
+I1202 01:55:04.128671 137274321021824 utils.py:1231] [57350] uptime = 360293.491027821
+I1202 01:55:04.128724 137274321021824 utils.py:1231] [57350] examples_seen = 58726400.0
+I1202 01:55:04.128776 137274321021824 utils.py:1231] [57350] progress = 0.5093114748274913
+I1202 01:55:04.128824 137274321021824 utils.py:1231] [57350] epoch = 45.83820844589347
+I1202 01:55:04.128879 137274321021824 utils.py:1231] [57350] img/sec/core = 164.21951717774735
+I1202 01:55:04.128948 137274321021824 utils.py:1231] [57350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.04720768314915
+I1202 01:55:04.128997 137274321021824 utils.py:1231] [57350] core_hours = 100.04720768314915
+I1202 01:55:04.129055 137274321021824 train.py:125] NOTE: Steps:57350/112603 [50.9%]
+Walltime:4d4h4m (0s eval)
+ETA:4d0h23m
+Total train time:8d4h26m
+I1202 02:00:15.827125 137274321021824 utils.py:1231] [57400] l2_params = 299.9297598612938
+I1202 02:00:15.827327 137274321021824 utils.py:1231] [57400] train/loss = 2.320397913455963
+I1202 02:00:15.827428 137274321021824 utils.py:1231] [57400] l2_grads = 1.5298210382461548
+I1202 02:00:15.827486 137274321021824 utils.py:1231] [57400] lr = 0.0005596030886620354
+I1202 02:00:15.827536 137274321021824 utils.py:1231] [57400] uptime = 360605.189898181
+I1202 02:00:15.827587 137274321021824 utils.py:1231] [57400] examples_seen = 58777600.0
+I1202 02:00:15.827645 137274321021824 utils.py:1231] [57400] progress = 0.5097555127305666
+I1202 02:00:15.827693 137274321021824 utils.py:1231] [57400] epoch = 45.87817201036243
+I1202 02:00:15.827771 137274321021824 utils.py:1231] [57400] img/sec/core = 164.261102200531
+I1202 02:00:15.827828 137274321021824 utils.py:1231] [57400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.13379070269362
+I1202 02:00:15.827878 137274321021824 utils.py:1231] [57400] core_hours = 100.13379070269362
+I1202 02:00:15.827976 137274321021824 train.py:125] NOTE: Steps:57400/112603 [51.0%]
+Walltime:4d4h10m (0s eval)
+ETA:4d0h18m
+Total train time:8d4h26m
+I1202 02:05:27.603580 137274321021824 utils.py:1231] [57450] l2_params = 299.84881878435397
+I1202 02:05:27.603791 137274321021824 utils.py:1231] [57450] train/loss = 2.427319973707199
+I1202 02:05:27.603890 137274321021824 utils.py:1231] [57450] l2_grads = 1.5127429962158203
+I1202 02:05:27.603981 137274321021824 utils.py:1231] [57450] lr = 0.0005588430043943738
+I1202 02:05:27.604039 137274321021824 utils.py:1231] [57450] uptime = 360916.966401164
+I1202 02:05:27.604101 137274321021824 utils.py:1231] [57450] examples_seen = 58828800.0
+I1202 02:05:27.604151 137274321021824 utils.py:1231] [57450] progress = 0.5101995506336421
+I1202 02:05:27.604198 137274321021824 utils.py:1231] [57450] epoch = 45.91813557483138
+I1202 02:05:27.604248 137274321021824 utils.py:1231] [57450] img/sec/core = 164.2202010418695
+I1202 02:05:27.604303 137274321021824 utils.py:1231] [57450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.22039528685556
+I1202 02:05:27.604353 137274321021824 utils.py:1231] [57450] core_hours = 100.22039528685556
+I1202 02:05:27.604416 137274321021824 train.py:125] NOTE: Steps:57450/112603 [51.0%]
+Walltime:4d4h15m (0s eval)
+ETA:4d0h13m
+Total train time:8d4h26m
+I1202 02:10:39.384508 137274321021824 utils.py:1231] [57500] l2_params = 299.7718103089318
+I1202 02:10:39.384712 137274321021824 utils.py:1231] [57500] train/loss = 2.3732071220874786
+I1202 02:10:39.384829 137274321021824 utils.py:1231] [57500] l2_grads = 1.6953142881393433
+I1202 02:10:39.384933 137274321021824 utils.py:1231] [57500] lr = 0.0005580827822107959
+I1202 02:10:39.384998 137274321021824 utils.py:1231] [57500] uptime = 361228.74735937896
+I1202 02:10:39.385059 137274321021824 utils.py:1231] [57500] examples_seen = 58880000.0
+I1202 02:10:39.385117 137274321021824 utils.py:1231] [57500] progress = 0.5106435885367175
+I1202 02:10:39.385179 137274321021824 utils.py:1231] [57500] epoch = 45.958099139300344
+I1202 02:10:39.385237 137274321021824 utils.py:1231] [57500] img/sec/core = 164.21785439732173
+I1202 02:10:39.385299 137274321021824 utils.py:1231] [57500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.30700110858193
+I1202 02:10:39.385359 137274321021824 utils.py:1231] [57500] core_hours = 100.30700110858193
+I1202 02:10:39.385425 137274321021824 train.py:125] NOTE: Steps:57500/112603 [51.1%]
+Walltime:4d4h20m (0s eval)
+ETA:4d0h7m
+Total train time:8d4h26m
+I1202 02:10:39.385524 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:57500/112603 [51.1%]
+Walltime:4d4h20m (0s eval)
+ETA:4d0h7m
+Total train time:8d4h26m
+I1202 02:12:13.880264 137274321021824 utils.py:1231] [57500] val/acc@1 = 0.6631656568877551
+I1202 02:12:13.880505 137274321021824 utils.py:1231] [57500] val/loss = 1.406849354961697
+I1202 02:12:13.880696 137274321021824 utils.py:1231] [57500] z/secs/eval/val = 94.49508948903531
+I1202 02:12:13.880771 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 94.49508948903531
+I1202 02:17:24.118463 137274321021824 utils.py:1231] [57550] l2_params = 299.68124392724917
+I1202 02:17:24.118699 137274321021824 utils.py:1231] [57550] train/loss = 2.3616280257701874
+I1202 02:17:24.118814 137274321021824 utils.py:1231] [57550] l2_grads = 1.6126158237457275
+I1202 02:17:24.118888 137274321021824 utils.py:1231] [57550] lr = 0.000557322423893106
+I1202 02:17:24.118952 137274321021824 utils.py:1231] [57550] uptime = 361633.481314124
+I1202 02:17:24.119002 137274321021824 utils.py:1231] [57550] examples_seen = 58931200.0
+I1202 02:17:24.119048 137274321021824 utils.py:1231] [57550] progress = 0.5110876264397929
+I1202 02:17:24.119094 137274321021824 utils.py:1231] [57550] epoch = 45.9980627037693
+I1202 02:17:24.119142 137274321021824 utils.py:1231] [57550] img/sec/core = 126.50285304639544
+I1202 02:17:24.119194 137274321021824 utils.py:1231] [57550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.41942720712223
+I1202 02:17:24.119241 137274321021824 utils.py:1231] [57550] core_hours = 100.41942720712223
+I1202 02:17:24.119303 137274321021824 train.py:125] NOTE: Steps:57550/112603 [51.1%]
+Walltime:4d4h27m (0s eval)
+ETA:4d0h3m
+Total train time:8d4h29m
+I1202 02:22:35.904885 137274321021824 utils.py:1231] [57600] l2_params = 299.5941936973683
+I1202 02:22:35.905203 137274321021824 utils.py:1231] [57600] train/loss = 4.90668922662735
+I1202 02:22:35.905360 137274321021824 utils.py:1231] [57600] l2_grads = 1.6246362924575806
+I1202 02:22:35.905437 137274321021824 utils.py:1231] [57600] lr = 0.0005565619312234284
+I1202 02:22:35.905504 137274321021824 utils.py:1231] [57600] uptime = 361945.267865411
+I1202 02:22:35.905556 137274321021824 utils.py:1231] [57600] examples_seen = 58982400.0
+I1202 02:22:35.905612 137274321021824 utils.py:1231] [57600] progress = 0.5115316643428683
+I1202 02:22:35.905665 137274321021824 utils.py:1231] [57600] epoch = 46.038026268238255
+I1202 02:22:35.905726 137274321021824 utils.py:1231] [57600] img/sec/core = 164.21490852846927
+I1202 02:22:35.905791 137274321021824 utils.py:1231] [57600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.50603458247973
+I1202 02:22:35.905842 137274321021824 utils.py:1231] [57600] core_hours = 100.50603458247973
+I1202 02:22:35.905914 137274321021824 train.py:125] NOTE: Steps:57600/112603 [51.2%]
+Walltime:4d4h32m (0s eval)
+ETA:3d23h58m
+Total train time:8d4h29m
+I1202 02:27:47.745432 137274321021824 utils.py:1231] [57650] l2_params = 299.4993406612513
+I1202 02:27:47.745657 137274321021824 utils.py:1231] [57650] train/loss = 2.3710479736328125
+I1202 02:27:47.745755 137274321021824 utils.py:1231] [57650] l2_grads = 1.6960588693618774
+I1202 02:27:47.745820 137274321021824 utils.py:1231] [57650] lr = 0.0005558013059842014
+I1202 02:27:47.745894 137274321021824 utils.py:1231] [57650] uptime = 362257.108248042
+I1202 02:27:47.745973 137274321021824 utils.py:1231] [57650] examples_seen = 59033600.0
+I1202 02:27:47.746025 137274321021824 utils.py:1231] [57650] progress = 0.5119757022459437
+I1202 02:27:47.746073 137274321021824 utils.py:1231] [57650] epoch = 46.07798983270721
+I1202 02:27:47.746131 137274321021824 utils.py:1231] [57650] img/sec/core = 164.18656098362237
+I1202 02:27:47.746187 137274321021824 utils.py:1231] [57650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.59265691098834
+I1202 02:27:47.746237 137274321021824 utils.py:1231] [57650] core_hours = 100.59265691098834
+I1202 02:27:47.746297 137274321021824 train.py:125] NOTE: Steps:57650/112603 [51.2%]
+Walltime:4d4h37m (0s eval)
+ETA:3d23h53m
+Total train time:8d4h29m
+I1202 02:32:59.528230 137274321021824 utils.py:1231] [57700] l2_params = 299.42235642911885
+I1202 02:32:59.528437 137274321021824 utils.py:1231] [57700] train/loss = 4.650014102458954
+I1202 02:32:59.528533 137274321021824 utils.py:1231] [57700] l2_grads = 1.4743493795394897
+I1202 02:32:59.528600 137274321021824 utils.py:1231] [57700] lr = 0.000555040549958175
+I1202 02:32:59.528666 137274321021824 utils.py:1231] [57700] uptime = 362568.891027455
+I1202 02:32:59.528725 137274321021824 utils.py:1231] [57700] examples_seen = 59084800.0
+I1202 02:32:59.528783 137274321021824 utils.py:1231] [57700] progress = 0.5124197401490191
+I1202 02:32:59.528841 137274321021824 utils.py:1231] [57700] epoch = 46.117953397176166
+I1202 02:32:59.528908 137274321021824 utils.py:1231] [57700] img/sec/core = 164.21689516142578
+I1202 02:32:59.528972 137274321021824 utils.py:1231] [57700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.67926323860304
+I1202 02:32:59.529030 137274321021824 utils.py:1231] [57700] core_hours = 100.67926323860304
+I1202 02:32:59.529096 137274321021824 train.py:125] NOTE: Steps:57700/112603 [51.2%]
+Walltime:4d4h42m (0s eval)
+ETA:3d23h48m
+Total train time:8d4h29m
+I1202 02:38:11.306136 137274321021824 utils.py:1231] [57750] l2_params = 299.3336611732251
+I1202 02:38:11.306410 137274321021824 utils.py:1231] [57750] train/loss = 2.1875221878290176
+I1202 02:38:11.306529 137274321021824 utils.py:1231] [57750] l2_grads = 1.6245304346084595
+I1202 02:38:11.306608 137274321021824 utils.py:1231] [57750] lr = 0.0005542796649284034
+I1202 02:38:11.306667 137274321021824 utils.py:1231] [57750] uptime = 362880.669027631
+I1202 02:38:11.531335 137274321021824 utils.py:1231] [57750] examples_seen = 59136000.0
+I1202 02:38:11.531563 137274321021824 utils.py:1231] [57750] progress = 0.5128637780520945
+I1202 02:38:11.531638 137274321021824 utils.py:1231] [57750] epoch = 46.15791696164513
+I1202 02:38:11.531696 137274321021824 utils.py:1231] [57750] img/sec/core = 164.21941243800254
+I1202 02:38:11.531757 137274321021824 utils.py:1231] [57750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.76586823865195
+I1202 02:38:11.531810 137274321021824 utils.py:1231] [57750] core_hours = 100.76586823865195
+I1202 02:38:11.531888 137274321021824 train.py:125] NOTE: Steps:57750/112603 [51.3%]
+Walltime:4d4h48m (0s eval)
+ETA:3d23h42m
+Total train time:8d4h29m
+I1202 02:43:22.692306 137274321021824 utils.py:1231] [57800] l2_params = 299.2419792111626
+I1202 02:43:22.692500 137274321021824 utils.py:1231] [57800] train/loss = 2.355449289083481
+I1202 02:43:22.692599 137274321021824 utils.py:1231] [57800] l2_grads = 1.7974275350570679
+I1202 02:43:22.692668 137274321021824 utils.py:1231] [57800] lr = 0.0005535186526782457
+I1202 02:43:22.692726 137274321021824 utils.py:1231] [57800] uptime = 363192.055088199
+I1202 02:43:22.692783 137274321021824 utils.py:1231] [57800] examples_seen = 59187200.0
+I1202 02:43:22.692842 137274321021824 utils.py:1231] [57800] progress = 0.5133078159551699
+I1202 02:43:22.692900 137274321021824 utils.py:1231] [57800] epoch = 46.19788052611408
+I1202 02:43:22.692972 137274321021824 utils.py:1231] [57800] img/sec/core = 164.42611434374817
+I1202 02:43:22.693033 137274321021824 utils.py:1231] [57800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.85236436658748
+I1202 02:43:22.693082 137274321021824 utils.py:1231] [57800] core_hours = 100.85236436658748
+I1202 02:43:22.693143 137274321021824 train.py:125] NOTE: Steps:57800/112603 [51.3%]
+Walltime:4d4h53m (0s eval)
+ETA:3d23h37m
+Total train time:8d4h28m
+I1202 02:48:34.478168 137274321021824 utils.py:1231] [57850] l2_params = 299.14589339364454
+I1202 02:48:34.478358 137274321021824 utils.py:1231] [57850] train/loss = 3.118516445159912
+I1202 02:48:34.478455 137274321021824 utils.py:1231] [57850] l2_grads = 1.4332406520843506
+I1202 02:48:34.478513 137274321021824 utils.py:1231] [57850] lr = 0.0005527575149913587
+I1202 02:48:34.478563 137274321021824 utils.py:1231] [57850] uptime = 363503.84092531
+I1202 02:48:34.478615 137274321021824 utils.py:1231] [57850] examples_seen = 59238400.0
+I1202 02:48:34.478673 137274321021824 utils.py:1231] [57850] progress = 0.5137518538582453
+I1202 02:48:34.478732 137274321021824 utils.py:1231] [57850] epoch = 46.23784409058304
+I1202 02:48:34.478780 137274321021824 utils.py:1231] [57850] img/sec/core = 164.21528467879466
+I1202 02:48:34.478833 137274321021824 utils.py:1231] [57850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 100.93897154356277
+I1202 02:48:34.478885 137274321021824 utils.py:1231] [57850] core_hours = 100.93897154356277
+I1202 02:48:34.478945 137274321021824 train.py:125] NOTE: Steps:57850/112603 [51.4%]
+Walltime:4d4h58m (0s eval)
+ETA:3d23h32m
+Total train time:8d4h28m
+I1202 02:53:46.257910 137274321021824 utils.py:1231] [57900] l2_params = 299.0571688316378
+I1202 02:53:46.258151 137274321021824 utils.py:1231] [57900] train/loss = 2.215279847383499
+I1202 02:53:46.258284 137274321021824 utils.py:1231] [57900] l2_grads = 1.5461015701293945
+I1202 02:53:46.258366 137274321021824 utils.py:1231] [57900] lr = 0.0005519962536516926
+I1202 02:53:46.258424 137274321021824 utils.py:1231] [57900] uptime = 363815.62078665197
+I1202 02:53:46.258473 137274321021824 utils.py:1231] [57900] examples_seen = 59289600.0
+I1202 02:53:46.258518 137274321021824 utils.py:1231] [57900] progress = 0.5141958917613207
+I1202 02:53:46.258563 137274321021824 utils.py:1231] [57900] epoch = 46.277807655051994
+I1202 02:53:46.258610 137274321021824 utils.py:1231] [57900] img/sec/core = 164.21843213227743
+I1202 02:53:46.258663 137274321021824 utils.py:1231] [57900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.02557706060222
+I1202 02:53:46.258711 137274321021824 utils.py:1231] [57900] core_hours = 101.02557706060222
+I1202 02:53:46.258766 137274321021824 train.py:125] NOTE: Steps:57900/112603 [51.4%]
+Walltime:4d5h3m (0s eval)
+ETA:3d23h27m
+Total train time:8d4h28m
+I1202 02:58:58.030021 137274321021824 utils.py:1231] [57950] l2_params = 298.96916882142807
+I1202 02:58:58.030243 137274321021824 utils.py:1231] [57950] train/loss = 2.7192652821540833
+I1202 02:58:58.030385 137274321021824 utils.py:1231] [57950] l2_grads = 1.5549812316894531
+I1202 02:58:58.030468 137274321021824 utils.py:1231] [57950] lr = 0.0005512348704434879
+I1202 02:58:58.030521 137274321021824 utils.py:1231] [57950] uptime = 364127.392882921
+I1202 02:58:58.030574 137274321021824 utils.py:1231] [57950] examples_seen = 59340800.0
+I1202 02:58:58.030621 137274321021824 utils.py:1231] [57950] progress = 0.5146399296643962
+I1202 02:58:58.030667 137274321021824 utils.py:1231] [57950] epoch = 46.31777121952095
+I1202 02:58:58.030715 137274321021824 utils.py:1231] [57950] img/sec/core = 164.22252219718516
+I1202 02:58:58.030769 137274321021824 utils.py:1231] [57950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.11218042067694
+I1202 02:58:58.030817 137274321021824 utils.py:1231] [57950] core_hours = 101.11218042067694
+I1202 02:58:58.030877 137274321021824 train.py:125] NOTE: Steps:57950/112603 [51.5%]
+Walltime:4d5h8m (0s eval)
+ETA:3d23h21m
+Total train time:8d4h28m
+I1202 03:04:09.813498 137274321021824 utils.py:1231] [58000] l2_params = 298.8870206646771
+I1202 03:04:09.813715 137274321021824 utils.py:1231] [58000] train/loss = 3.6837131083011627
+I1202 03:04:09.813817 137274321021824 utils.py:1231] [58000] l2_grads = 1.335170865058899
+I1202 03:04:09.813906 137274321021824 utils.py:1231] [58000] lr = 0.0005504733671512701
+I1202 03:04:09.813974 137274321021824 utils.py:1231] [58000] uptime = 364439.176335279
+I1202 03:04:09.814053 137274321021824 utils.py:1231] [58000] examples_seen = 59392000.0
+I1202 03:04:09.814112 137274321021824 utils.py:1231] [58000] progress = 0.5150839675674715
+I1202 03:04:09.814167 137274321021824 utils.py:1231] [58000] epoch = 46.35773478398991
+I1202 03:04:09.814224 137274321021824 utils.py:1231] [58000] img/sec/core = 164.21654072007686
+I1202 03:04:09.814286 137274321021824 utils.py:1231] [58000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.19878693522084
+I1202 03:04:09.814341 137274321021824 utils.py:1231] [58000] core_hours = 101.19878693522084
+I1202 03:04:09.814408 137274321021824 train.py:125] NOTE: Steps:58000/112603 [51.5%]
+Walltime:4d5h13m (0s eval)
+ETA:3d23h16m
+Total train time:8d4h28m
+I1202 03:09:21.684859 137274321021824 utils.py:1231] [58050] l2_params = 298.813927910941
+I1202 03:09:21.685164 137274321021824 utils.py:1231] [58050] train/loss = 2.2885574102401733
+I1202 03:09:21.685403 137274321021824 utils.py:1231] [58050] l2_grads = 1.7386397123336792
+I1202 03:09:21.685547 137274321021824 utils.py:1231] [58050] lr = 0.0005497117455598471
+I1202 03:09:21.685667 137274321021824 utils.py:1231] [58050] uptime = 364751.048019058
+I1202 03:09:21.685783 137274321021824 utils.py:1231] [58050] examples_seen = 59443200.0
+I1202 03:09:21.685902 137274321021824 utils.py:1231] [58050] progress = 0.515528005470547
+I1202 03:09:21.685985 137274321021824 utils.py:1231] [58050] epoch = 46.39769834845887
+I1202 03:09:21.686044 137274321021824 utils.py:1231] [58050] img/sec/core = 164.17008232232433
+I1202 03:09:21.686119 137274321021824 utils.py:1231] [58050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.28541795849279
+I1202 03:09:21.686215 137274321021824 utils.py:1231] [58050] core_hours = 101.28541795849279
+I1202 03:09:21.686295 137274321021824 train.py:125] NOTE: Steps:58050/112603 [51.6%]
+Walltime:4d5h19m (0s eval)
+ETA:3d23h11m
+Total train time:8d4h28m
+I1202 03:14:31.938426 137274321021824 utils.py:1231] [58100] l2_params = 298.722502587392
+I1202 03:14:31.938645 137274321021824 utils.py:1231] [58100] train/loss = 2.2856554090976715
+I1202 03:14:31.938751 137274321021824 utils.py:1231] [58100] l2_grads = 1.672379493713379
+I1202 03:14:31.938840 137274321021824 utils.py:1231] [58100] lr = 0.0005489500074543025
+I1202 03:14:31.938937 137274321021824 utils.py:1231] [58100] uptime = 365061.301295264
+I1202 03:14:31.939005 137274321021824 utils.py:1231] [58100] examples_seen = 59494400.0
+I1202 03:14:31.939064 137274321021824 utils.py:1231] [58100] progress = 0.5159720433736223
+I1202 03:14:31.939119 137274321021824 utils.py:1231] [58100] epoch = 46.43766191292782
+I1202 03:14:31.939175 137274321021824 utils.py:1231] [58100] img/sec/core = 165.02646040072548
+I1202 03:14:31.939237 137274321021824 utils.py:1231] [58100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.37159942410555
+I1202 03:14:31.939297 137274321021824 utils.py:1231] [58100] core_hours = 101.37159942410555
+I1202 03:14:31.939380 137274321021824 train.py:125] NOTE: Steps:58100/112603 [51.6%]
+Walltime:4d5h24m (0s eval)
+ETA:3d23h5m
+Total train time:8d4h28m
+I1202 03:19:43.705076 137274321021824 utils.py:1231] [58150] l2_params = 298.6385319702845
+I1202 03:19:43.705275 137274321021824 utils.py:1231] [58150] train/loss = 2.3351165652275085
+I1202 03:19:43.705380 137274321021824 utils.py:1231] [58150] l2_grads = 1.7349101305007935
+I1202 03:19:43.705448 137274321021824 utils.py:1231] [58150] lr = 0.000548188154619994
+I1202 03:19:43.705506 137274321021824 utils.py:1231] [58150] uptime = 365373.067867723
+I1202 03:19:43.705561 137274321021824 utils.py:1231] [58150] examples_seen = 59545600.0
+I1202 03:19:43.927902 137274321021824 utils.py:1231] [58150] progress = 0.5164160812766978
+I1202 03:19:43.928183 137274321021824 utils.py:1231] [58150] epoch = 46.47762547739678
+I1202 03:19:43.928297 137274321021824 utils.py:1231] [58150] img/sec/core = 164.2254318548861
+I1202 03:19:43.928396 137274321021824 utils.py:1231] [58150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.4582012497886
+I1202 03:19:43.928458 137274321021824 utils.py:1231] [58150] core_hours = 101.4582012497886
+I1202 03:19:43.928527 137274321021824 train.py:125] NOTE: Steps:58150/112603 [51.6%]
+Walltime:4d5h29m (0s eval)
+ETA:3d23h0m
+Total train time:8d4h28m
+I1202 03:24:55.603701 137274321021824 utils.py:1231] [58200] l2_params = 298.5490744976866
+I1202 03:24:55.603904 137274321021824 utils.py:1231] [58200] train/loss = 2.3738392293453217
+I1202 03:24:55.604005 137274321021824 utils.py:1231] [58200] l2_grads = 1.6300466060638428
+I1202 03:24:55.604069 137274321021824 utils.py:1231] [58200] lr = 0.0005474261888425485
+I1202 03:24:55.604165 137274321021824 utils.py:1231] [58200] uptime = 365684.966522035
+I1202 03:24:55.604230 137274321021824 utils.py:1231] [58200] examples_seen = 59596800.0
+I1202 03:24:55.604280 137274321021824 utils.py:1231] [58200] progress = 0.5168601191797731
+I1202 03:24:55.604328 137274321021824 utils.py:1231] [58200] epoch = 46.51758904186573
+I1202 03:24:55.604379 137274321021824 utils.py:1231] [58200] img/sec/core = 164.155886189818
+I1202 03:24:55.604434 137274321021824 utils.py:1231] [58200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.54483976487528
+I1202 03:24:55.604483 137274321021824 utils.py:1231] [58200] core_hours = 101.54483976487528
+I1202 03:24:55.604542 137274321021824 train.py:125] NOTE: Steps:58200/112603 [51.7%]
+Walltime:4d5h34m (0s eval)
+ETA:3d22h55m
+Total train time:8d4h28m
+I1202 03:30:07.354415 137274321021824 utils.py:1231] [58250] l2_params = 298.4610737227953
+I1202 03:30:07.354637 137274321021824 utils.py:1231] [58250] train/loss = 2.375891298055649
+I1202 03:30:07.354745 137274321021824 utils.py:1231] [58250] l2_grads = 1.549149513244629
+I1202 03:30:07.354820 137274321021824 utils.py:1231] [58250] lr = 0.0005466641119078575
+I1202 03:30:07.354898 137274321021824 utils.py:1231] [58250] uptime = 365996.717259538
+I1202 03:30:07.354967 137274321021824 utils.py:1231] [58250] examples_seen = 59648000.0
+I1202 03:30:07.355027 137274321021824 utils.py:1231] [58250] progress = 0.5173041570828486
+I1202 03:30:07.355085 137274321021824 utils.py:1231] [58250] epoch = 46.557552606334696
+I1202 03:30:07.355142 137274321021824 utils.py:1231] [58250] img/sec/core = 164.23377346303118
+I1202 03:30:07.355205 137274321021824 utils.py:1231] [58250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.63143719195945
+I1202 03:30:07.355261 137274321021824 utils.py:1231] [58250] core_hours = 101.63143719195945
+I1202 03:30:07.355332 137274321021824 train.py:125] NOTE: Steps:58250/112603 [51.7%]
+Walltime:4d5h39m (0s eval)
+ETA:3d22h50m
+Total train time:8d4h28m
+I1202 03:35:19.120974 137274321021824 utils.py:1231] [58300] l2_params = 298.3722850591543
+I1202 03:35:19.121211 137274321021824 utils.py:1231] [58300] train/loss = 2.1259716153144836
+I1202 03:35:19.121318 137274321021824 utils.py:1231] [58300] l2_grads = 1.5578289031982422
+I1202 03:35:19.121378 137274321021824 utils.py:1231] [58300] lr = 0.000545901925602073
+I1202 03:35:19.121431 137274321021824 utils.py:1231] [58300] uptime = 366308.483793083
+I1202 03:35:19.121486 137274321021824 utils.py:1231] [58300] examples_seen = 59699200.0
+I1202 03:35:19.121542 137274321021824 utils.py:1231] [58300] progress = 0.517748194985924
+I1202 03:35:19.121592 137274321021824 utils.py:1231] [58300] epoch = 46.59751617080365
+I1202 03:35:19.121644 137274321021824 utils.py:1231] [58300] img/sec/core = 164.22545235315394
+I1202 03:35:19.121711 137274321021824 utils.py:1231] [58300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.71803900683305
+I1202 03:35:19.121767 137274321021824 utils.py:1231] [58300] core_hours = 101.71803900683305
+I1202 03:35:19.121828 137274321021824 train.py:125] NOTE: Steps:58300/112603 [51.8%]
+Walltime:4d5h45m (0s eval)
+ETA:3d22h44m
+Total train time:8d4h28m
+I1202 03:40:30.889657 137274321021824 utils.py:1231] [58350] l2_params = 298.28251218833464
+I1202 03:40:30.889953 137274321021824 utils.py:1231] [58350] train/loss = 4.575597941875458
+I1202 03:40:30.890226 137274321021824 utils.py:1231] [58350] l2_grads = 1.5511474609375
+I1202 03:40:30.890322 137274321021824 utils.py:1231] [58350] lr = 0.0005451396317116024
+I1202 03:40:30.890396 137274321021824 utils.py:1231] [58350] uptime = 366620.252753394
+I1202 03:40:30.890466 137274321021824 utils.py:1231] [58350] examples_seen = 59750400.0
+I1202 03:40:30.890537 137274321021824 utils.py:1231] [58350] progress = 0.5181922328889994
+I1202 03:40:30.890598 137274321021824 utils.py:1231] [58350] epoch = 46.637479735272606
+I1202 03:40:30.890653 137274321021824 utils.py:1231] [58350] img/sec/core = 164.2241740451747
+I1202 03:40:30.890713 137274321021824 utils.py:1231] [58350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.80464149580833
+I1202 03:40:30.890773 137274321021824 utils.py:1231] [58350] core_hours = 101.80464149580833
+I1202 03:40:30.890836 137274321021824 train.py:125] NOTE: Steps:58350/112603 [51.8%]
+Walltime:4d5h50m (0s eval)
+ETA:3d22h39m
+Total train time:8d4h28m
+I1202 03:45:42.652280 137274321021824 utils.py:1231] [58400] l2_params = 298.19313195554247
+I1202 03:45:42.652527 137274321021824 utils.py:1231] [58400] train/loss = 2.698868840932846
+I1202 03:45:42.652647 137274321021824 utils.py:1231] [58400] l2_grads = 1.5859339237213135
+I1202 03:45:42.652735 137274321021824 utils.py:1231] [58400] lr = 0.0005443772320231064
+I1202 03:45:42.652858 137274321021824 utils.py:1231] [58400] uptime = 366932.015214098
+I1202 03:45:42.652933 137274321021824 utils.py:1231] [58400] examples_seen = 59801600.0
+I1202 03:45:42.652991 137274321021824 utils.py:1231] [58400] progress = 0.5186362707920749
+I1202 03:45:42.653044 137274321021824 utils.py:1231] [58400] epoch = 46.67744329974156
+I1202 03:45:42.653099 137274321021824 utils.py:1231] [58400] img/sec/core = 164.22759778194617
+I1202 03:45:42.653159 137274321021824 utils.py:1231] [58400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.89124217933721
+I1202 03:45:42.653213 137274321021824 utils.py:1231] [58400] core_hours = 101.89124217933721
+I1202 03:45:42.653276 137274321021824 train.py:125] NOTE: Steps:58400/112603 [51.9%]
+Walltime:4d5h55m (0s eval)
+ETA:3d22h34m
+Total train time:8d4h27m
+I1202 03:50:54.410873 137274321021824 utils.py:1231] [58450] l2_params = 298.1173441225774
+I1202 03:50:54.411139 137274321021824 utils.py:1231] [58450] train/loss = 3.6576043367385864
+I1202 03:50:54.411278 137274321021824 utils.py:1231] [58450] l2_grads = 1.4783575534820557
+I1202 03:50:54.411350 137274321021824 utils.py:1231] [58450] lr = 0.0005436147283234936
+I1202 03:50:54.411409 137274321021824 utils.py:1231] [58450] uptime = 367243.773766279
+I1202 03:50:54.411465 137274321021824 utils.py:1231] [58450] examples_seen = 59852800.0
+I1202 03:50:54.411518 137274321021824 utils.py:1231] [58450] progress = 0.5190803086951502
+I1202 03:50:54.411569 137274321021824 utils.py:1231] [58450] epoch = 46.717406864210524
+I1202 03:50:54.411624 137274321021824 utils.py:1231] [58450] img/sec/core = 164.22965670649089
+I1202 03:50:54.411681 137274321021824 utils.py:1231] [58450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 101.97784177716527
+I1202 03:50:54.411744 137274321021824 utils.py:1231] [58450] core_hours = 101.97784177716527
+I1202 03:50:54.411815 137274321021824 train.py:125] NOTE: Steps:58450/112603 [51.9%]
+Walltime:4d6h0m (0s eval)
+ETA:3d22h29m
+Total train time:8d4h27m
+I1202 03:56:06.191081 137274321021824 utils.py:1231] [58500] l2_params = 298.038594958681
+I1202 03:56:06.191368 137274321021824 utils.py:1231] [58500] train/loss = 2.418010711669922
+I1202 03:56:06.191486 137274321021824 utils.py:1231] [58500] l2_grads = 1.701897144317627
+I1202 03:56:06.191564 137274321021824 utils.py:1231] [58500] lr = 0.0005428521223999158
+I1202 03:56:06.191627 137274321021824 utils.py:1231] [58500] uptime = 367555.553988172
+I1202 03:56:06.191682 137274321021824 utils.py:1231] [58500] examples_seen = 59904000.0
+I1202 03:56:06.191732 137274321021824 utils.py:1231] [58500] progress = 0.5195243465982257
+I1202 03:56:06.191781 137274321021824 utils.py:1231] [58500] epoch = 46.75737042867948
+I1202 03:56:06.191834 137274321021824 utils.py:1231] [58500] img/sec/core = 164.21824222568415
+I1202 03:56:06.191913 137274321021824 utils.py:1231] [58500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.06444739435777
+I1202 03:56:06.191980 137274321021824 utils.py:1231] [58500] core_hours = 102.06444739435777
+I1202 03:56:06.192051 137274321021824 train.py:125] NOTE: Steps:58500/112603 [52.0%]
+Walltime:4d6h5m (0s eval)
+ETA:3d22h23m
+Total train time:8d4h27m
+I1202 04:01:17.974235 137274321021824 utils.py:1231] [58550] l2_params = 297.9540400896359
+I1202 04:01:18.201580 137274321021824 utils.py:1231] [58550] train/loss = 2.1907251477241516
+I1202 04:01:18.202010 137274321021824 utils.py:1231] [58550] l2_grads = 1.6701635122299194
+I1202 04:01:18.202157 137274321021824 utils.py:1231] [58550] lr = 0.000542089416039764
+I1202 04:01:18.202252 137274321021824 utils.py:1231] [58550] uptime = 367867.564610696
+I1202 04:01:18.202313 137274321021824 utils.py:1231] [58550] examples_seen = 59955200.0
+I1202 04:01:18.202373 137274321021824 utils.py:1231] [58550] progress = 0.519968384501301
+I1202 04:01:18.202439 137274321021824 utils.py:1231] [58550] epoch = 46.797333993148435
+I1202 04:01:18.202512 137274321021824 utils.py:1231] [58550] img/sec/core = 164.09697716640733
+I1202 04:01:18.202575 137274321021824 utils.py:1231] [58550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.15111701172555
+I1202 04:01:18.202641 137274321021824 utils.py:1231] [58550] core_hours = 102.15111701172555
+I1202 04:01:18.202741 137274321021824 train.py:125] NOTE: Steps:58550/112603 [52.0%]
+Walltime:4d6h11m (0s eval)
+ETA:3d22h18m
+Total train time:8d4h27m
+I1202 04:06:29.975599 137274321021824 utils.py:1231] [58600] l2_params = 297.85299927011624
+I1202 04:06:29.975793 137274321021824 utils.py:1231] [58600] train/loss = 2.1758870482444763
+I1202 04:06:29.975906 137274321021824 utils.py:1231] [58600] l2_grads = 1.595837116241455
+I1202 04:06:29.975980 137274321021824 utils.py:1231] [58600] lr = 0.0005413266110306665
+I1202 04:06:29.976041 137274321021824 utils.py:1231] [58600] uptime = 368179.338402791
+I1202 04:06:29.976102 137274321021824 utils.py:1231] [58600] examples_seen = 60006400.0
+I1202 04:06:29.976163 137274321021824 utils.py:1231] [58600] progress = 0.5204124224043765
+I1202 04:06:29.976219 137274321021824 utils.py:1231] [58600] epoch = 46.83729755761739
+I1202 04:06:29.976275 137274321021824 utils.py:1231] [58600] img/sec/core = 164.2216289443476
+I1202 04:06:29.976338 137274321021824 utils.py:1231] [58600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.23772084286306
+I1202 04:06:29.976398 137274321021824 utils.py:1231] [58600] core_hours = 102.23772084286306
+I1202 04:06:29.976466 137274321021824 train.py:125] NOTE: Steps:58600/112603 [52.0%]
+Walltime:4d6h16m (0s eval)
+ETA:3d22h13m
+Total train time:8d4h27m
+I1202 04:11:41.752159 137274321021824 utils.py:1231] [58650] l2_params = 297.7651183573485
+I1202 04:11:41.752370 137274321021824 utils.py:1231] [58650] train/loss = 2.2664433121681213
+I1202 04:11:41.752468 137274321021824 utils.py:1231] [58650] l2_grads = 1.6612424850463867
+I1202 04:11:41.752554 137274321021824 utils.py:1231] [58650] lr = 0.0005405637091604805
+I1202 04:11:41.752620 137274321021824 utils.py:1231] [58650] uptime = 368491.114981755
+I1202 04:11:41.752681 137274321021824 utils.py:1231] [58650] examples_seen = 60057600.0
+I1202 04:11:41.752743 137274321021824 utils.py:1231] [58650] progress = 0.5208564603074518
+I1202 04:11:41.752808 137274321021824 utils.py:1231] [58650] epoch = 46.877261122086345
+I1202 04:11:41.752864 137274321021824 utils.py:1231] [58650] img/sec/core = 164.22016102087431
+I1202 04:11:41.752928 137274321021824 utils.py:1231] [58650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.32432544813082
+I1202 04:11:41.752982 137274321021824 utils.py:1231] [58650] core_hours = 102.32432544813082
+I1202 04:11:41.753041 137274321021824 train.py:125] NOTE: Steps:58650/112603 [52.1%]
+Walltime:4d6h21m (0s eval)
+ETA:3d22h7m
+Total train time:8d4h27m
+I1202 04:16:53.541745 137274321021824 utils.py:1231] [58700] l2_params = 297.67490780753985
+I1202 04:16:53.542003 137274321021824 utils.py:1231] [58700] train/loss = 4.394439101219177
+I1202 04:16:53.542137 137274321021824 utils.py:1231] [58700] l2_grads = 1.4297858476638794
+I1202 04:16:53.542231 137274321021824 utils.py:1231] [58700] lr = 0.0005398007122172922
+I1202 04:16:53.542310 137274321021824 utils.py:1231] [58700] uptime = 368802.904671494
+I1202 04:16:53.542381 137274321021824 utils.py:1231] [58700] examples_seen = 60108800.0
+I1202 04:16:53.542442 137274321021824 utils.py:1231] [58700] progress = 0.5213004982105273
+I1202 04:16:53.542503 137274321021824 utils.py:1231] [58700] epoch = 46.91722468655531
+I1202 04:16:53.542572 137274321021824 utils.py:1231] [58700] img/sec/core = 164.21325555331654
+I1202 04:16:53.542649 137274321021824 utils.py:1231] [58700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.41093369528055
+I1202 04:16:53.542709 137274321021824 utils.py:1231] [58700] core_hours = 102.41093369528055
+I1202 04:16:53.542783 137274321021824 train.py:125] NOTE: Steps:58700/112603 [52.1%]
+Walltime:4d6h26m (0s eval)
+ETA:3d22h2m
+Total train time:8d4h27m
+I1202 04:22:05.332088 137274321021824 utils.py:1231] [58750] l2_params = 297.5944717359018
+I1202 04:22:05.332343 137274321021824 utils.py:1231] [58750] train/loss = 2.3544618487358093
+I1202 04:22:05.332462 137274321021824 utils.py:1231] [58750] l2_grads = 1.6365289688110352
+I1202 04:22:05.332538 137274321021824 utils.py:1231] [58750] lr = 0.0005390376219894087
+I1202 04:22:05.332598 137274321021824 utils.py:1231] [58750] uptime = 369114.694960365
+I1202 04:22:05.332652 137274321021824 utils.py:1231] [58750] examples_seen = 60160000.0
+I1202 04:22:05.332718 137274321021824 utils.py:1231] [58750] progress = 0.5217445361136026
+I1202 04:22:05.332791 137274321021824 utils.py:1231] [58750] epoch = 46.95718825102426
+I1202 04:22:05.332850 137274321021824 utils.py:1231] [58750] img/sec/core = 164.2129400033279
+I1202 04:22:05.332910 137274321021824 utils.py:1231] [58750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.49754210885584
+I1202 04:22:05.332962 137274321021824 utils.py:1231] [58750] core_hours = 102.49754210885584
+I1202 04:22:05.333022 137274321021824 train.py:125] NOTE: Steps:58750/112603 [52.2%]
+Walltime:4d6h31m (0s eval)
+ETA:3d21h57m
+Total train time:8d4h27m
+I1202 04:27:17.097743 137274321021824 utils.py:1231] [58800] l2_params = 297.4982240178087
+I1202 04:27:17.097989 137274321021824 utils.py:1231] [58800] train/loss = 2.394764930009842
+I1202 04:27:17.098124 137274321021824 utils.py:1231] [58800] l2_grads = 1.6334396600723267
+I1202 04:27:17.098200 137274321021824 utils.py:1231] [58800] lr = 0.0005382744402653575
+I1202 04:27:17.098256 137274321021824 utils.py:1231] [58800] uptime = 369426.460617737
+I1202 04:27:17.098315 137274321021824 utils.py:1231] [58800] examples_seen = 60211200.0
+I1202 04:27:17.098371 137274321021824 utils.py:1231] [58800] progress = 0.5221885740166781
+I1202 04:27:17.098427 137274321021824 utils.py:1231] [58800] epoch = 46.99715181549322
+I1202 04:27:17.098484 137274321021824 utils.py:1231] [58800] img/sec/core = 164.22591388542483
+I1202 04:27:17.098551 137274321021824 utils.py:1231] [58800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.58414368034805
+I1202 04:27:17.098614 137274321021824 utils.py:1231] [58800] core_hours = 102.58414368034805
+I1202 04:27:17.098688 137274321021824 train.py:125] NOTE: Steps:58800/112603 [52.2%]
+Walltime:4d6h37m (0s eval)
+ETA:3d21h52m
+Total train time:8d4h27m
+I1202 04:32:28.894885 137274321021824 utils.py:1231] [58850] l2_params = 297.41900758495854
+I1202 04:32:28.895100 137274321021824 utils.py:1231] [58850] train/loss = 3.021057039499283
+I1202 04:32:28.895204 137274321021824 utils.py:1231] [58850] l2_grads = 1.4699126482009888
+I1202 04:32:28.895269 137274321021824 utils.py:1231] [58850] lr = 0.0005375111688338793
+I1202 04:32:28.895325 137274321021824 utils.py:1231] [58850] uptime = 369738.25768653397
+I1202 04:32:28.895378 137274321021824 utils.py:1231] [58850] examples_seen = 60262400.0
+I1202 04:32:28.895432 137274321021824 utils.py:1231] [58850] progress = 0.5226326119197535
+I1202 04:32:28.895483 137274321021824 utils.py:1231] [58850] epoch = 47.037115379962174
+I1202 04:32:28.895538 137274321021824 utils.py:1231] [58850] img/sec/core = 164.20936924630146
+I1202 04:32:28.895598 137274321021824 utils.py:1231] [58850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.67075397723609
+I1202 04:32:28.895653 137274321021824 utils.py:1231] [58850] core_hours = 102.67075397723609
+I1202 04:32:28.895716 137274321021824 train.py:125] NOTE: Steps:58850/112603 [52.3%]
+Walltime:4d6h42m (0s eval)
+ETA:3d21h46m
+Total train time:8d4h27m
+I1202 04:37:40.696093 137274321021824 utils.py:1231] [58900] l2_params = 297.32572437790486
+I1202 04:37:40.696319 137274321021824 utils.py:1231] [58900] train/loss = 4.686167478561401
+I1202 04:37:40.696458 137274321021824 utils.py:1231] [58900] l2_grads = 1.5186353921890259
+I1202 04:37:40.696566 137274321021824 utils.py:1231] [58900] lr = 0.0005367478094839259
+I1202 04:37:40.696651 137274321021824 utils.py:1231] [58900] uptime = 370050.059007887
+I1202 04:37:40.696735 137274321021824 utils.py:1231] [58900] examples_seen = 60313600.0
+I1202 04:37:40.696816 137274321021824 utils.py:1231] [58900] progress = 0.5230766498228289
+I1202 04:37:40.696939 137274321021824 utils.py:1231] [58900] epoch = 47.07707894443113
+I1202 04:37:40.697014 137274321021824 utils.py:1231] [58900] img/sec/core = 164.20712964854243
+I1202 04:37:40.697088 137274321021824 utils.py:1231] [58900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.7573654553897
+I1202 04:37:40.697152 137274321021824 utils.py:1231] [58900] core_hours = 102.7573654553897
+I1202 04:37:40.697225 137274321021824 train.py:125] NOTE: Steps:58900/112603 [52.3%]
+Walltime:4d6h47m (0s eval)
+ETA:3d21h41m
+Total train time:8d4h27m
+I1202 04:42:52.478796 137274321021824 utils.py:1231] [58950] l2_params = 297.22533869226316
+I1202 04:42:52.479036 137274321021824 utils.py:1231] [58950] train/loss = 2.2061559855937958
+I1202 04:42:52.479159 137274321021824 utils.py:1231] [58950] l2_grads = 1.5822182893753052
+I1202 04:42:52.479246 137274321021824 utils.py:1231] [58950] lr = 0.0005359843640046549
+I1202 04:42:52.479315 137274321021824 utils.py:1231] [58950] uptime = 370361.84167616
+I1202 04:42:52.479387 137274321021824 utils.py:1231] [58950] examples_seen = 60364800.0
+I1202 04:42:52.479459 137274321021824 utils.py:1231] [58950] progress = 0.5235206877259043
+I1202 04:42:52.479521 137274321021824 utils.py:1231] [58950] epoch = 47.11704250890009
+I1202 04:42:52.479583 137274321021824 utils.py:1231] [58950] img/sec/core = 164.2169536991871
+I1202 04:42:52.479667 137274321021824 utils.py:1231] [58950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.8439717521322
+I1202 04:42:52.479737 137274321021824 utils.py:1231] [58950] core_hours = 102.8439717521322
+I1202 04:42:52.479832 137274321021824 train.py:125] NOTE: Steps:58950/112603 [52.4%]
+Walltime:4d6h52m (0s eval)
+ETA:3d21h36m
+Total train time:8d4h27m
+I1202 04:48:04.259658 137274321021824 utils.py:1231] [59000] l2_params = 297.143190697596
+I1202 04:48:04.259893 137274321021824 utils.py:1231] [59000] train/loss = 3.366920530796051
+I1202 04:48:04.260009 137274321021824 utils.py:1231] [59000] l2_grads = 1.383514165878296
+I1202 04:48:04.260083 137274321021824 utils.py:1231] [59000] lr = 0.0005352208341854257
+I1202 04:48:04.260148 137274321021824 utils.py:1231] [59000] uptime = 370673.622505936
+I1202 04:48:04.260216 137274321021824 utils.py:1231] [59000] examples_seen = 60416000.0
+I1202 04:48:04.260272 137274321021824 utils.py:1231] [59000] progress = 0.5239647256289797
+I1202 04:48:04.260325 137274321021824 utils.py:1231] [59000] epoch = 47.15700607336905
+I1202 04:48:04.260386 137274321021824 utils.py:1231] [59000] img/sec/core = 164.21792204729678
+I1202 04:48:04.260450 137274321021824 utils.py:1231] [59000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 102.93057753818111
+I1202 04:48:04.260502 137274321021824 utils.py:1231] [59000] core_hours = 102.93057753818111
+I1202 04:48:04.260564 137274321021824 train.py:125] NOTE: Steps:59000/112603 [52.4%]
+Walltime:4d6h57m (0s eval)
+ETA:3d21h31m
+Total train time:8d4h27m
+I1202 04:53:16.398890 137274321021824 utils.py:1231] [59050] l2_params = 297.07236416130183
+I1202 04:53:16.399133 137274321021824 utils.py:1231] [59050] train/loss = 3.4589312374591827
+I1202 04:53:16.399258 137274321021824 utils.py:1231] [59050] l2_grads = 1.4880445003509521
+I1202 04:53:16.399334 137274321021824 utils.py:1231] [59050] lr = 0.0005344572218157951
+I1202 04:53:16.399396 137274321021824 utils.py:1231] [59050] uptime = 370985.761757681
+I1202 04:53:16.399449 137274321021824 utils.py:1231] [59050] examples_seen = 60467200.0
+I1202 04:53:16.399505 137274321021824 utils.py:1231] [59050] progress = 0.5244087635320551
+I1202 04:53:16.399554 137274321021824 utils.py:1231] [59050] epoch = 47.196969637838
+I1202 04:53:16.399606 137274321021824 utils.py:1231] [59050] img/sec/core = 164.02935457099957
+I1202 04:53:16.399661 137274321021824 utils.py:1231] [59050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.01728288588805
+I1202 04:53:16.399716 137274321021824 utils.py:1231] [59050] core_hours = 103.01728288588805
+I1202 04:53:16.399786 137274321021824 train.py:125] NOTE: Steps:59050/112603 [52.4%]
+Walltime:4d7h3m (0s eval)
+ETA:3d21h25m
+Total train time:8d4h27m
+I1202 04:58:28.182544 137274321021824 utils.py:1231] [59100] l2_params = 296.97101555453537
+I1202 04:58:28.182796 137274321021824 utils.py:1231] [59100] train/loss = 2.154524102807045
+I1202 04:58:28.182896 137274321021824 utils.py:1231] [59100] l2_grads = 1.628462553024292
+I1202 04:58:28.182961 137274321021824 utils.py:1231] [59100] lr = 0.0005336935286855137
+I1202 04:58:28.183015 137274321021824 utils.py:1231] [59100] uptime = 371297.545376893
+I1202 04:58:28.183069 137274321021824 utils.py:1231] [59100] examples_seen = 60518400.0
+I1202 04:58:28.183119 137274321021824 utils.py:1231] [59100] progress = 0.5248528014351305
+I1202 04:58:28.183171 137274321021824 utils.py:1231] [59100] epoch = 47.23693320230696
+I1202 04:58:28.183224 137274321021824 utils.py:1231] [59100] img/sec/core = 164.21645283801374
+I1202 04:58:28.183282 137274321021824 utils.py:1231] [59100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.10388944678029
+I1202 04:58:28.183334 137274321021824 utils.py:1231] [59100] core_hours = 103.10388944678029
+I1202 04:58:28.183399 137274321021824 train.py:125] NOTE: Steps:59100/112603 [52.5%]
+Walltime:4d7h8m (0s eval)
+ETA:3d21h20m
+Total train time:8d4h27m
+I1202 05:03:39.953408 137274321021824 utils.py:1231] [59150] l2_params = 296.9086863697146
+I1202 05:03:39.953662 137274321021824 utils.py:1231] [59150] train/loss = 2.858738988637924
+I1202 05:03:39.953793 137274321021824 utils.py:1231] [59150] l2_grads = 1.4673264026641846
+I1202 05:03:39.953889 137274321021824 utils.py:1231] [59150] lr = 0.0005329297565845217
+I1202 05:03:39.953976 137274321021824 utils.py:1231] [59150] uptime = 371609.316332836
+I1202 05:03:39.954049 137274321021824 utils.py:1231] [59150] examples_seen = 60569600.0
+I1202 05:03:39.954104 137274321021824 utils.py:1231] [59150] progress = 0.5252968393382059
+I1202 05:03:39.954159 137274321021824 utils.py:1231] [59150] epoch = 47.27689676677591
+I1202 05:03:39.954218 137274321021824 utils.py:1231] [59150] img/sec/core = 164.22312285357404
+I1202 05:03:39.954279 137274321021824 utils.py:1231] [59150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.19049249009777
+I1202 05:03:39.954334 137274321021824 utils.py:1231] [59150] core_hours = 103.19049249009777
+I1202 05:03:39.954396 137274321021824 train.py:125] NOTE: Steps:59150/112603 [52.5%]
+Walltime:4d7h13m (0s eval)
+ETA:3d21h15m
+Total train time:8d4h26m
+I1202 05:08:51.719044 137274321021824 utils.py:1231] [59200] l2_params = 296.8232956065856
+I1202 05:08:51.719299 137274321021824 utils.py:1231] [59200] train/loss = 2.7674224972724915
+I1202 05:08:51.719431 137274321021824 utils.py:1231] [59200] l2_grads = 1.4440102577209473
+I1202 05:08:51.719503 137274321021824 utils.py:1231] [59200] lr = 0.0005321659073029437
+I1202 05:08:51.719572 137274321021824 utils.py:1231] [59200] uptime = 371921.081933824
+I1202 05:08:51.719642 137274321021824 utils.py:1231] [59200] examples_seen = 60620800.0
+I1202 05:08:51.719705 137274321021824 utils.py:1231] [59200] progress = 0.5257408772412813
+I1202 05:08:51.719764 137274321021824 utils.py:1231] [59200] epoch = 47.316860331244875
+I1202 05:08:51.719826 137274321021824 utils.py:1231] [59200] img/sec/core = 164.22594358628595
+I1202 05:08:51.719902 137274321021824 utils.py:1231] [59200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.27709404592777
+I1202 05:08:51.719959 137274321021824 utils.py:1231] [59200] core_hours = 103.27709404592777
+I1202 05:08:51.720024 137274321021824 train.py:125] NOTE: Steps:59200/112603 [52.6%]
+Walltime:4d7h18m (0s eval)
+ETA:3d21h10m
+Total train time:8d4h26m
+I1202 05:14:03.480106 137274321021824 utils.py:1231] [59250] l2_params = 296.7415113935267
+I1202 05:14:03.480361 137274321021824 utils.py:1231] [59250] train/loss = 2.1655569076538086
+I1202 05:14:03.480537 137274321021824 utils.py:1231] [59250] l2_grads = 1.642745852470398
+I1202 05:14:03.480643 137274321021824 utils.py:1231] [59250] lr = 0.0005314019826310855
+I1202 05:14:03.480729 137274321021824 utils.py:1231] [59250] uptime = 372232.843086737
+I1202 05:14:03.480819 137274321021824 utils.py:1231] [59250] examples_seen = 60672000.0
+I1202 05:14:03.480906 137274321021824 utils.py:1231] [59250] progress = 0.5261849151443567
+I1202 05:14:03.480985 137274321021824 utils.py:1231] [59250] epoch = 47.35682389571383
+I1202 05:14:03.481067 137274321021824 utils.py:1231] [59250] img/sec/core = 164.2282866919321
+I1202 05:14:03.481156 137274321021824 utils.py:1231] [59250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.36369436618138
+I1202 05:14:03.481221 137274321021824 utils.py:1231] [59250] core_hours = 103.36369436618138
+I1202 05:14:03.481288 137274321021824 train.py:125] NOTE: Steps:59250/112603 [52.6%]
+Walltime:4d7h23m (0s eval)
+ETA:3d21h4m
+Total train time:8d4h26m
+I1202 05:19:15.202736 137274321021824 utils.py:1231] [59300] l2_params = 296.6377582718955
+I1202 05:19:15.202939 137274321021824 utils.py:1231] [59300] train/loss = 4.679825127124786
+I1202 05:19:15.203036 137274321021824 utils.py:1231] [59300] l2_grads = 1.7061196565628052
+I1202 05:19:15.203096 137274321021824 utils.py:1231] [59300] lr = 0.00053063798435943
+I1202 05:19:15.203148 137274321021824 utils.py:1231] [59300] uptime = 372544.565509359
+I1202 05:19:15.203202 137274321021824 utils.py:1231] [59300] examples_seen = 60723200.0
+I1202 05:19:15.203250 137274321021824 utils.py:1231] [59300] progress = 0.5266289530474322
+I1202 05:19:15.203298 137274321021824 utils.py:1231] [59300] epoch = 47.396787460182786
+I1202 05:19:15.203348 137274321021824 utils.py:1231] [59300] img/sec/core = 164.24869141377565
+I1202 05:19:15.203404 137274321021824 utils.py:1231] [59300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.45028392802082
+I1202 05:19:15.203458 137274321021824 utils.py:1231] [59300] core_hours = 103.45028392802082
+I1202 05:19:15.203517 137274321021824 train.py:125] NOTE: Steps:59300/112603 [52.7%]
+Walltime:4d7h29m (0s eval)
+ETA:3d20h59m
+Total train time:8d4h26m
+I1202 05:24:26.978364 137274321021824 utils.py:1231] [59350] l2_params = 296.54063602442784
+I1202 05:24:26.978577 137274321021824 utils.py:1231] [59350] train/loss = 4.241087079048157
+I1202 05:24:26.978667 137274321021824 utils.py:1231] [59350] l2_grads = 1.4580037593841553
+I1202 05:24:26.978723 137274321021824 utils.py:1231] [59350] lr = 0.0005298739142786326
+I1202 05:24:26.978772 137274321021824 utils.py:1231] [59350] uptime = 372856.341134874
+I1202 05:24:26.978821 137274321021824 utils.py:1231] [59350] examples_seen = 60774400.0
+I1202 05:24:26.978868 137274321021824 utils.py:1231] [59350] progress = 0.5270729909505075
+I1202 05:24:26.978917 137274321021824 utils.py:1231] [59350] epoch = 47.43675102465174
+I1202 05:24:26.978964 137274321021824 utils.py:1231] [59350] img/sec/core = 164.2206632267384
+I1202 05:24:26.979015 137274321021824 utils.py:1231] [59350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.53688826844164
+I1202 05:24:26.979061 137274321021824 utils.py:1231] [59350] core_hours = 103.53688826844164
+I1202 05:24:26.979117 137274321021824 train.py:125] NOTE: Steps:59350/112603 [52.7%]
+Walltime:4d7h34m (0s eval)
+ETA:3d20h54m
+Total train time:8d4h26m
+I1202 05:29:38.762361 137274321021824 utils.py:1231] [59400] l2_params = 296.46112819831063
+I1202 05:29:38.762565 137274321021824 utils.py:1231] [59400] train/loss = 2.380006730556488
+I1202 05:29:38.762673 137274321021824 utils.py:1231] [59400] l2_grads = 1.6842610836029053
+I1202 05:29:38.762741 137274321021824 utils.py:1231] [59400] lr = 0.0005291097741795154
+I1202 05:29:38.762799 137274321021824 utils.py:1231] [59400] uptime = 373168.125160434
+I1202 05:29:38.762873 137274321021824 utils.py:1231] [59400] examples_seen = 60825600.0
+I1202 05:29:38.762934 137274321021824 utils.py:1231] [59400] progress = 0.527517028853583
+I1202 05:29:38.762988 137274321021824 utils.py:1231] [59400] epoch = 47.476714589120704
+I1202 05:29:38.763044 137274321021824 utils.py:1231] [59400] img/sec/core = 164.21623881478905
+I1202 05:29:38.763104 137274321021824 utils.py:1231] [59400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.62349494220832
+I1202 05:29:38.763159 137274321021824 utils.py:1231] [59400] core_hours = 103.62349494220832
+I1202 05:29:38.763221 137274321021824 train.py:125] NOTE: Steps:59400/112603 [52.8%]
+Walltime:4d7h39m (0s eval)
+ETA:3d20h48m
+Total train time:8d4h26m
+I1202 05:34:50.548544 137274321021824 utils.py:1231] [59450] l2_params = 296.3608501194681
+I1202 05:34:50.548831 137274321021824 utils.py:1231] [59450] train/loss = 2.6728204488754272
+I1202 05:34:50.549056 137274321021824 utils.py:1231] [59450] l2_grads = 1.550041675567627
+I1202 05:34:50.549200 137274321021824 utils.py:1231] [59450] lr = 0.0005283455658530663
+I1202 05:34:50.549316 137274321021824 utils.py:1231] [59450] uptime = 373479.911673091
+I1202 05:34:50.549415 137274321021824 utils.py:1231] [59450] examples_seen = 60876800.0
+I1202 05:34:50.549531 137274321021824 utils.py:1231] [59450] progress = 0.5279610667566583
+I1202 05:34:50.549611 137274321021824 utils.py:1231] [59450] epoch = 47.51667815358966
+I1202 05:34:50.549706 137274321021824 utils.py:1231] [59450] img/sec/core = 164.21492887450202
+I1202 05:34:50.549800 137274321021824 utils.py:1231] [59450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.71010230683528
+I1202 05:34:50.549879 137274321021824 utils.py:1231] [59450] core_hours = 103.71010230683528
+I1202 05:34:50.549994 137274321021824 train.py:125] NOTE: Steps:59450/112603 [52.8%]
+Walltime:4d7h44m (0s eval)
+ETA:3d20h43m
+Total train time:8d4h26m
+I1202 05:40:02.302145 137274321021824 utils.py:1231] [59500] l2_params = 296.26477846966543
+I1202 05:40:02.302419 137274321021824 utils.py:1231] [59500] train/loss = 2.5338591933250427
+I1202 05:40:02.302600 137274321021824 utils.py:1231] [59500] l2_grads = 1.670183777809143
+I1202 05:40:02.302697 137274321021824 utils.py:1231] [59500] lr = 0.0005275812910904326
+I1202 05:40:02.302763 137274321021824 utils.py:1231] [59500] uptime = 373791.665123822
+I1202 05:40:02.302834 137274321021824 utils.py:1231] [59500] examples_seen = 60928000.0
+I1202 05:40:02.302908 137274321021824 utils.py:1231] [59500] progress = 0.5284051046597338
+I1202 05:40:02.302970 137274321021824 utils.py:1231] [59500] epoch = 47.556641718058614
+I1202 05:40:02.303030 137274321021824 utils.py:1231] [59500] img/sec/core = 164.23234411663375
+I1202 05:40:02.303092 137274321021824 utils.py:1231] [59500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.79670048759388
+I1202 05:40:02.303154 137274321021824 utils.py:1231] [59500] core_hours = 103.79670048759388
+I1202 05:40:02.303231 137274321021824 train.py:125] NOTE: Steps:59500/112603 [52.8%]
+Walltime:4d7h49m (0s eval)
+ETA:3d20h38m
+Total train time:8d4h26m
+I1202 05:45:14.068214 137274321021824 utils.py:1231] [59550] l2_params = 296.17352101405174
+I1202 05:45:14.068486 137274321021824 utils.py:1231] [59550] train/loss = 2.2614502906799316
+I1202 05:45:14.068615 137274321021824 utils.py:1231] [59550] l2_grads = 1.661726474761963
+I1202 05:45:14.068708 137274321021824 utils.py:1231] [59550] lr = 0.0005268169516829176
+I1202 05:45:14.068784 137274321021824 utils.py:1231] [59550] uptime = 374103.431145322
+I1202 05:45:14.068863 137274321021824 utils.py:1231] [59550] examples_seen = 60979200.0
+I1202 05:45:14.068928 137274321021824 utils.py:1231] [59550] progress = 0.5288491425628091
+I1202 05:45:14.068987 137274321021824 utils.py:1231] [59550] epoch = 47.59660528252757
+I1202 05:45:14.069044 137274321021824 utils.py:1231] [59550] img/sec/core = 164.22572207728948
+I1202 05:45:14.069106 137274321021824 utils.py:1231] [59550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.88330216023276
+I1202 05:45:14.069157 137274321021824 utils.py:1231] [59550] core_hours = 103.88330216023276
+I1202 05:45:14.069223 137274321021824 train.py:125] NOTE: Steps:59550/112603 [52.9%]
+Walltime:4d7h55m (0s eval)
+ETA:3d20h33m
+Total train time:8d4h26m
+I1202 05:50:25.831829 137274321021824 utils.py:1231] [59600] l2_params = 296.0916808116951
+I1202 05:50:25.832098 137274321021824 utils.py:1231] [59600] train/loss = 2.8064883053302765
+I1202 05:50:25.832194 137274321021824 utils.py:1231] [59600] l2_grads = 1.4530882835388184
+I1202 05:50:25.832256 137274321021824 utils.py:1231] [59600] lr = 0.0005260525494219763
+I1202 05:50:25.832309 137274321021824 utils.py:1231] [59600] uptime = 374415.19467024796
+I1202 05:50:25.832362 137274321021824 utils.py:1231] [59600] examples_seen = 61030400.0
+I1202 05:50:25.832411 137274321021824 utils.py:1231] [59600] progress = 0.5292931804658846
+I1202 05:50:25.832460 137274321021824 utils.py:1231] [59600] epoch = 47.636568846996525
+I1202 05:50:25.832513 137274321021824 utils.py:1231] [59600] img/sec/core = 164.2270371819688
+I1202 05:50:25.832569 137274321021824 utils.py:1231] [59600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 103.96990313937887
+I1202 05:50:25.832618 137274321021824 utils.py:1231] [59600] core_hours = 103.96990313937887
+I1202 05:50:25.832677 137274321021824 train.py:125] NOTE: Steps:59600/112603 [52.9%]
+Walltime:4d8h0m (0s eval)
+ETA:3d20h27m
+Total train time:8d4h26m
+I1202 05:55:37.598728 137274321021824 utils.py:1231] [59650] l2_params = 296.00318898687306
+I1202 05:55:37.598984 137274321021824 utils.py:1231] [59650] train/loss = 2.9091480672359467
+I1202 05:55:37.599167 137274321021824 utils.py:1231] [59650] l2_grads = 1.4834191799163818
+I1202 05:55:37.599237 137274321021824 utils.py:1231] [59650] lr = 0.0005252880860992101
+I1202 05:55:37.599295 137274321021824 utils.py:1231] [59650] uptime = 374726.961656671
+I1202 05:55:37.599355 137274321021824 utils.py:1231] [59650] examples_seen = 61081600.0
+I1202 05:55:37.599411 137274321021824 utils.py:1231] [59650] progress = 0.5297372183689599
+I1202 05:55:37.599478 137274321021824 utils.py:1231] [59650] epoch = 47.67653241146549
+I1202 05:55:37.599535 137274321021824 utils.py:1231] [59650] img/sec/core = 164.22521379644604
+I1202 05:55:37.599596 137274321021824 utils.py:1231] [59650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.05650508005195
+I1202 05:55:37.599666 137274321021824 utils.py:1231] [59650] core_hours = 104.05650508005195
+I1202 05:55:37.599731 137274321021824 train.py:125] NOTE: Steps:59650/112603 [53.0%]
+Walltime:4d8h5m (0s eval)
+ETA:3d20h22m
+Total train time:8d4h26m
+I1202 06:00:49.370126 137274321021824 utils.py:1231] [59700] l2_params = 295.90869264775876
+I1202 06:00:49.370339 137274321021824 utils.py:1231] [59700] train/loss = 3.678310513496399
+I1202 06:00:49.370434 137274321021824 utils.py:1231] [59700] l2_grads = 1.5480031967163086
+I1202 06:00:49.370493 137274321021824 utils.py:1231] [59700] lr = 0.0005245235635063641
+I1202 06:00:49.370549 137274321021824 utils.py:1231] [59700] uptime = 375038.73291173
+I1202 06:00:49.370604 137274321021824 utils.py:1231] [59700] examples_seen = 61132800.0
+I1202 06:00:49.370653 137274321021824 utils.py:1231] [59700] progress = 0.5301812562720354
+I1202 06:00:49.370700 137274321021824 utils.py:1231] [59700] epoch = 47.71649597593444
+I1202 06:00:49.370750 137274321021824 utils.py:1231] [59700] img/sec/core = 164.2229652965023
+I1202 06:00:49.370809 137274321021824 utils.py:1231] [59700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.14310820645723
+I1202 06:00:49.370858 137274321021824 utils.py:1231] [59700] core_hours = 104.14310820645723
+I1202 06:00:49.370924 137274321021824 train.py:125] NOTE: Steps:59700/112603 [53.0%]
+Walltime:4d8h10m (0s eval)
+ETA:3d20h17m
+Total train time:8d4h26m
+I1202 06:06:01.217416 137274321021824 utils.py:1231] [59750] l2_params = 295.8229179455979
+I1202 06:06:01.217638 137274321021824 utils.py:1231] [59750] train/loss = 3.1430985927581787
+I1202 06:06:01.217749 137274321021824 utils.py:1231] [59750] l2_grads = 1.4785895347595215
+I1202 06:06:01.217821 137274321021824 utils.py:1231] [59750] lr = 0.0005237589834353221
+I1202 06:06:01.217886 137274321021824 utils.py:1231] [59750] uptime = 375350.58024283696
+I1202 06:06:01.217947 137274321021824 utils.py:1231] [59750] examples_seen = 61184000.0
+I1202 06:06:01.218004 137274321021824 utils.py:1231] [59750] progress = 0.5306252941751108
+I1202 06:06:01.218060 137274321021824 utils.py:1231] [59750] epoch = 47.7564595404034
+I1202 06:06:01.218117 137274321021824 utils.py:1231] [59750] img/sec/core = 164.18290263462288
+I1202 06:06:01.218184 137274321021824 utils.py:1231] [59750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.22973246509805
+I1202 06:06:01.218240 137274321021824 utils.py:1231] [59750] core_hours = 104.22973246509805
+I1202 06:06:01.218306 137274321021824 train.py:125] NOTE: Steps:59750/112603 [53.1%]
+Walltime:4d8h15m (0s eval)
+ETA:3d20h12m
+Total train time:8d4h26m
+I1202 06:11:13.001012 137274321021824 utils.py:1231] [59800] l2_params = 295.74039088761464
+I1202 06:11:13.001226 137274321021824 utils.py:1231] [59800] train/loss = 2.239173710346222
+I1202 06:11:13.001335 137274321021824 utils.py:1231] [59800] l2_grads = 1.7183607816696167
+I1202 06:11:13.001409 137274321021824 utils.py:1231] [59800] lr = 0.0005229943476781026
+I1202 06:11:13.001471 137274321021824 utils.py:1231] [59800] uptime = 375662.36383254296
+I1202 06:11:13.001538 137274321021824 utils.py:1231] [59800] examples_seen = 61235200.0
+I1202 06:11:13.001606 137274321021824 utils.py:1231] [59800] progress = 0.5310693320781862
+I1202 06:11:13.001667 137274321021824 utils.py:1231] [59800] epoch = 47.79642310487235
+I1202 06:11:13.001735 137274321021824 utils.py:1231] [59800] img/sec/core = 164.21646837885126
+I1202 06:11:13.001810 137274321021824 utils.py:1231] [59800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.31633901779416
+I1202 06:11:13.001868 137274321021824 utils.py:1231] [59800] core_hours = 104.31633901779416
+I1202 06:11:13.001955 137274321021824 train.py:125] NOTE: Steps:59800/112603 [53.1%]
+Walltime:4d8h21m (0s eval)
+ETA:3d20h6m
+Total train time:8d4h26m
+I1202 06:16:24.802237 137274321021824 utils.py:1231] [59850] l2_params = 295.65364957110785
+I1202 06:16:24.802433 137274321021824 utils.py:1231] [59850] train/loss = 2.154087573289871
+I1202 06:16:24.802520 137274321021824 utils.py:1231] [59850] l2_grads = 1.6434845924377441
+I1202 06:16:24.802577 137274321021824 utils.py:1231] [59850] lr = 0.0005222296580268554
+I1202 06:16:24.802626 137274321021824 utils.py:1231] [59850] uptime = 375974.164988717
+I1202 06:16:24.802677 137274321021824 utils.py:1231] [59850] examples_seen = 61286400.0
+I1202 06:16:24.802725 137274321021824 utils.py:1231] [59850] progress = 0.5315133699812616
+I1202 06:16:24.802771 137274321021824 utils.py:1231] [59850] epoch = 47.83638666934131
+I1202 06:16:24.802822 137274321021824 utils.py:1231] [59850] img/sec/core = 164.2072166384816
+I1202 06:16:24.802876 137274321021824 utils.py:1231] [59850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.40295045006472
+I1202 06:16:24.802931 137274321021824 utils.py:1231] [59850] core_hours = 104.40295045006472
+I1202 06:16:24.802991 137274321021824 train.py:125] NOTE: Steps:59850/112603 [53.2%]
+Walltime:4d8h26m (0s eval)
+ETA:3d20h1m
+Total train time:8d4h25m
+I1202 06:21:36.591584 137274321021824 utils.py:1231] [59900] l2_params = 295.5777896043948
+I1202 06:21:36.591796 137274321021824 utils.py:1231] [59900] train/loss = 4.230537712574005
+I1202 06:21:36.591901 137274321021824 utils.py:1231] [59900] l2_grads = 1.4865602254867554
+I1202 06:21:36.591960 137274321021824 utils.py:1231] [59900] lr = 0.0005214649162738554
+I1202 06:21:36.592021 137274321021824 utils.py:1231] [59900] uptime = 376285.954383665
+I1202 06:21:36.592084 137274321021824 utils.py:1231] [59900] examples_seen = 61337600.0
+I1202 06:21:36.592131 137274321021824 utils.py:1231] [59900] progress = 0.531957407884337
+I1202 06:21:36.592177 137274321021824 utils.py:1231] [59900] epoch = 47.87635023381027
+I1202 06:21:36.592226 137274321021824 utils.py:1231] [59900] img/sec/core = 164.21341081386214
+I1202 06:21:36.592279 137274321021824 utils.py:1231] [59900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.48955861532805
+I1202 06:21:36.592328 137274321021824 utils.py:1231] [59900] core_hours = 104.48955861532805
+I1202 06:21:36.592387 137274321021824 train.py:125] NOTE: Steps:59900/112603 [53.2%]
+Walltime:4d8h31m (0s eval)
+ETA:3d19h56m
+Total train time:8d4h25m
+I1202 06:26:48.378663 137274321021824 utils.py:1231] [59950] l2_params = 295.4815652199417
+I1202 06:26:48.378870 137274321021824 utils.py:1231] [59950] train/loss = 4.8199750781059265
+I1202 06:26:48.378970 137274321021824 utils.py:1231] [59950] l2_grads = 1.5084328651428223
+I1202 06:26:48.379039 137274321021824 utils.py:1231] [59950] lr = 0.0005207001242115007
+I1202 06:26:48.379090 137274321021824 utils.py:1231] [59950] uptime = 376597.74145241897
+I1202 06:26:48.379143 137274321021824 utils.py:1231] [59950] examples_seen = 61388800.0
+I1202 06:26:48.379192 137274321021824 utils.py:1231] [59950] progress = 0.5324014457874124
+I1202 06:26:48.379241 137274321021824 utils.py:1231] [59950] epoch = 47.916313798279226
+I1202 06:26:48.379293 137274321021824 utils.py:1231] [59950] img/sec/core = 164.2146359841559
+I1202 06:26:48.379349 137274321021824 utils.py:1231] [59950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.57616613442637
+I1202 06:26:48.379400 137274321021824 utils.py:1231] [59950] core_hours = 104.57616613442637
+I1202 06:26:48.379460 137274321021824 train.py:125] NOTE: Steps:59950/112603 [53.2%]
+Walltime:4d8h36m (0s eval)
+ETA:3d19h51m
+Total train time:8d4h25m
+I1202 06:32:00.161416 137274321021824 utils.py:1231] [60000] l2_params = 295.3985499457244
+I1202 06:32:00.161622 137274321021824 utils.py:1231] [60000] train/loss = 3.0618233382701874
+I1202 06:32:00.161732 137274321021824 utils.py:1231] [60000] l2_grads = 1.6128909587860107
+I1202 06:32:00.161827 137274321021824 utils.py:1231] [60000] lr = 0.0005199352836323064
+I1202 06:32:00.161924 137274321021824 utils.py:1231] [60000] uptime = 376909.524280162
+I1202 06:32:00.162003 137274321021824 utils.py:1231] [60000] examples_seen = 61440000.0
+I1202 06:32:00.162068 137274321021824 utils.py:1231] [60000] progress = 0.5328454836904878
+I1202 06:32:00.162131 137274321021824 utils.py:1231] [60000] epoch = 47.95627736274818
+I1202 06:32:00.162192 137274321021824 utils.py:1231] [60000] img/sec/core = 164.21686970585014
+I1202 06:32:00.162256 137274321021824 utils.py:1231] [60000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.6627724754661
+I1202 06:32:00.162312 137274321021824 utils.py:1231] [60000] core_hours = 104.6627724754661
+I1202 06:32:00.162391 137274321021824 train.py:125] NOTE: Steps:60000/112603 [53.3%]
+Walltime:4d8h41m (0s eval)
+ETA:3d19h45m
+Total train time:8d4h25m
+I1202 06:32:00.537401 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:60000/112603 [53.3%]
+Walltime:4d8h41m (0s eval)
+ETA:3d19h45m
+Total train time:8d4h25m
+I1202 06:33:36.936101 137274321021824 utils.py:1231] [60000] val/acc@1 = 0.6681481186224489
+I1202 06:33:36.936354 137274321021824 utils.py:1231] [60000] val/loss = 1.3596173404734961
+I1202 06:33:36.936573 137274321021824 utils.py:1231] [60000] z/secs/eval/val = 96.39891782798804
+I1202 06:33:36.936683 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 96.39891782798804
+I1202 06:38:47.223788 137274321021824 utils.py:1231] [60050] l2_params = 295.3162703197728
+I1202 06:38:47.224023 137274321021824 utils.py:1231] [60050] train/loss = 2.4442699253559113
+I1202 06:38:47.224118 137274321021824 utils.py:1231] [60050] l2_grads = 1.7433117628097534
+I1202 06:38:47.224179 137274321021824 utils.py:1231] [60050] lr = 0.0005191703963289016
+I1202 06:38:47.224231 137274321021824 utils.py:1231] [60050] uptime = 377316.586593181
+I1202 06:38:47.224283 137274321021824 utils.py:1231] [60050] examples_seen = 61491200.0
+I1202 06:38:47.224333 137274321021824 utils.py:1231] [60050] progress = 0.5332895215935632
+I1202 06:38:47.224388 137274321021824 utils.py:1231] [60050] epoch = 47.99624092721714
+I1202 06:38:47.224444 137274321021824 utils.py:1231] [60050] img/sec/core = 125.77926858487957
+I1202 06:38:47.224498 137274321021824 utils.py:1231] [60050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.77584534019361
+I1202 06:38:47.224549 137274321021824 utils.py:1231] [60050] core_hours = 104.77584534019361
+I1202 06:38:47.224610 137274321021824 train.py:125] NOTE: Steps:60050/112603 [53.3%]
+Walltime:4d8h48m (0s eval)
+ETA:3d19h41m
+Total train time:8d4h28m
+I1202 06:43:58.884585 137274321021824 utils.py:1231] [60100] l2_params = 295.21818254456065
+I1202 06:43:58.884862 137274321021824 utils.py:1231] [60100] train/loss = 4.182109326124191
+I1202 06:43:58.885087 137274321021824 utils.py:1231] [60100] l2_grads = 1.6313775777816772
+I1202 06:43:58.885192 137274321021824 utils.py:1231] [60100] lr = 0.0005184054640940257
+I1202 06:43:58.885266 137274321021824 utils.py:1231] [60100] uptime = 377628.247624493
+I1202 06:43:58.885339 137274321021824 utils.py:1231] [60100] examples_seen = 61542400.0
+I1202 06:43:58.885428 137274321021824 utils.py:1231] [60100] progress = 0.5337335594966386
+I1202 06:43:58.885524 137274321021824 utils.py:1231] [60100] epoch = 48.03620449168609
+I1202 06:43:58.885605 137274321021824 utils.py:1231] [60100] img/sec/core = 164.28104528971699
+I1202 06:43:58.885699 137274321021824 utils.py:1231] [60100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.86241784889138
+I1202 06:43:58.885752 137274321021824 utils.py:1231] [60100] core_hours = 104.86241784889138
+I1202 06:43:58.885835 137274321021824 train.py:125] NOTE: Steps:60100/112603 [53.4%]
+Walltime:4d8h53m (0s eval)
+ETA:3d19h36m
+Total train time:8d4h28m
+I1202 06:49:10.644951 137274321021824 utils.py:1231] [60150] l2_params = 295.1379983430566
+I1202 06:49:10.645235 137274321021824 utils.py:1231] [60150] train/loss = 2.284355252981186
+I1202 06:49:10.645364 137274321021824 utils.py:1231] [60150] l2_grads = 1.699129343032837
+I1202 06:49:10.645459 137274321021824 utils.py:1231] [60150] lr = 0.0005176404887205218
+I1202 06:49:10.645531 137274321021824 utils.py:1231] [60150] uptime = 377940.00789254
+I1202 06:49:10.645610 137274321021824 utils.py:1231] [60150] examples_seen = 61593600.0
+I1202 06:49:10.645668 137274321021824 utils.py:1231] [60150] progress = 0.534177597399714
+I1202 06:49:10.645723 137274321021824 utils.py:1231] [60150] epoch = 48.076168056155055
+I1202 06:49:10.645781 137274321021824 utils.py:1231] [60150] img/sec/core = 164.22875281937445
+I1202 06:49:10.645853 137274321021824 utils.py:1231] [60150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 104.94901792334889
+I1202 06:49:10.645930 137274321021824 utils.py:1231] [60150] core_hours = 104.94901792334889
+I1202 06:49:10.646000 137274321021824 train.py:125] NOTE: Steps:60150/112603 [53.4%]
+Walltime:4d8h59m (0s eval)
+ETA:3d19h31m
+Total train time:8d4h28m
+I1202 06:54:22.392695 137274321021824 utils.py:1231] [60200] l2_params = 295.038947040359
+I1202 06:54:22.392975 137274321021824 utils.py:1231] [60200] train/loss = 2.222465008497238
+I1202 06:54:22.393158 137274321021824 utils.py:1231] [60200] l2_grads = 1.5704312324523926
+I1202 06:54:22.393298 137274321021824 utils.py:1231] [60200] lr = 0.0005168754720013362
+I1202 06:54:22.393416 137274321021824 utils.py:1231] [60200] uptime = 378251.755768308
+I1202 06:54:22.393515 137274321021824 utils.py:1231] [60200] examples_seen = 61644800.0
+I1202 06:54:22.393601 137274321021824 utils.py:1231] [60200] progress = 0.5346216353027895
+I1202 06:54:22.393695 137274321021824 utils.py:1231] [60200] epoch = 48.11613162062401
+I1202 06:54:22.393774 137274321021824 utils.py:1231] [60200] img/sec/core = 164.2352810708605
+I1202 06:54:22.393850 137274321021824 utils.py:1231] [60200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.03561455550667
+I1202 06:54:22.393929 137274321021824 utils.py:1231] [60200] core_hours = 105.03561455550667
+I1202 06:54:22.394002 137274321021824 train.py:125] NOTE: Steps:60200/112603 [53.5%]
+Walltime:4d9h4m (0s eval)
+ETA:3d19h26m
+Total train time:8d4h28m
+I1202 06:59:34.160840 137274321021824 utils.py:1231] [60250] l2_params = 294.939657814841
+I1202 06:59:34.161060 137274321021824 utils.py:1231] [60250] train/loss = 3.16524401307106
+I1202 06:59:34.161177 137274321021824 utils.py:1231] [60250] l2_grads = 1.4826531410217285
+I1202 06:59:34.161252 137274321021824 utils.py:1231] [60250] lr = 0.0005161104157295096
+I1202 06:59:34.161313 137274321021824 utils.py:1231] [60250] uptime = 378563.523674305
+I1202 06:59:34.161381 137274321021824 utils.py:1231] [60250] examples_seen = 61696000.0
+I1202 06:59:34.161446 137274321021824 utils.py:1231] [60250] progress = 0.5350656732058648
+I1202 06:59:34.161503 137274321021824 utils.py:1231] [60250] epoch = 48.156095185092965
+I1202 06:59:34.161566 137274321021824 utils.py:1231] [60250] img/sec/core = 164.22472940653915
+I1202 06:59:34.161628 137274321021824 utils.py:1231] [60250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.12221675161693
+I1202 06:59:34.161686 137274321021824 utils.py:1231] [60250] core_hours = 105.12221675161693
+I1202 06:59:34.161753 137274321021824 train.py:125] NOTE: Steps:60250/112603 [53.5%]
+Walltime:4d9h9m (0s eval)
+ETA:3d19h20m
+Total train time:8d4h28m
+I1202 07:04:44.674000 137274321021824 utils.py:1231] [60300] l2_params = 294.8467491005457
+I1202 07:04:44.674195 137274321021824 utils.py:1231] [60300] train/loss = 2.145082399249077
+I1202 07:04:44.674287 137274321021824 utils.py:1231] [60300] l2_grads = 1.7566533088684082
+I1202 07:04:44.674345 137274321021824 utils.py:1231] [60300] lr = 0.0005153453216981775
+I1202 07:04:44.674397 137274321021824 utils.py:1231] [60300] uptime = 378874.036758481
+I1202 07:04:44.674448 137274321021824 utils.py:1231] [60300] examples_seen = 61747200.0
+I1202 07:04:44.674497 137274321021824 utils.py:1231] [60300] progress = 0.5355097111089403
+I1202 07:04:44.674544 137274321021824 utils.py:1231] [60300] epoch = 48.19605874956192
+I1202 07:04:44.674594 137274321021824 utils.py:1231] [60300] img/sec/core = 164.88838187243738
+I1202 07:04:44.674652 137274321021824 utils.py:1231] [60300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.20847038611028
+I1202 07:04:44.674703 137274321021824 utils.py:1231] [60300] core_hours = 105.20847038611028
+I1202 07:04:44.674764 137274321021824 train.py:125] NOTE: Steps:60300/112603 [53.6%]
+Walltime:4d9h14m (0s eval)
+ETA:3d19h15m
+Total train time:8d4h28m
+I1202 07:09:56.446894 137274321021824 utils.py:1231] [60350] l2_params = 294.7555352280529
+I1202 07:09:56.447165 137274321021824 utils.py:1231] [60350] train/loss = 3.452998995780945
+I1202 07:09:56.447295 137274321021824 utils.py:1231] [60350] l2_grads = 1.503095269203186
+I1202 07:09:56.447376 137274321021824 utils.py:1231] [60350] lr = 0.0005145801917005631
+I1202 07:09:56.447438 137274321021824 utils.py:1231] [60350] uptime = 379185.809799787
+I1202 07:09:56.447491 137274321021824 utils.py:1231] [60350] examples_seen = 61798400.0
+I1202 07:09:56.447541 137274321021824 utils.py:1231] [60350] progress = 0.5359537490120156
+I1202 07:09:56.447590 137274321021824 utils.py:1231] [60350] epoch = 48.236022314030876
+I1202 07:09:56.447642 137274321021824 utils.py:1231] [60350] img/sec/core = 164.2220244108508
+I1202 07:09:56.447704 137274321021824 utils.py:1231] [60350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.29507400869528
+I1202 07:09:56.447755 137274321021824 utils.py:1231] [60350] core_hours = 105.29507400869528
+I1202 07:09:56.447817 137274321021824 train.py:125] NOTE: Steps:60350/112603 [53.6%]
+Walltime:4d9h19m (0s eval)
+ETA:3d19h10m
+Total train time:8d4h28m
+I1202 07:15:08.228350 137274321021824 utils.py:1231] [60400] l2_params = 294.66828910247386
+I1202 07:15:08.228565 137274321021824 utils.py:1231] [60400] train/loss = 2.1491039097309113
+I1202 07:15:08.228671 137274321021824 utils.py:1231] [60400] l2_grads = 1.7233684062957764
+I1202 07:15:08.228763 137274321021824 utils.py:1231] [60400] lr = 0.0005138150275299745
+I1202 07:15:08.228843 137274321021824 utils.py:1231] [60400] uptime = 379497.59119807096
+I1202 07:15:08.228946 137274321021824 utils.py:1231] [60400] examples_seen = 61849600.0
+I1202 07:15:08.229017 137274321021824 utils.py:1231] [60400] progress = 0.5363977869150911
+I1202 07:15:08.229074 137274321021824 utils.py:1231] [60400] epoch = 48.27598587849984
+I1202 07:15:08.229133 137274321021824 utils.py:1231] [60400] img/sec/core = 164.21762260931646
+I1202 07:15:08.229206 137274321021824 utils.py:1231] [60400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.38167995266305
+I1202 07:15:08.229270 137274321021824 utils.py:1231] [60400] core_hours = 105.38167995266305
+I1202 07:15:08.229338 137274321021824 train.py:125] NOTE: Steps:60400/112603 [53.6%]
+Walltime:4d9h24m (0s eval)
+ETA:3d19h4m
+Total train time:8d4h28m
+I1202 07:20:18.764162 137274321021824 utils.py:1231] [60450] l2_params = 294.58214776312235
+I1202 07:20:18.764370 137274321021824 utils.py:1231] [60450] train/loss = 2.943434715270996
+I1202 07:20:18.764459 137274321021824 utils.py:1231] [60450] l2_grads = 1.545562982559204
+I1202 07:20:18.764516 137274321021824 utils.py:1231] [60450] lr = 0.0005130498309797987
+I1202 07:20:18.764564 137274321021824 utils.py:1231] [60450] uptime = 379808.12692691997
+I1202 07:20:18.764618 137274321021824 utils.py:1231] [60450] examples_seen = 61900800.0
+I1202 07:20:18.764665 137274321021824 utils.py:1231] [60450] progress = 0.5368418248181664
+I1202 07:20:18.764713 137274321021824 utils.py:1231] [60450] epoch = 48.315949442968794
+I1202 07:20:18.764761 137274321021824 utils.py:1231] [60450] img/sec/core = 164.87635799517136
+I1202 07:20:18.764814 137274321021824 utils.py:1231] [60450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.46793987734331
+I1202 07:20:18.764863 137274321021824 utils.py:1231] [60450] core_hours = 105.46793987734331
+I1202 07:20:18.764946 137274321021824 train.py:125] NOTE: Steps:60450/112603 [53.7%]
+Walltime:4d9h30m (0s eval)
+ETA:3d18h59m
+Total train time:8d4h27m
+I1202 07:25:30.561484 137274321021824 utils.py:1231] [60500] l2_params = 294.4932585597975
+I1202 07:25:30.561763 137274321021824 utils.py:1231] [60500] train/loss = 2.227520704269409
+I1202 07:25:30.561904 137274321021824 utils.py:1231] [60500] l2_grads = 1.6648255586624146
+I1202 07:25:30.562036 137274321021824 utils.py:1231] [60500] lr = 0.0005122846038434993
+I1202 07:25:30.562148 137274321021824 utils.py:1231] [60500] uptime = 380119.924500493
+I1202 07:25:30.562220 137274321021824 utils.py:1231] [60500] examples_seen = 61952000.0
+I1202 07:25:30.562281 137274321021824 utils.py:1231] [60500] progress = 0.5372858627212419
+I1202 07:25:30.562342 137274321021824 utils.py:1231] [60500] epoch = 48.35591300743775
+I1202 07:25:30.562408 137274321021824 utils.py:1231] [60500] img/sec/core = 164.20910340410458
+I1202 07:25:30.562489 137274321021824 utils.py:1231] [60500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.55455031444694
+I1202 07:25:30.562586 137274321021824 utils.py:1231] [60500] core_hours = 105.55455031444694
+I1202 07:25:30.562684 137274321021824 train.py:125] NOTE: Steps:60500/112603 [53.7%]
+Walltime:4d9h35m (0s eval)
+ETA:3d18h54m
+Total train time:8d4h27m
+I1202 07:30:42.392444 137274321021824 utils.py:1231] [60550] l2_params = 294.40522354824896
+I1202 07:30:42.392668 137274321021824 utils.py:1231] [60550] train/loss = 2.11660398542881
+I1202 07:30:42.392770 137274321021824 utils.py:1231] [60550] l2_grads = 1.7443031072616577
+I1202 07:30:42.392840 137274321021824 utils.py:1231] [60550] lr = 0.0005115193479146117
+I1202 07:30:42.392908 137274321021824 utils.py:1231] [60550] uptime = 380431.755269132
+I1202 07:30:42.392967 137274321021824 utils.py:1231] [60550] examples_seen = 62003200.0
+I1202 07:30:42.393023 137274321021824 utils.py:1231] [60550] progress = 0.5377299006243172
+I1202 07:30:42.393078 137274321021824 utils.py:1231] [60550] epoch = 48.395876571906705
+I1202 07:30:42.393134 137274321021824 utils.py:1231] [60550] img/sec/core = 164.19162298661743
+I1202 07:30:42.393195 137274321021824 utils.py:1231] [60550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.64116997240221
+I1202 07:30:42.393246 137274321021824 utils.py:1231] [60550] core_hours = 105.64116997240221
+I1202 07:30:42.393310 137274321021824 train.py:125] NOTE: Steps:60550/112603 [53.8%]
+Walltime:4d9h40m (0s eval)
+ETA:3d18h49m
+Total train time:8d4h27m
+I1202 07:35:54.182666 137274321021824 utils.py:1231] [60600] l2_params = 294.3119670061513
+I1202 07:35:54.182912 137274321021824 utils.py:1231] [60600] train/loss = 2.2295336425304413
+I1202 07:35:54.183017 137274321021824 utils.py:1231] [60600] l2_grads = 1.6563092470169067
+I1202 07:35:54.183086 137274321021824 utils.py:1231] [60600] lr = 0.000510754064986739
+I1202 07:35:54.183147 137274321021824 utils.py:1231] [60600] uptime = 380743.545508631
+I1202 07:35:54.183208 137274321021824 utils.py:1231] [60600] examples_seen = 62054400.0
+I1202 07:35:54.183269 137274321021824 utils.py:1231] [60600] progress = 0.5381739385273927
+I1202 07:35:54.183326 137274321021824 utils.py:1231] [60600] epoch = 48.43584013637567
+I1202 07:35:54.183385 137274321021824 utils.py:1231] [60600] img/sec/core = 164.21296600647042
+I1202 07:35:54.183446 137274321021824 utils.py:1231] [60600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.72777837226305
+I1202 07:35:54.183504 137274321021824 utils.py:1231] [60600] core_hours = 105.72777837226305
+I1202 07:35:54.183572 137274321021824 train.py:125] NOTE: Steps:60600/112603 [53.8%]
+Walltime:4d9h45m (0s eval)
+ETA:3d18h43m
+Total train time:8d4h27m
+I1202 07:41:05.962893 137274321021824 utils.py:1231] [60650] l2_params = 294.2263912566999
+I1202 07:41:05.963092 137274321021824 utils.py:1231] [60650] train/loss = 2.160022869706154
+I1202 07:41:05.963192 137274321021824 utils.py:1231] [60650] l2_grads = 1.6498347520828247
+I1202 07:41:05.963249 137274321021824 utils.py:1231] [60650] lr = 0.0005099887568535463
+I1202 07:41:05.963300 137274321021824 utils.py:1231] [60650] uptime = 381055.325662478
+I1202 07:41:05.963351 137274321021824 utils.py:1231] [60650] examples_seen = 62105600.0
+I1202 07:41:05.963401 137274321021824 utils.py:1231] [60650] progress = 0.5386179764304682
+I1202 07:41:05.963448 137274321021824 utils.py:1231] [60650] epoch = 48.47580370084462
+I1202 07:41:05.963525 137274321021824 utils.py:1231] [60650] img/sec/core = 164.21827806629412
+I1202 07:41:05.963580 137274321021824 utils.py:1231] [60650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.81438397055389
+I1202 07:41:05.963632 137274321021824 utils.py:1231] [60650] core_hours = 105.81438397055389
+I1202 07:41:05.963691 137274321021824 train.py:125] NOTE: Steps:60650/112603 [53.9%]
+Walltime:4d9h50m (0s eval)
+ETA:3d18h38m
+Total train time:8d4h27m
+I1202 07:46:17.745356 137274321021824 utils.py:1231] [60700] l2_params = 294.14448058733467
+I1202 07:46:17.745583 137274321021824 utils.py:1231] [60700] train/loss = 2.4013363122940063
+I1202 07:46:17.745682 137274321021824 utils.py:1231] [60700] l2_grads = 1.5840420722961426
+I1202 07:46:17.745741 137274321021824 utils.py:1231] [60700] lr = 0.000509223425308759
+I1202 07:46:17.745793 137274321021824 utils.py:1231] [60700] uptime = 381367.108155431
+I1202 07:46:17.745846 137274321021824 utils.py:1231] [60700] examples_seen = 62156800.0
+I1202 07:46:17.745903 137274321021824 utils.py:1231] [60700] progress = 0.5390620143335435
+I1202 07:46:17.745957 137274321021824 utils.py:1231] [60700] epoch = 48.51576726531358
+I1202 07:46:17.746008 137274321021824 utils.py:1231] [60700] img/sec/core = 164.21704604086813
+I1202 07:46:17.746063 137274321021824 utils.py:1231] [60700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.90099021859639
+I1202 07:46:17.746114 137274321021824 utils.py:1231] [60700] core_hours = 105.90099021859639
+I1202 07:46:17.746175 137274321021824 train.py:125] NOTE: Steps:60700/112603 [53.9%]
+Walltime:4d9h56m (0s eval)
+ETA:3d18h33m
+Total train time:8d4h27m
+I1202 07:51:29.519854 137274321021824 utils.py:1231] [60750] l2_params = 294.03715202484636
+I1202 07:51:29.520072 137274321021824 utils.py:1231] [60750] train/loss = 2.5974367558956146
+I1202 07:51:29.520174 137274321021824 utils.py:1231] [60750] l2_grads = 1.613297462463379
+I1202 07:51:29.520251 137274321021824 utils.py:1231] [60750] lr = 0.0005084580721461579
+I1202 07:51:29.520305 137274321021824 utils.py:1231] [60750] uptime = 381678.882667271
+I1202 07:51:29.520355 137274321021824 utils.py:1231] [60750] examples_seen = 62208000.0
+I1202 07:51:29.520401 137274321021824 utils.py:1231] [60750] progress = 0.539506052236619
+I1202 07:51:29.520448 137274321021824 utils.py:1231] [60750] epoch = 48.55573082978253
+I1202 07:51:29.520496 137274321021824 utils.py:1231] [60750] img/sec/core = 164.22124983158386
+I1202 07:51:29.520548 137274321021824 utils.py:1231] [60750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 105.98759424966306
+I1202 07:51:29.520595 137274321021824 utils.py:1231] [60750] core_hours = 105.98759424966306
+I1202 07:51:29.520653 137274321021824 train.py:125] NOTE: Steps:60750/112603 [54.0%]
+Walltime:4d10h1m (0s eval)
+ETA:3d18h28m
+Total train time:8d4h27m
+I1202 07:56:41.288775 137274321021824 utils.py:1231] [60800] l2_params = 293.9412105085
+I1202 07:56:41.289078 137274321021824 utils.py:1231] [60800] train/loss = 2.3003086149692535
+I1202 07:56:41.289183 137274321021824 utils.py:1231] [60800] l2_grads = 1.5831153392791748
+I1202 07:56:41.289265 137274321021824 utils.py:1231] [60800] lr = 0.0005076926991595728
+I1202 07:56:41.289326 137274321021824 utils.py:1231] [60800] uptime = 381990.651687518
+I1202 07:56:41.289397 137274321021824 utils.py:1231] [60800] examples_seen = 62259200.0
+I1202 07:56:41.289455 137274321021824 utils.py:1231] [60800] progress = 0.5399500901396943
+I1202 07:56:41.289513 137274321021824 utils.py:1231] [60800] epoch = 48.59569439425149
+I1202 07:56:41.289568 137274321021824 utils.py:1231] [60800] img/sec/core = 164.22414247393655
+I1202 07:56:41.289627 137274321021824 utils.py:1231] [60800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.07419675528722
+I1202 07:56:41.289681 137274321021824 utils.py:1231] [60800] core_hours = 106.07419675528722
+I1202 07:56:41.289750 137274321021824 train.py:125] NOTE: Steps:60800/112603 [54.0%]
+Walltime:4d10h6m (0s eval)
+ETA:3d18h22m
+Total train time:8d4h27m
+I1202 08:01:53.056241 137274321021824 utils.py:1231] [60850] l2_params = 293.8655815397042
+I1202 08:01:53.056583 137274321021824 utils.py:1231] [60850] train/loss = 2.240939348936081
+I1202 08:01:53.056774 137274321021824 utils.py:1231] [60850] l2_grads = 1.6978647708892822
+I1202 08:01:53.056855 137274321021824 utils.py:1231] [60850] lr = 0.0005069273081428801
+I1202 08:01:53.056932 137274321021824 utils.py:1231] [60850] uptime = 382302.41929338296
+I1202 08:01:53.056990 137274321021824 utils.py:1231] [60850] examples_seen = 62310400.0
+I1202 08:01:53.057046 137274321021824 utils.py:1231] [60850] progress = 0.5403941280427698
+I1202 08:01:53.057101 137274321021824 utils.py:1231] [60850] epoch = 48.63565795872045
+I1202 08:01:53.057158 137274321021824 utils.py:1231] [60850] img/sec/core = 164.2248875021859
+I1202 08:01:53.057218 137274321021824 utils.py:1231] [60850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.16079886802748
+I1202 08:01:53.057272 137274321021824 utils.py:1231] [60850] core_hours = 106.16079886802748
+I1202 08:01:53.057342 137274321021824 train.py:125] NOTE: Steps:60850/112603 [54.0%]
+Walltime:4d10h11m (0s eval)
+ETA:3d18h17m
+Total train time:8d4h27m
+I1202 08:07:00.655267 137274321021824 utils.py:1231] [60900] l2_params = 293.78007785127886
+I1202 08:07:00.655478 137274321021824 utils.py:1231] [60900] train/loss = 2.9650618135929108
+I1202 08:07:00.655583 137274321021824 utils.py:1231] [60900] l2_grads = 1.5482045412063599
+I1202 08:07:00.655668 137274321021824 utils.py:1231] [60900] lr = 0.0005061619008900007
+I1202 08:07:00.655738 137274321021824 utils.py:1231] [60900] uptime = 382610.0180996
+I1202 08:07:00.655800 137274321021824 utils.py:1231] [60900] examples_seen = 62361600.0
+I1202 08:07:00.655858 137274321021824 utils.py:1231] [60900] progress = 0.5408381659458451
+I1202 08:07:00.655925 137274321021824 utils.py:1231] [60900] epoch = 48.675621523189406
+I1202 08:07:00.655983 137274321021824 utils.py:1231] [60900] img/sec/core = 166.4505809683689
+I1202 08:07:00.656044 137274321021824 utils.py:1231] [60900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.24624298086556
+I1202 08:07:00.656100 137274321021824 utils.py:1231] [60900] core_hours = 106.24624298086556
+I1202 08:07:00.656163 137274321021824 train.py:125] NOTE: Steps:60900/112603 [54.1%]
+Walltime:4d10h16m (0s eval)
+ETA:3d18h12m
+Total train time:8d4h27m
+I1202 08:12:11.450032 137274321021824 utils.py:1231] [60950] l2_params = 293.6913263219212
+I1202 08:12:11.450271 137274321021824 utils.py:1231] [60950] train/loss = 3.5601473450660706
+I1202 08:12:11.450373 137274321021824 utils.py:1231] [60950] l2_grads = 1.5763863325119019
+I1202 08:12:11.450442 137274321021824 utils.py:1231] [60950] lr = 0.0005053964791948911
+I1202 08:12:11.450507 137274321021824 utils.py:1231] [60950] uptime = 382920.812868314
+I1202 08:12:11.450568 137274321021824 utils.py:1231] [60950] examples_seen = 62412800.0
+I1202 08:12:11.450626 137274321021824 utils.py:1231] [60950] progress = 0.5412822038489206
+I1202 08:12:11.450682 137274321021824 utils.py:1231] [60950] epoch = 48.71558508765836
+I1202 08:12:11.450741 137274321021824 utils.py:1231] [60950] img/sec/core = 164.73893756917084
+I1202 08:12:11.450804 137274321021824 utils.py:1231] [60950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.33257486106388
+I1202 08:12:11.450861 137274321021824 utils.py:1231] [60950] core_hours = 106.33257486106388
+I1202 08:12:11.450933 137274321021824 train.py:125] NOTE: Steps:60950/112603 [54.1%]
+Walltime:4d10h22m (0s eval)
+ETA:3d18h6m
+Total train time:8d4h27m
+I1202 08:17:22.292800 137274321021824 utils.py:1231] [61000] l2_params = 293.6115995650382
+I1202 08:17:22.293001 137274321021824 utils.py:1231] [61000] train/loss = 2.521895319223404
+I1202 08:17:22.293108 137274321021824 utils.py:1231] [61000] l2_grads = 1.7006789445877075
+I1202 08:17:22.293180 137274321021824 utils.py:1231] [61000] lr = 0.0005046310448515427
+I1202 08:17:22.293245 137274321021824 utils.py:1231] [61000] uptime = 383231.655605786
+I1202 08:17:22.293306 137274321021824 utils.py:1231] [61000] examples_seen = 62464000.0
+I1202 08:17:22.293364 137274321021824 utils.py:1231] [61000] progress = 0.5417262417519959
+I1202 08:17:22.293422 137274321021824 utils.py:1231] [61000] epoch = 48.75554865212732
+I1202 08:17:22.293488 137274321021824 utils.py:1231] [61000] img/sec/core = 164.7135153177139
+I1202 08:17:22.293550 137274321021824 utils.py:1231] [61000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.41892006591723
+I1202 08:17:22.293607 137274321021824 utils.py:1231] [61000] core_hours = 106.41892006591723
+I1202 08:17:22.293673 137274321021824 train.py:125] NOTE: Steps:61000/112603 [54.2%]
+Walltime:4d10h27m (0s eval)
+ETA:3d18h1m
+Total train time:8d4h27m
+I1202 08:22:29.841030 137274321021824 utils.py:1231] [61050] l2_params = 293.49706646699735
+I1202 08:22:29.841261 137274321021824 utils.py:1231] [61050] train/loss = 2.252588927745819
+I1202 08:22:29.841355 137274321021824 utils.py:1231] [61050] l2_grads = 1.7304447889328003
+I1202 08:22:29.841437 137274321021824 utils.py:1231] [61050] lr = 0.0005038655996539759
+I1202 08:22:29.841504 137274321021824 utils.py:1231] [61050] uptime = 383539.203865596
+I1202 08:22:29.841565 137274321021824 utils.py:1231] [61050] examples_seen = 62515200.0
+I1202 08:22:29.841634 137274321021824 utils.py:1231] [61050] progress = 0.5421702796550714
+I1202 08:22:29.841685 137274321021824 utils.py:1231] [61050] epoch = 48.79551221659627
+I1202 08:22:29.841737 137274321021824 utils.py:1231] [61050] img/sec/core = 166.477937581689
+I1202 08:22:29.841794 137274321021824 utils.py:1231] [61050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.50435013808666
+I1202 08:22:29.841853 137274321021824 utils.py:1231] [61050] core_hours = 106.50435013808666
+I1202 08:22:29.841921 137274321021824 train.py:125] NOTE: Steps:61050/112603 [54.2%]
+Walltime:4d10h32m (0s eval)
+ETA:3d17h56m
+Total train time:8d4h26m
+I1202 08:27:40.443410 137274321021824 utils.py:1231] [61100] l2_params = 293.4129381984802
+I1202 08:27:40.443618 137274321021824 utils.py:1231] [61100] train/loss = 4.734778583049774
+I1202 08:27:40.443720 137274321021824 utils.py:1231] [61100] l2_grads = 1.5121796131134033
+I1202 08:27:40.443792 137274321021824 utils.py:1231] [61100] lr = 0.0005031001453962377
+I1202 08:27:40.443855 137274321021824 utils.py:1231] [61100] uptime = 383849.806216169
+I1202 08:27:40.443925 137274321021824 utils.py:1231] [61100] examples_seen = 62566400.0
+I1202 08:27:40.443983 137274321021824 utils.py:1231] [61100] progress = 0.5426143175581467
+I1202 08:27:40.444038 137274321021824 utils.py:1231] [61100] epoch = 48.835475781065234
+I1202 08:27:40.444095 137274321021824 utils.py:1231] [61100] img/sec/core = 164.84099333294344
+I1202 08:27:40.444162 137274321021824 utils.py:1231] [61100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.59062856880138
+I1202 08:27:40.444218 137274321021824 utils.py:1231] [61100] core_hours = 106.59062856880138
+I1202 08:27:40.444287 137274321021824 train.py:125] NOTE: Steps:61100/112603 [54.3%]
+Walltime:4d10h37m (0s eval)
+ETA:3d17h51m
+Total train time:8d4h26m
+I1202 08:32:52.203686 137274321021824 utils.py:1231] [61150] l2_params = 293.3217464692903
+I1202 08:32:52.203877 137274321021824 utils.py:1231] [61150] train/loss = 4.256334185600281
+I1202 08:32:52.203988 137274321021824 utils.py:1231] [61150] l2_grads = 1.5010322332382202
+I1202 08:32:52.204055 137274321021824 utils.py:1231] [61150] lr = 0.000502334683872395
+I1202 08:32:52.204113 137274321021824 utils.py:1231] [61150] uptime = 384161.566474778
+I1202 08:32:52.204169 137274321021824 utils.py:1231] [61150] examples_seen = 62617600.0
+I1202 08:32:52.204230 137274321021824 utils.py:1231] [61150] progress = 0.5430583554612222
+I1202 08:32:52.204283 137274321021824 utils.py:1231] [61150] epoch = 48.87543934553419
+I1202 08:32:52.204338 137274321021824 utils.py:1231] [61150] img/sec/core = 164.22875779112664
+I1202 08:32:52.204399 137274321021824 utils.py:1231] [61150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.67722864063721
+I1202 08:32:52.204457 137274321021824 utils.py:1231] [61150] core_hours = 106.67722864063721
+I1202 08:32:52.204519 137274321021824 train.py:125] NOTE: Steps:61150/112603 [54.3%]
+Walltime:4d10h42m (0s eval)
+ETA:3d17h45m
+Total train time:8d4h26m
+I1202 08:38:03.965185 137274321021824 utils.py:1231] [61200] l2_params = 293.23328624420975
+I1202 08:38:03.965408 137274321021824 utils.py:1231] [61200] train/loss = 2.2798406183719635
+I1202 08:38:03.965508 137274321021824 utils.py:1231] [61200] l2_grads = 1.6510971784591675
+I1202 08:38:03.965576 137274321021824 utils.py:1231] [61200] lr = 0.0005015692168765335
+I1202 08:38:03.965646 137274321021824 utils.py:1231] [61200] uptime = 384473.328008001
+I1202 08:38:03.965717 137274321021824 utils.py:1231] [61200] examples_seen = 62668800.0
+I1202 08:38:03.965773 137274321021824 utils.py:1231] [61200] progress = 0.5435023933642976
+I1202 08:38:03.965827 137274321021824 utils.py:1231] [61200] epoch = 48.915402910003145
+I1202 08:38:03.965887 137274321021824 utils.py:1231] [61200] img/sec/core = 164.22808635400526
+I1202 08:38:03.965953 137274321021824 utils.py:1231] [61200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.7638290665325
+I1202 08:38:03.966007 137274321021824 utils.py:1231] [61200] core_hours = 106.7638290665325
+I1202 08:38:03.966074 137274321021824 train.py:125] NOTE: Steps:61200/112603 [54.4%]
+Walltime:4d10h47m (0s eval)
+ETA:3d17h40m
+Total train time:8d4h26m
+I1202 08:43:15.734161 137274321021824 utils.py:1231] [61250] l2_params = 293.13036511194247
+I1202 08:43:15.734389 137274321021824 utils.py:1231] [61250] train/loss = 2.8315942585468292
+I1202 08:43:15.734510 137274321021824 utils.py:1231] [61250] l2_grads = 1.4310754537582397
+I1202 08:43:15.734580 137274321021824 utils.py:1231] [61250] lr = 0.0005008037462027497
+I1202 08:43:15.734638 137274321021824 utils.py:1231] [61250] uptime = 384785.09699540096
+I1202 08:43:15.734703 137274321021824 utils.py:1231] [61250] examples_seen = 62720000.0
+I1202 08:43:15.734750 137274321021824 utils.py:1231] [61250] progress = 0.543946431267373
+I1202 08:43:15.734798 137274321021824 utils.py:1231] [61250] epoch = 48.9553664744721
+I1202 08:43:15.734849 137274321021824 utils.py:1231] [61250] img/sec/core = 164.2241597760707
+I1202 08:43:15.734913 137274321021824 utils.py:1231] [61250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.85043156303249
+I1202 08:43:15.734971 137274321021824 utils.py:1231] [61250] core_hours = 106.85043156303249
+I1202 08:43:15.735037 137274321021824 train.py:125] NOTE: Steps:61250/112603 [54.4%]
+Walltime:4d10h53m (0s eval)
+ETA:3d17h35m
+Total train time:8d4h26m
+I1202 08:48:27.509420 137274321021824 utils.py:1231] [61300] l2_params = 293.048543140603
+I1202 08:48:27.509671 137274321021824 utils.py:1231] [61300] train/loss = 2.658630609512329
+I1202 08:48:27.509790 137274321021824 utils.py:1231] [61300] l2_grads = 1.6295783519744873
+I1202 08:48:27.509896 137274321021824 utils.py:1231] [61300] lr = 0.0005000382736451501
+I1202 08:48:27.509969 137274321021824 utils.py:1231] [61300] uptime = 385096.872331732
+I1202 08:48:27.510019 137274321021824 utils.py:1231] [61300] examples_seen = 62771200.0
+I1202 08:48:27.510065 137274321021824 utils.py:1231] [61300] progress = 0.5443904691704484
+I1202 08:48:27.510110 137274321021824 utils.py:1231] [61300] epoch = 48.995330038941056
+I1202 08:48:27.510159 137274321021824 utils.py:1231] [61300] img/sec/core = 164.22081554788252
+I1202 08:48:27.510212 137274321021824 utils.py:1231] [61300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 106.93703582312445
+I1202 08:48:27.510280 137274321021824 utils.py:1231] [61300] core_hours = 106.93703582312445
+I1202 08:48:27.510366 137274321021824 train.py:125] NOTE: Steps:61300/112603 [54.4%]
+Walltime:4d10h58m (0s eval)
+ETA:3d17h30m
+Total train time:8d4h26m
+I1202 08:53:39.477913 137274321021824 utils.py:1231] [61350] l2_params = 292.952281247425
+I1202 08:53:39.478224 137274321021824 utils.py:1231] [61350] train/loss = 4.446922779083252
+I1202 08:53:39.478455 137274321021824 utils.py:1231] [61350] l2_grads = 1.6013423204421997
+I1202 08:53:39.478560 137274321021824 utils.py:1231] [61350] lr = 0.0004992728009978449
+I1202 08:53:39.478647 137274321021824 utils.py:1231] [61350] uptime = 385408.841004009
+I1202 08:53:39.478739 137274321021824 utils.py:1231] [61350] examples_seen = 62822400.0
+I1202 08:53:39.478850 137274321021824 utils.py:1231] [61350] progress = 0.5448345070735238
+I1202 08:53:39.478941 137274321021824 utils.py:1231] [61350] epoch = 49.03529360341002
+I1202 08:53:39.479014 137274321021824 utils.py:1231] [61350] img/sec/core = 164.11904319206542
+I1202 08:53:39.479083 137274321021824 utils.py:1231] [61350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.02369378764584
+I1202 08:53:39.479151 137274321021824 utils.py:1231] [61350] core_hours = 107.02369378764584
+I1202 08:53:39.479227 137274321021824 train.py:125] NOTE: Steps:61350/112603 [54.5%]
+Walltime:4d11h3m (0s eval)
+ETA:3d17h24m
+Total train time:8d4h26m
+I1202 08:58:51.245420 137274321021824 utils.py:1231] [61400] l2_params = 292.8676661735447
+I1202 08:58:51.245627 137274321021824 utils.py:1231] [61400] train/loss = 4.016349911689758
+I1202 08:58:51.245731 137274321021824 utils.py:1231] [61400] l2_grads = 1.4899168014526367
+I1202 08:58:51.245799 137274321021824 utils.py:1231] [61400] lr = 0.0004985073300549449
+I1202 08:58:51.245861 137274321021824 utils.py:1231] [61400] uptime = 385720.608221097
+I1202 08:58:51.245925 137274321021824 utils.py:1231] [61400] examples_seen = 62873600.0
+I1202 08:58:51.245993 137274321021824 utils.py:1231] [61400] progress = 0.5452785449765992
+I1202 08:58:51.246048 137274321021824 utils.py:1231] [61400] epoch = 49.07525716787897
+I1202 08:58:51.246104 137274321021824 utils.py:1231] [61400] img/sec/core = 164.22509229234848
+I1202 08:58:51.246161 137274321021824 utils.py:1231] [61400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.11029579239249
+I1202 08:58:51.246215 137274321021824 utils.py:1231] [61400] core_hours = 107.11029579239249
+I1202 08:58:51.246282 137274321021824 train.py:125] NOTE: Steps:61400/112603 [54.5%]
+Walltime:4d11h8m (0s eval)
+ETA:3d17h19m
+Total train time:8d4h26m
+I1202 09:04:02.396405 137274321021824 utils.py:1231] [61450] l2_params = 292.79386998617355
+I1202 09:04:02.396686 137274321021824 utils.py:1231] [61450] train/loss = 3.2056581377983093
+I1202 09:04:02.396815 137274321021824 utils.py:1231] [61450] l2_grads = 1.4795405864715576
+I1202 09:04:02.396911 137274321021824 utils.py:1231] [61450] lr = 0.0004977418626105569
+I1202 09:04:02.396975 137274321021824 utils.py:1231] [61450] uptime = 386031.759336147
+I1202 09:04:02.397036 137274321021824 utils.py:1231] [61450] examples_seen = 62924800.0
+I1202 09:04:02.397093 137274321021824 utils.py:1231] [61450] progress = 0.5457225828796746
+I1202 09:04:02.397153 137274321021824 utils.py:1231] [61450] epoch = 49.11522073234793
+I1202 09:04:02.397219 137274321021824 utils.py:1231] [61450] img/sec/core = 164.5502700248075
+I1202 09:04:02.397299 137274321021824 utils.py:1231] [61450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.19672665768417
+I1202 09:04:02.397371 137274321021824 utils.py:1231] [61450] core_hours = 107.19672665768417
+I1202 09:04:02.397443 137274321021824 train.py:125] NOTE: Steps:61450/112603 [54.6%]
+Walltime:4d11h13m (0s eval)
+ETA:3d17h14m
+Total train time:8d4h26m
+I1202 09:09:12.909080 137274321021824 utils.py:1231] [61500] l2_params = 292.70952062024367
+I1202 09:09:12.909335 137274321021824 utils.py:1231] [61500] train/loss = 3.0824449360370636
+I1202 09:09:12.909514 137274321021824 utils.py:1231] [61500] l2_grads = 1.5425076484680176
+I1202 09:09:12.909609 137274321021824 utils.py:1231] [61500] lr = 0.0004969764004587793
+I1202 09:09:12.909703 137274321021824 utils.py:1231] [61500] uptime = 386342.272064813
+I1202 09:09:12.909767 137274321021824 utils.py:1231] [61500] examples_seen = 62976000.0
+I1202 09:09:12.909822 137274321021824 utils.py:1231] [61500] progress = 0.54616662078275
+I1202 09:09:12.909879 137274321021824 utils.py:1231] [61500] epoch = 49.155184296816884
+I1202 09:09:12.909957 137274321021824 utils.py:1231] [61500] img/sec/core = 164.88857065524854
+I1202 09:09:12.910013 137274321021824 utils.py:1231] [61500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.28298019342472
+I1202 09:09:12.910071 137274321021824 utils.py:1231] [61500] core_hours = 107.28298019342472
+I1202 09:09:12.910147 137274321021824 train.py:125] NOTE: Steps:61500/112603 [54.6%]
+Walltime:4d11h19m (0s eval)
+ETA:3d17h8m
+Total train time:8d4h26m
+I1202 09:14:24.691350 137274321021824 utils.py:1231] [61550] l2_params = 292.610849944387
+I1202 09:14:24.691534 137274321021824 utils.py:1231] [61550] train/loss = 3.106774866580963
+I1202 09:14:24.691625 137274321021824 utils.py:1231] [61550] l2_grads = 1.4575212001800537
+I1202 09:14:24.691689 137274321021824 utils.py:1231] [61550] lr = 0.0004962109453936971
+I1202 09:14:24.691739 137274321021824 utils.py:1231] [61550] uptime = 386654.054101197
+I1202 09:14:24.691790 137274321021824 utils.py:1231] [61550] examples_seen = 63027200.0
+I1202 09:14:24.691838 137274321021824 utils.py:1231] [61550] progress = 0.5466106586858254
+I1202 09:14:24.691888 137274321021824 utils.py:1231] [61550] epoch = 49.19514786128585
+I1202 09:14:24.691946 137274321021824 utils.py:1231] [61550] img/sec/core = 164.2172865178895
+I1202 09:14:24.691999 137274321021824 utils.py:1231] [61550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.36958631464249
+I1202 09:14:24.692050 137274321021824 utils.py:1231] [61550] core_hours = 107.36958631464249
+I1202 09:14:24.692108 137274321021824 train.py:125] NOTE: Steps:61550/112603 [54.7%]
+Walltime:4d11h24m (0s eval)
+ETA:3d17h3m
+Total train time:8d4h26m
+I1202 09:19:36.465502 137274321021824 utils.py:1231] [61600] l2_params = 292.51857298771546
+I1202 09:19:36.465717 137274321021824 utils.py:1231] [61600] train/loss = 2.3057554960250854
+I1202 09:19:36.465903 137274321021824 utils.py:1231] [61600] l2_grads = 1.6051260232925415
+I1202 09:19:36.465988 137274321021824 utils.py:1231] [61600] lr = 0.0004954454992093808
+I1202 09:19:36.466040 137274321021824 utils.py:1231] [61600] uptime = 386965.82840317796
+I1202 09:19:36.466089 137274321021824 utils.py:1231] [61600] examples_seen = 63078400.0
+I1202 09:19:36.466134 137274321021824 utils.py:1231] [61600] progress = 0.5470546965889008
+I1202 09:19:36.466179 137274321021824 utils.py:1231] [61600] epoch = 49.2351114257548
+I1202 09:19:36.466226 137274321021824 utils.py:1231] [61600] img/sec/core = 164.2213603708907
+I1202 09:19:36.466279 137274321021824 utils.py:1231] [61600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.45619028741498
+I1202 09:19:36.466326 137274321021824 utils.py:1231] [61600] core_hours = 107.45619028741498
+I1202 09:19:36.466383 137274321021824 train.py:125] NOTE: Steps:61600/112603 [54.7%]
+Walltime:4d11h29m (0s eval)
+ETA:3d16h58m
+Total train time:8d4h25m
+I1202 09:24:46.841279 137274321021824 utils.py:1231] [61650] l2_params = 292.4100295759523
+I1202 09:24:46.841519 137274321021824 utils.py:1231] [61650] train/loss = 2.7638562619686127
+I1202 09:24:46.841618 137274321021824 utils.py:1231] [61650] l2_grads = 1.5199111700057983
+I1202 09:24:46.841688 137274321021824 utils.py:1231] [61650] lr = 0.0004946800636998787
+I1202 09:24:46.841756 137274321021824 utils.py:1231] [61650] uptime = 387276.204118263
+I1202 09:24:46.841824 137274321021824 utils.py:1231] [61650] examples_seen = 63129600.0
+I1202 09:24:46.841871 137274321021824 utils.py:1231] [61650] progress = 0.5474987344919763
+I1202 09:24:46.841923 137274321021824 utils.py:1231] [61650] epoch = 49.27507499022376
+I1202 09:24:46.841972 137274321021824 utils.py:1231] [61650] img/sec/core = 164.96135976996763
+I1202 09:24:46.842025 137274321021824 utils.py:1231] [61650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.54240576382749
+I1202 09:24:46.842075 137274321021824 utils.py:1231] [61650] core_hours = 107.54240576382749
+I1202 09:24:46.842135 137274321021824 train.py:125] NOTE: Steps:61650/112603 [54.7%]
+Walltime:4d11h34m (0s eval)
+ETA:3d16h53m
+Total train time:8d4h25m
+I1202 09:29:58.610554 137274321021824 utils.py:1231] [61700] l2_params = 292.32177844195974
+I1202 09:29:58.610771 137274321021824 utils.py:1231] [61700] train/loss = 2.164633870124817
+I1202 09:29:58.610861 137274321021824 utils.py:1231] [61700] l2_grads = 1.7466486692428589
+I1202 09:29:58.610950 137274321021824 utils.py:1231] [61700] lr = 0.0004939146406592148
+I1202 09:29:58.611000 137274321021824 utils.py:1231] [61700] uptime = 387587.97336254996
+I1202 09:29:58.611050 137274321021824 utils.py:1231] [61700] examples_seen = 63180800.0
+I1202 09:29:58.611097 137274321021824 utils.py:1231] [61700] progress = 0.5479427723950516
+I1202 09:29:58.611143 137274321021824 utils.py:1231] [61700] epoch = 49.31503855469271
+I1202 09:29:58.611193 137274321021824 utils.py:1231] [61700] img/sec/core = 164.22402446108381
+I1202 09:29:58.611245 137274321021824 utils.py:1231] [61700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.629008331685
+I1202 09:29:58.611292 137274321021824 utils.py:1231] [61700] core_hours = 107.629008331685
+I1202 09:29:58.611349 137274321021824 train.py:125] NOTE: Steps:61700/112603 [54.8%]
+Walltime:4d11h39m (0s eval)
+ETA:3d16h47m
+Total train time:8d4h25m
+I1202 09:35:10.383011 137274321021824 utils.py:1231] [61750] l2_params = 292.2353962251852
+I1202 09:35:10.383278 137274321021824 utils.py:1231] [61750] train/loss = 2.404238283634186
+I1202 09:35:10.383403 137274321021824 utils.py:1231] [61750] l2_grads = 1.699280858039856
+I1202 09:35:10.383495 137274321021824 utils.py:1231] [61750] lr = 0.0004931492318813837
+I1202 09:35:10.383574 137274321021824 utils.py:1231] [61750] uptime = 387899.745929253
+I1202 09:35:10.383647 137274321021824 utils.py:1231] [61750] examples_seen = 63232000.0
+I1202 09:35:10.383714 137274321021824 utils.py:1231] [61750] progress = 0.5483868102981271
+I1202 09:35:10.383769 137274321021824 utils.py:1231] [61750] epoch = 49.35500211916167
+I1202 09:35:10.383825 137274321021824 utils.py:1231] [61750] img/sec/core = 164.2222744016115
+I1202 09:35:10.383916 137274321021824 utils.py:1231] [61750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.71561182243583
+I1202 09:35:10.383984 137274321021824 utils.py:1231] [61750] core_hours = 107.71561182243583
+I1202 09:35:10.384049 137274321021824 train.py:125] NOTE: Steps:61750/112603 [54.8%]
+Walltime:4d11h44m (0s eval)
+ETA:3d16h42m
+Total train time:8d4h25m
+I1202 09:40:21.267576 137274321021824 utils.py:1231] [61800] l2_params = 292.15211927437934
+I1202 09:40:21.267832 137274321021824 utils.py:1231] [61800] train/loss = 4.041780233383179
+I1202 09:40:21.267938 137274321021824 utils.py:1231] [61800] l2_grads = 1.393593430519104
+I1202 09:40:21.268005 137274321021824 utils.py:1231] [61800] lr = 0.0004923838391603458
+I1202 09:40:21.268064 137274321021824 utils.py:1231] [61800] uptime = 388210.630425484
+I1202 09:40:21.268122 137274321021824 utils.py:1231] [61800] examples_seen = 63283200.0
+I1202 09:40:21.268177 137274321021824 utils.py:1231] [61800] progress = 0.5488308482012024
+I1202 09:40:21.268231 137274321021824 utils.py:1231] [61800] epoch = 49.39496568363063
+I1202 09:40:21.268287 137274321021824 utils.py:1231] [61800] img/sec/core = 164.6913905991612
+I1202 09:40:21.268357 137274321021824 utils.py:1231] [61800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.80196862694444
+I1202 09:40:21.268412 137274321021824 utils.py:1231] [61800] core_hours = 107.80196862694444
+I1202 09:40:21.268476 137274321021824 train.py:125] NOTE: Steps:61800/112603 [54.9%]
+Walltime:4d11h50m (0s eval)
+ETA:3d16h37m
+Total train time:8d4h25m
+I1202 09:45:31.982828 137274321021824 utils.py:1231] [61850] l2_params = 292.05539948003405
+I1202 09:45:31.983088 137274321021824 utils.py:1231] [61850] train/loss = 2.1774661540985107
+I1202 09:45:31.983223 137274321021824 utils.py:1231] [61850] l2_grads = 1.7166248559951782
+I1202 09:45:31.983309 137274321021824 utils.py:1231] [61850] lr = 0.0004916184642900242
+I1202 09:45:31.983402 137274321021824 utils.py:1231] [61850] uptime = 388521.345762925
+I1202 09:45:31.983468 137274321021824 utils.py:1231] [61850] examples_seen = 63334400.0
+I1202 09:45:31.983528 137274321021824 utils.py:1231] [61850] progress = 0.5492748861042779
+I1202 09:45:31.983586 137274321021824 utils.py:1231] [61850] epoch = 49.434929248099586
+I1202 09:45:31.983661 137274321021824 utils.py:1231] [61850] img/sec/core = 164.78105143335623
+I1202 09:45:31.983721 137274321021824 utils.py:1231] [61850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.88827844290027
+I1202 09:45:31.983782 137274321021824 utils.py:1231] [61850] core_hours = 107.88827844290027
+I1202 09:45:31.983845 137274321021824 train.py:125] NOTE: Steps:61850/112603 [54.9%]
+Walltime:4d11h55m (0s eval)
+ETA:3d16h32m
+Total train time:8d4h25m
+I1202 09:50:43.769707 137274321021824 utils.py:1231] [61900] l2_params = 291.9572987418129
+I1202 09:50:43.769969 137274321021824 utils.py:1231] [61900] train/loss = 2.3416420221328735
+I1202 09:50:43.770105 137274321021824 utils.py:1231] [61900] l2_grads = 1.6724079847335815
+I1202 09:50:43.770200 137274321021824 utils.py:1231] [61900] lr = 0.000490853109064301
+I1202 09:50:43.770282 137274321021824 utils.py:1231] [61900] uptime = 388833.13263432
+I1202 09:50:43.770364 137274321021824 utils.py:1231] [61900] examples_seen = 63385600.0
+I1202 09:50:43.770435 137274321021824 utils.py:1231] [61900] progress = 0.5497189240073532
+I1202 09:50:43.770503 137274321021824 utils.py:1231] [61900] epoch = 49.47489281256854
+I1202 09:50:43.770565 137274321021824 utils.py:1231] [61900] img/sec/core = 164.21473993089532
+I1202 09:50:43.770646 137274321021824 utils.py:1231] [61900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 107.97488590717667
+I1202 09:50:43.770706 137274321021824 utils.py:1231] [61900] core_hours = 107.97488590717667
+I1202 09:50:43.770776 137274321021824 train.py:125] NOTE: Steps:61900/112603 [55.0%]
+Walltime:4d12h0m (0s eval)
+ETA:3d16h26m
+Total train time:8d4h25m
+I1202 09:55:55.560780 137274321021824 utils.py:1231] [61950] l2_params = 291.8812584188687
+I1202 09:55:55.560981 137274321021824 utils.py:1231] [61950] train/loss = 4.189966380596161
+I1202 09:55:55.561094 137274321021824 utils.py:1231] [61950] l2_grads = 1.5985307693481445
+I1202 09:55:55.561156 137274321021824 utils.py:1231] [61950] lr = 0.0004900877752770113
+I1202 09:55:55.561208 137274321021824 utils.py:1231] [61950] uptime = 389144.923570437
+I1202 09:55:55.561263 137274321021824 utils.py:1231] [61950] examples_seen = 63436800.0
+I1202 09:55:55.561312 137274321021824 utils.py:1231] [61950] progress = 0.5501629619104287
+I1202 09:55:55.561366 137274321021824 utils.py:1231] [61950] epoch = 49.514856377037496
+I1202 09:55:55.561418 137274321021824 utils.py:1231] [61950] img/sec/core = 164.21259911412804
+I1202 09:55:55.561475 137274321021824 utils.py:1231] [61950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.0614945005425
+I1202 09:55:55.561528 137274321021824 utils.py:1231] [61950] core_hours = 108.0614945005425
+I1202 09:55:55.561592 137274321021824 train.py:125] NOTE: Steps:61950/112603 [55.0%]
+Walltime:4d12h5m (0s eval)
+ETA:3d16h21m
+Total train time:8d4h25m
+I1202 10:01:07.335114 137274321021824 utils.py:1231] [62000] l2_params = 291.7907110624877
+I1202 10:01:07.335354 137274321021824 utils.py:1231] [62000] train/loss = 2.1778292506933212
+I1202 10:01:07.335477 137274321021824 utils.py:1231] [62000] l2_grads = 1.7449300289154053
+I1202 10:01:07.335569 137274321021824 utils.py:1231] [62000] lr = 0.0004893224647219407
+I1202 10:01:07.335633 137274321021824 utils.py:1231] [62000] uptime = 389456.69799519697
+I1202 10:01:07.335711 137274321021824 utils.py:1231] [62000] examples_seen = 63488000.0
+I1202 10:01:07.335769 137274321021824 utils.py:1231] [62000] progress = 0.550606999813504
+I1202 10:01:07.335823 137274321021824 utils.py:1231] [62000] epoch = 49.55481994150645
+I1202 10:01:07.335884 137274321021824 utils.py:1231] [62000] img/sec/core = 164.22129569934842
+I1202 10:01:07.335956 137274321021824 utils.py:1231] [62000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.14809850742026
+I1202 10:01:07.336009 137274321021824 utils.py:1231] [62000] core_hours = 108.14809850742026
+I1202 10:01:07.336072 137274321021824 train.py:125] NOTE: Steps:62000/112603 [55.1%]
+Walltime:4d12h10m (0s eval)
+ETA:3d16h16m
+Total train time:8d4h25m
+I1202 10:06:19.479408 137274321021824 utils.py:1231] [62050] l2_params = 291.69366651800584
+I1202 10:06:19.479609 137274321021824 utils.py:1231] [62050] train/loss = 3.7980493307113647
+I1202 10:06:19.479712 137274321021824 utils.py:1231] [62050] l2_grads = 1.4496763944625854
+I1202 10:06:19.479786 137274321021824 utils.py:1231] [62050] lr = 0.0004885571791928197
+I1202 10:06:19.479846 137274321021824 utils.py:1231] [62050] uptime = 389768.842207625
+I1202 10:06:19.479925 137274321021824 utils.py:1231] [62050] examples_seen = 63539200.0
+I1202 10:06:19.479973 137274321021824 utils.py:1231] [62050] progress = 0.5510510377165795
+I1202 10:06:19.480020 137274321021824 utils.py:1231] [62050] epoch = 49.594783505975414
+I1202 10:06:19.480068 137274321021824 utils.py:1231] [62050] img/sec/core = 164.02674777065587
+I1202 10:06:19.480119 137274321021824 utils.py:1231] [62050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.23480523309472
+I1202 10:06:19.480167 137274321021824 utils.py:1231] [62050] core_hours = 108.23480523309472
+I1202 10:06:19.480224 137274321021824 train.py:125] NOTE: Steps:62050/112603 [55.1%]
+Walltime:4d12h16m (0s eval)
+ETA:3d16h10m
+Total train time:8d4h25m
+I1202 10:11:31.248672 137274321021824 utils.py:1231] [62100] l2_params = 291.6124817634115
+I1202 10:11:31.248901 137274321021824 utils.py:1231] [62100] train/loss = 2.7187952995300293
+I1202 10:11:31.249032 137274321021824 utils.py:1231] [62100] l2_grads = 1.6731083393096924
+I1202 10:11:31.249121 137274321021824 utils.py:1231] [62100] lr = 0.0004877919204833206
+I1202 10:11:31.249207 137274321021824 utils.py:1231] [62100] uptime = 390080.611564292
+I1202 10:11:31.249277 137274321021824 utils.py:1231] [62100] examples_seen = 63590400.0
+I1202 10:11:31.249341 137274321021824 utils.py:1231] [62100] progress = 0.5514950756196549
+I1202 10:11:31.249397 137274321021824 utils.py:1231] [62100] epoch = 49.63474707044437
+I1202 10:11:31.249460 137274321021824 utils.py:1231] [62100] img/sec/core = 164.2239652650841
+I1202 10:11:31.249529 137274321021824 utils.py:1231] [62100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.32140783216889
+I1202 10:11:31.249591 137274321021824 utils.py:1231] [62100] core_hours = 108.32140783216889
+I1202 10:11:31.249660 137274321021824 train.py:125] NOTE: Steps:62100/112603 [55.1%]
+Walltime:4d12h21m (0s eval)
+ETA:3d16h5m
+Total train time:8d4h25m
+I1202 10:16:43.024055 137274321021824 utils.py:1231] [62150] l2_params = 291.5286379307747
+I1202 10:16:43.024251 137274321021824 utils.py:1231] [62150] train/loss = 2.7013773024082184
+I1202 10:16:43.024357 137274321021824 utils.py:1231] [62150] l2_grads = 1.5559464693069458
+I1202 10:16:43.024425 137274321021824 utils.py:1231] [62150] lr = 0.0004870266903870524
+I1202 10:16:43.024490 137274321021824 utils.py:1231] [62150] uptime = 390392.386851051
+I1202 10:16:43.024563 137274321021824 utils.py:1231] [62150] examples_seen = 63641600.0
+I1202 10:16:43.024636 137274321021824 utils.py:1231] [62150] progress = 0.5519391135227303
+I1202 10:16:43.024707 137274321021824 utils.py:1231] [62150] epoch = 49.674710634913325
+I1202 10:16:43.024797 137274321021824 utils.py:1231] [62150] img/sec/core = 164.22084165886574
+I1202 10:16:43.024888 137274321021824 utils.py:1231] [62150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.40801207849083
+I1202 10:16:43.024952 137274321021824 utils.py:1231] [62150] core_hours = 108.40801207849083
+I1202 10:16:43.025015 137274321021824 train.py:125] NOTE: Steps:62150/112603 [55.2%]
+Walltime:4d12h26m (0s eval)
+ETA:3d16h0m
+Total train time:8d4h25m
+I1202 10:21:53.429455 137274321021824 utils.py:1231] [62200] l2_params = 291.4342106599371
+I1202 10:21:53.429670 137274321021824 utils.py:1231] [62200] train/loss = 4.482192695140839
+I1202 10:21:53.429772 137274321021824 utils.py:1231] [62200] l2_grads = 1.4957304000854492
+I1202 10:21:53.429846 137274321021824 utils.py:1231] [62200] lr = 0.00048626149069755706
+I1202 10:21:53.429911 137274321021824 utils.py:1231] [62200] uptime = 390702.79227185197
+I1202 10:21:53.429971 137274321021824 utils.py:1231] [62200] examples_seen = 63692800.0
+I1202 10:21:53.430027 137274321021824 utils.py:1231] [62200] progress = 0.5523831514258057
+I1202 10:21:53.430083 137274321021824 utils.py:1231] [62200] epoch = 49.71467419938228
+I1202 10:21:53.430141 137274321021824 utils.py:1231] [62200] img/sec/core = 164.9455730118529
+I1202 10:21:53.430207 137274321021824 utils.py:1231] [62200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.4942358064911
+I1202 10:21:53.430259 137274321021824 utils.py:1231] [62200] core_hours = 108.4942358064911
+I1202 10:21:53.430320 137274321021824 train.py:125] NOTE: Steps:62200/112603 [55.2%]
+Walltime:4d12h31m (0s eval)
+ETA:3d15h55m
+Total train time:8d4h25m
+I1202 10:27:03.053510 137274321021824 utils.py:1231] [62250] l2_params = 291.34510639085244
+I1202 10:27:03.053724 137274321021824 utils.py:1231] [62250] train/loss = 2.2947084605693817
+I1202 10:27:03.053826 137274321021824 utils.py:1231] [62250] l2_grads = 1.6121770143508911
+I1202 10:27:03.053902 137274321021824 utils.py:1231] [62250] lr = 0.00048549632320830574
+I1202 10:27:03.053965 137274321021824 utils.py:1231] [62250] uptime = 391012.41632523
+I1202 10:27:03.054025 137274321021824 utils.py:1231] [62250] examples_seen = 63744000.0
+I1202 10:27:03.054083 137274321021824 utils.py:1231] [62250] progress = 0.5528271893288811
+I1202 10:27:03.054139 137274321021824 utils.py:1231] [62250] epoch = 49.754637763851235
+I1202 10:27:03.054198 137274321021824 utils.py:1231] [62250] img/sec/core = 165.36182974612765
+I1202 10:27:03.054259 137274321021824 utils.py:1231] [62250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.58024248798499
+I1202 10:27:03.054316 137274321021824 utils.py:1231] [62250] core_hours = 108.58024248798499
+I1202 10:27:03.054379 137274321021824 train.py:125] NOTE: Steps:62250/112603 [55.3%]
+Walltime:4d12h36m (0s eval)
+ETA:3d15h49m
+Total train time:8d4h24m
+I1202 10:32:14.834675 137274321021824 utils.py:1231] [62300] l2_params = 291.24662285144274
+I1202 10:32:14.834913 137274321021824 utils.py:1231] [62300] train/loss = 2.32657727599144
+I1202 10:32:14.835043 137274321021824 utils.py:1231] [62300] l2_grads = 1.7068153619766235
+I1202 10:32:14.835100 137274321021824 utils.py:1231] [62300] lr = 0.0004847311897126943
+I1202 10:32:14.835149 137274321021824 utils.py:1231] [62300] uptime = 391324.19751131
+I1202 10:32:14.835199 137274321021824 utils.py:1231] [62300] examples_seen = 63795200.0
+I1202 10:32:14.835247 137274321021824 utils.py:1231] [62300] progress = 0.5532712272319565
+I1202 10:32:14.835292 137274321021824 utils.py:1231] [62300] epoch = 49.7946013283202
+I1202 10:32:14.835349 137274321021824 utils.py:1231] [62300] img/sec/core = 164.21773437881626
+I1202 10:32:14.835406 137274321021824 utils.py:1231] [62300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.66684837300723
+I1202 10:32:14.835453 137274321021824 utils.py:1231] [62300] core_hours = 108.66684837300723
+I1202 10:32:14.835510 137274321021824 train.py:125] NOTE: Steps:62300/112603 [55.3%]
+Walltime:4d12h42m (0s eval)
+ETA:3d15h44m
+Total train time:8d4h24m
+I1202 10:37:25.401803 137274321021824 utils.py:1231] [62350] l2_params = 291.17152309248047
+I1202 10:37:25.402077 137274321021824 utils.py:1231] [62350] train/loss = 3.4533610939979553
+I1202 10:37:25.402307 137274321021824 utils.py:1231] [62350] l2_grads = 1.443175196647644
+I1202 10:37:25.402406 137274321021824 utils.py:1231] [62350] lr = 0.0004839660920040378
+I1202 10:37:25.402471 137274321021824 utils.py:1231] [62350] uptime = 391634.764824545
+I1202 10:37:25.402525 137274321021824 utils.py:1231] [62350] examples_seen = 63846400.0
+I1202 10:37:25.402578 137274321021824 utils.py:1231] [62350] progress = 0.5537152651350319
+I1202 10:37:25.402635 137274321021824 utils.py:1231] [62350] epoch = 49.83456489278915
+I1202 10:37:25.402691 137274321021824 utils.py:1231] [62350] img/sec/core = 164.85959023400076
+I1202 10:37:25.402750 137274321021824 utils.py:1231] [62350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.75311707112806
+I1202 10:37:25.402817 137274321021824 utils.py:1231] [62350] core_hours = 108.75311707112806
+I1202 10:37:25.402879 137274321021824 train.py:125] NOTE: Steps:62350/112603 [55.4%]
+Walltime:4d12h47m (0s eval)
+ETA:3d15h39m
+Total train time:8d4h24m
+I1202 10:42:35.605496 137274321021824 utils.py:1231] [62400] l2_params = 291.08033451253505
+I1202 10:42:35.605740 137274321021824 utils.py:1231] [62400] train/loss = 2.2096824049949646
+I1202 10:42:35.605859 137274321021824 utils.py:1231] [62400] l2_grads = 1.7892875671386719
+I1202 10:42:35.605933 137274321021824 utils.py:1231] [62400] lr = 0.00048320103187556826
+I1202 10:42:35.605991 137274321021824 utils.py:1231] [62400] uptime = 391944.96835354296
+I1202 10:42:35.606047 137274321021824 utils.py:1231] [62400] examples_seen = 63897600.0
+I1202 10:42:35.606094 137274321021824 utils.py:1231] [62400] progress = 0.5541593030381073
+I1202 10:42:35.606141 137274321021824 utils.py:1231] [62400] epoch = 49.87452845725811
+I1202 10:42:35.606191 137274321021824 utils.py:1231] [62400] img/sec/core = 165.05292562399475
+I1202 10:42:35.606245 137274321021824 utils.py:1231] [62400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.83928471807192
+I1202 10:42:35.606295 137274321021824 utils.py:1231] [62400] core_hours = 108.83928471807192
+I1202 10:42:35.606355 137274321021824 train.py:125] NOTE: Steps:62400/112603 [55.4%]
+Walltime:4d12h52m (0s eval)
+ETA:3d15h34m
+Total train time:8d4h24m
+I1202 10:47:47.397239 137274321021824 utils.py:1231] [62450] l2_params = 290.9802108253247
+I1202 10:47:47.397552 137274321021824 utils.py:1231] [62450] train/loss = 2.2511297166347504
+I1202 10:47:47.397730 137274321021824 utils.py:1231] [62450] l2_grads = 1.7391233444213867
+I1202 10:47:47.397805 137274321021824 utils.py:1231] [62450] lr = 0.00048243601112042976
+I1202 10:47:47.397862 137274321021824 utils.py:1231] [62450] uptime = 392256.760224975
+I1202 10:47:47.397924 137274321021824 utils.py:1231] [62450] examples_seen = 63948800.0
+I1202 10:47:47.397971 137274321021824 utils.py:1231] [62450] progress = 0.5546033409411827
+I1202 10:47:47.398017 137274321021824 utils.py:1231] [62450] epoch = 49.914492021727064
+I1202 10:47:47.398064 137274321021824 utils.py:1231] [62450] img/sec/core = 164.2121065082346
+I1202 10:47:47.398118 137274321021824 utils.py:1231] [62450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 108.9258935712475
+I1202 10:47:47.398167 137274321021824 utils.py:1231] [62450] core_hours = 108.9258935712475
+I1202 10:47:47.398226 137274321021824 train.py:125] NOTE: Steps:62450/112603 [55.5%]
+Walltime:4d12h57m (0s eval)
+ETA:3d15h28m
+Total train time:8d4h24m
+I1202 10:52:59.187544 137274321021824 utils.py:1231] [62500] l2_params = 290.87784156345765
+I1202 10:52:59.187864 137274321021824 utils.py:1231] [62500] train/loss = 4.159871757030487
+I1202 10:52:59.188028 137274321021824 utils.py:1231] [62500] l2_grads = 1.47421133518219
+I1202 10:52:59.188088 137274321021824 utils.py:1231] [62500] lr = 0.0004816710315316735
+I1202 10:52:59.188139 137274321021824 utils.py:1231] [62500] uptime = 392568.550501297
+I1202 10:52:59.188196 137274321021824 utils.py:1231] [62500] examples_seen = 64000000.0
+I1202 10:52:59.188250 137274321021824 utils.py:1231] [62500] progress = 0.5550473788442581
+I1202 10:52:59.188297 137274321021824 utils.py:1231] [62500] epoch = 49.954455586196026
+I1202 10:52:59.188352 137274321021824 utils.py:1231] [62500] img/sec/core = 164.2129466126297
+I1202 10:52:59.188422 137274321021824 utils.py:1231] [62500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.01250198133694
+I1202 10:52:59.188471 137274321021824 utils.py:1231] [62500] core_hours = 109.01250198133694
+I1202 10:52:59.188561 137274321021824 train.py:125] NOTE: Steps:62500/112603 [55.5%]
+Walltime:4d13h2m (0s eval)
+ETA:3d15h23m
+Total train time:8d4h24m
+I1202 10:52:59.188669 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:62500/112603 [55.5%]
+Walltime:4d13h2m (0s eval)
+ETA:3d15h23m
+Total train time:8d4h24m
+I1202 10:54:34.147052 137274321021824 utils.py:1231] [62500] val/acc@1 = 0.6760403380102041
+I1202 10:54:34.147296 137274321021824 utils.py:1231] [62500] val/loss = 1.3180131820999845
+I1202 10:54:34.147443 137274321021824 utils.py:1231] [62500] z/secs/eval/val = 94.95870455401018
+I1202 10:54:34.147516 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 94.95870455401018
+I1202 10:59:35.228159 137274321021824 utils.py:1231] [62550] l2_params = 290.79410862540726
+I1202 10:59:35.228484 137274321021824 utils.py:1231] [62550] train/loss = 2.451757997274399
+I1202 10:59:35.228650 137274321021824 utils.py:1231] [62550] l2_grads = 1.6675022840499878
+I1202 10:59:35.228742 137274321021824 utils.py:1231] [62550] lr = 0.00048090609490225497
+I1202 10:59:35.228821 137274321021824 utils.py:1231] [62550] uptime = 392964.5911826
+I1202 10:59:35.228899 137274321021824 utils.py:1231] [62550] examples_seen = 64051200.0
+I1202 10:59:35.228954 137274321021824 utils.py:1231] [62550] progress = 0.5554914167473336
+I1202 10:59:35.229009 137274321021824 utils.py:1231] [62550] epoch = 49.99441915066498
+I1202 10:59:35.229069 137274321021824 utils.py:1231] [62550] img/sec/core = 129.27964832185148
+I1202 10:59:35.229134 137274321021824 utils.py:1231] [62550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.12251328169889
+I1202 10:59:35.229188 137274321021824 utils.py:1231] [62550] core_hours = 109.12251328169889
+I1202 10:59:35.229256 137274321021824 train.py:125] NOTE: Steps:62550/112603 [55.5%]
+Walltime:4d13h9m (0s eval)
+ETA:3d15h19m
+Total train time:8d4h26m
+I1202 11:04:16.488275 137274321021824 utils.py:1231] [62600] l2_params = 290.70773141948644
+I1202 11:04:16.488480 137274321021824 utils.py:1231] [62600] train/loss = 3.3124676048755646
+I1202 11:04:16.488571 137274321021824 utils.py:1231] [62600] l2_grads = 1.579344630241394
+I1202 11:04:16.488627 137274321021824 utils.py:1231] [62600] lr = 0.0004801412030250284
+I1202 11:04:16.488676 137274321021824 utils.py:1231] [62600] uptime = 393245.851038163
+I1202 11:04:16.488726 137274321021824 utils.py:1231] [62600] examples_seen = 64102400.0
+I1202 11:04:16.488772 137274321021824 utils.py:1231] [62600] progress = 0.5559354546504089
+I1202 11:04:16.488817 137274321021824 utils.py:1231] [62600] epoch = 50.03438271513394
+I1202 11:04:16.488864 137274321021824 utils.py:1231] [62600] img/sec/core = 182.03806546623358
+I1202 11:04:16.488923 137274321021824 utils.py:1231] [62600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.20064101935527
+I1202 11:04:16.488971 137274321021824 utils.py:1231] [62600] core_hours = 109.20064101935527
+I1202 11:04:16.489028 137274321021824 train.py:125] NOTE: Steps:62600/112603 [55.6%]
+Walltime:4d13h14m (0s eval)
+ETA:3d15h13m
+Total train time:8d4h25m
+I1202 11:08:49.010560 137274321021824 utils.py:1231] [62650] l2_params = 290.62136204729376
+I1202 11:08:49.010764 137274321021824 utils.py:1231] [62650] train/loss = 2.1749828904867172
+I1202 11:08:49.010854 137274321021824 utils.py:1231] [62650] l2_grads = 1.8400152921676636
+I1202 11:08:49.010918 137274321021824 utils.py:1231] [62650] lr = 0.00047937635769274267
+I1202 11:08:49.010972 137274321021824 utils.py:1231] [62650] uptime = 393518.37333468196
+I1202 11:08:49.011022 137274321021824 utils.py:1231] [62650] examples_seen = 64153600.0
+I1202 11:08:49.011070 137274321021824 utils.py:1231] [62650] progress = 0.5563794925534844
+I1202 11:08:49.011117 137274321021824 utils.py:1231] [62650] epoch = 50.07434627960289
+I1202 11:08:49.011167 137274321021824 utils.py:1231] [62650] img/sec/core = 187.87453597007908
+I1202 11:08:49.011223 137274321021824 utils.py:1231] [62650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.27634165727721
+I1202 11:08:49.011273 137274321021824 utils.py:1231] [62650] core_hours = 109.27634165727721
+I1202 11:08:49.011331 137274321021824 train.py:125] NOTE: Steps:62650/112603 [55.6%]
+Walltime:4d13h18m (0s eval)
+ETA:3d15h7m
+Total train time:8d4h24m
+I1202 11:13:21.004757 137274321021824 utils.py:1231] [62700] l2_params = 290.5333197700399
+I1202 11:13:21.004975 137274321021824 utils.py:1231] [62700] train/loss = 2.10734660923481
+I1202 11:13:21.005087 137274321021824 utils.py:1231] [62700] l2_grads = 1.7069453001022339
+I1202 11:13:21.005156 137274321021824 utils.py:1231] [62700] lr = 0.00047861156069803877
+I1202 11:13:21.005217 137274321021824 utils.py:1231] [62700] uptime = 393790.367578423
+I1202 11:13:21.005282 137274321021824 utils.py:1231] [62700] examples_seen = 64204800.0
+I1202 11:13:21.005339 137274321021824 utils.py:1231] [62700] progress = 0.5568235304565597
+I1202 11:13:21.005409 137274321021824 utils.py:1231] [62700] epoch = 50.11430984407185
+I1202 11:13:21.005476 137274321021824 utils.py:1231] [62700] img/sec/core = 188.2392777721697
+I1202 11:13:21.005560 137274321021824 utils.py:1231] [62700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.35189561387193
+I1202 11:13:21.005634 137274321021824 utils.py:1231] [62700] core_hours = 109.35189561387193
+I1202 11:13:21.005701 137274321021824 train.py:125] NOTE: Steps:62700/112603 [55.7%]
+Walltime:4d13h23m (0s eval)
+ETA:3d15h2m
+Total train time:8d4h23m
+I1202 11:17:53.106372 137274321021824 utils.py:1231] [62750] l2_params = 290.46060526480574
+I1202 11:17:53.106662 137274321021824 utils.py:1231] [62750] train/loss = 3.846334159374237
+I1202 11:17:53.106815 137274321021824 utils.py:1231] [62750] l2_grads = 1.5022519826889038
+I1202 11:17:53.106890 137274321021824 utils.py:1231] [62750] lr = 0.0004778468138334433
+I1202 11:17:53.106951 137274321021824 utils.py:1231] [62750] uptime = 394062.469312275
+I1202 11:17:53.107031 137274321021824 utils.py:1231] [62750] examples_seen = 64256000.0
+I1202 11:17:53.107096 137274321021824 utils.py:1231] [62750] progress = 0.5572675683596352
+I1202 11:17:53.107151 137274321021824 utils.py:1231] [62750] epoch = 50.15427340854081
+I1202 11:17:53.107219 137274321021824 utils.py:1231] [62750] img/sec/core = 188.1649163906019
+I1202 11:17:53.107291 137274321021824 utils.py:1231] [62750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.42747942883084
+I1202 11:17:53.107354 137274321021824 utils.py:1231] [62750] core_hours = 109.42747942883084
+I1202 11:17:53.107416 137274321021824 train.py:125] NOTE: Steps:62750/112603 [55.7%]
+Walltime:4d13h27m (0s eval)
+ETA:3d14h56m
+Total train time:8d4h22m
+I1202 11:22:24.756806 137274321021824 utils.py:1231] [62800] l2_params = 290.3716962398951
+I1202 11:22:24.757024 137274321021824 utils.py:1231] [62800] train/loss = 2.5961559414863586
+I1202 11:22:24.757140 137274321021824 utils.py:1231] [62800] l2_grads = 1.6018811464309692
+I1202 11:22:24.757214 137274321021824 utils.py:1231] [62800] lr = 0.0004770821188913667
+I1202 11:22:24.757269 137274321021824 utils.py:1231] [62800] uptime = 394334.119631233
+I1202 11:22:24.757334 137274321021824 utils.py:1231] [62800] examples_seen = 64307200.0
+I1202 11:22:24.757383 137274321021824 utils.py:1231] [62800] progress = 0.5577116062627105
+I1202 11:22:24.757430 137274321021824 utils.py:1231] [62800] epoch = 50.194236973009765
+I1202 11:22:24.757480 137274321021824 utils.py:1231] [62800] img/sec/core = 188.47759942413143
+I1202 11:22:24.757546 137274321021824 utils.py:1231] [62800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.5029378507636
+I1202 11:22:24.757596 137274321021824 utils.py:1231] [62800] core_hours = 109.5029378507636
+I1202 11:22:24.757654 137274321021824 train.py:125] NOTE: Steps:62800/112603 [55.8%]
+Walltime:4d13h32m (0s eval)
+ETA:3d14h50m
+Total train time:8d4h20m
+I1202 11:26:55.598110 137274321021824 utils.py:1231] [62850] l2_params = 290.275917354438
+I1202 11:26:55.598309 137274321021824 utils.py:1231] [62850] train/loss = 2.6389599442481995
+I1202 11:26:55.598394 137274321021824 utils.py:1231] [62850] l2_grads = 1.6296788454055786
+I1202 11:26:55.598448 137274321021824 utils.py:1231] [62850] lr = 0.000476317477664096
+I1202 11:26:55.598497 137274321021824 utils.py:1231] [62850] uptime = 394604.960859438
+I1202 11:26:55.598548 137274321021824 utils.py:1231] [62850] examples_seen = 64358400.0
+I1202 11:26:55.598594 137274321021824 utils.py:1231] [62850] progress = 0.558155644165786
+I1202 11:26:55.598649 137274321021824 utils.py:1231] [62850] epoch = 50.23420053747872
+I1202 11:26:55.598702 137274321021824 utils.py:1231] [62850] img/sec/core = 189.04064325555476
+I1202 11:26:55.598760 137274321021824 utils.py:1231] [62850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.57817152526499
+I1202 11:26:55.598805 137274321021824 utils.py:1231] [62850] core_hours = 109.57817152526499
+I1202 11:26:55.598861 137274321021824 train.py:125] NOTE: Steps:62850/112603 [55.8%]
+Walltime:4d13h36m (0s eval)
+ETA:3d14h44m
+Total train time:8d4h19m
+I1202 11:31:32.692954 137274321021824 utils.py:1231] [62900] l2_params = 290.1917466852789
+I1202 11:31:32.693183 137274321021824 utils.py:1231] [62900] train/loss = 2.1776697635650635
+I1202 11:31:32.693287 137274321021824 utils.py:1231] [62900] l2_grads = 1.7689570188522339
+I1202 11:31:32.693384 137274321021824 utils.py:1231] [62900] lr = 0.000475552891943793
+I1202 11:31:32.693492 137274321021824 utils.py:1231] [62900] uptime = 394882.055831795
+I1202 11:31:32.693578 137274321021824 utils.py:1231] [62900] examples_seen = 64409600.0
+I1202 11:31:32.693657 137274321021824 utils.py:1231] [62900] progress = 0.5585996820688613
+I1202 11:31:32.693729 137274321021824 utils.py:1231] [62900] epoch = 50.274164101947676
+I1202 11:31:32.693790 137274321021824 utils.py:1231] [62900] img/sec/core = 184.7741933550474
+I1202 11:31:32.693867 137274321021824 utils.py:1231] [62900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.65514235091972
+I1202 11:31:32.693965 137274321021824 utils.py:1231] [62900] core_hours = 109.65514235091972
+I1202 11:31:32.694110 137274321021824 train.py:125] NOTE: Steps:62900/112603 [55.9%]
+Walltime:4d13h41m (0s eval)
+ETA:3d14h39m
+Total train time:8d4h18m
+I1202 11:36:16.513190 137274321021824 utils.py:1231] [62950] l2_params = 290.1132189380494
+I1202 11:36:16.513384 137274321021824 utils.py:1231] [62950] train/loss = 2.1405283957719803
+I1202 11:36:16.513482 137274321021824 utils.py:1231] [62950] l2_grads = 1.742801547050476
+I1202 11:36:16.513555 137274321021824 utils.py:1231] [62950] lr = 0.00047478836352249024
+I1202 11:36:16.513630 137274321021824 utils.py:1231] [62950] uptime = 395165.875974969
+I1202 11:36:16.513684 137274321021824 utils.py:1231] [62950] examples_seen = 64460800.0
+I1202 11:36:16.513728 137274321021824 utils.py:1231] [62950] progress = 0.5590437199719368
+I1202 11:36:16.513773 137274321021824 utils.py:1231] [62950] epoch = 50.31412766641663
+I1202 11:36:16.513820 137274321021824 utils.py:1231] [62950] img/sec/core = 180.39593464867485
+I1202 11:36:16.513872 137274321021824 utils.py:1231] [62950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.73398127957917
+I1202 11:36:16.513924 137274321021824 utils.py:1231] [62950] core_hours = 109.73398127957917
+I1202 11:36:16.513979 137274321021824 train.py:125] NOTE: Steps:62950/112603 [55.9%]
+Walltime:4d13h46m (0s eval)
+ETA:3d14h33m
+Total train time:8d4h17m
+I1202 11:41:12.433490 137274321021824 utils.py:1231] [63000] l2_params = 290.00764248577514
+I1202 11:41:12.433700 137274321021824 utils.py:1231] [63000] train/loss = 4.6309614181518555
+I1202 11:41:12.433812 137274321021824 utils.py:1231] [63000] l2_grads = 1.6655360460281372
+I1202 11:41:12.433901 137274321021824 utils.py:1231] [63000] lr = 0.00047402389419208535
+I1202 11:41:12.433999 137274321021824 utils.py:1231] [63000] uptime = 395461.796359013
+I1202 11:41:12.434062 137274321021824 utils.py:1231] [63000] examples_seen = 64512000.0
+I1202 11:41:12.434119 137274321021824 utils.py:1231] [63000] progress = 0.5594877578750123
+I1202 11:41:12.434171 137274321021824 utils.py:1231] [63000] epoch = 50.354091230885594
+I1202 11:41:12.434225 137274321021824 utils.py:1231] [63000] img/sec/core = 173.01951051937925
+I1202 11:41:12.434283 137274321021824 utils.py:1231] [63000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.81618138625805
+I1202 11:41:12.434342 137274321021824 utils.py:1231] [63000] core_hours = 109.81618138625805
+I1202 11:41:12.434416 137274321021824 train.py:125] NOTE: Steps:63000/112603 [55.9%]
+Walltime:4d13h51m (0s eval)
+ETA:3d14h27m
+Total train time:8d4h17m
+I1202 11:46:07.963097 137274321021824 utils.py:1231] [63050] l2_params = 289.9345908547365
+I1202 11:46:07.963336 137274321021824 utils.py:1231] [63050] train/loss = 2.204306975007057
+I1202 11:46:07.963465 137274321021824 utils.py:1231] [63050] l2_grads = 1.7148172855377197
+I1202 11:46:07.963557 137274321021824 utils.py:1231] [63050] lr = 0.00047325948574433713
+I1202 11:46:07.963632 137274321021824 utils.py:1231] [63050] uptime = 395757.325993333
+I1202 11:46:07.963712 137274321021824 utils.py:1231] [63050] examples_seen = 64563200.0
+I1202 11:46:07.963768 137274321021824 utils.py:1231] [63050] progress = 0.5599317957780876
+I1202 11:46:07.963832 137274321021824 utils.py:1231] [63050] epoch = 50.39405479535455
+I1202 11:46:07.963898 137274321021824 utils.py:1231] [63050] img/sec/core = 173.24827717466775
+I1202 11:46:07.963970 137274321021824 utils.py:1231] [63050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.89827295134695
+I1202 11:46:07.964025 137274321021824 utils.py:1231] [63050] core_hours = 109.89827295134695
+I1202 11:46:07.964089 137274321021824 train.py:125] NOTE: Steps:63050/112603 [56.0%]
+Walltime:4d13h55m (0s eval)
+ETA:3d14h22m
+Total train time:8d4h16m
+I1202 11:51:03.960278 137274321021824 utils.py:1231] [63100] l2_params = 289.83450550698416
+I1202 11:51:03.960552 137274321021824 utils.py:1231] [63100] train/loss = 2.2907527685165405
+I1202 11:51:03.960694 137274321021824 utils.py:1231] [63100] l2_grads = 1.858561396598816
+I1202 11:51:03.960809 137274321021824 utils.py:1231] [63100] lr = 0.000472495139970862
+I1202 11:51:03.960912 137274321021824 utils.py:1231] [63100] uptime = 396053.323271907
+I1202 11:51:03.960987 137274321021824 utils.py:1231] [63100] examples_seen = 64614400.0
+I1202 11:51:03.961072 137274321021824 utils.py:1231] [63100] progress = 0.5603758336811631
+I1202 11:51:03.961188 137274321021824 utils.py:1231] [63100] epoch = 50.434018359823504
+I1202 11:51:03.961266 137274321021824 utils.py:1231] [63100] img/sec/core = 172.9745633022878
+I1202 11:51:03.961347 137274321021824 utils.py:1231] [63100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 109.9804944176175
+I1202 11:51:03.961426 137274321021824 utils.py:1231] [63100] core_hours = 109.9804944176175
+I1202 11:51:03.961517 137274321021824 train.py:125] NOTE: Steps:63100/112603 [56.0%]
+Walltime:4d14h0m (0s eval)
+ETA:3d14h17m
+Total train time:8d4h16m
+I1202 11:56:03.709648 137274321021824 utils.py:1231] [63150] l2_params = 289.7444485569896
+I1202 11:56:03.709912 137274321021824 utils.py:1231] [63150] train/loss = 2.1897361129522324
+I1202 11:56:03.710046 137274321021824 utils.py:1231] [63150] l2_grads = 1.7139307260513306
+I1202 11:56:03.710129 137274321021824 utils.py:1231] [63150] lr = 0.0004717308586631294
+I1202 11:56:03.710208 137274321021824 utils.py:1231] [63150] uptime = 396353.072564302
+I1202 11:56:03.710287 137274321021824 utils.py:1231] [63150] examples_seen = 64665600.0
+I1202 11:56:03.710357 137274321021824 utils.py:1231] [63150] progress = 0.5608198715842384
+I1202 11:56:03.710416 137274321021824 utils.py:1231] [63150] epoch = 50.47398192429246
+I1202 11:56:03.710471 137274321021824 utils.py:1231] [63150] img/sec/core = 170.80941072757216
+I1202 11:56:03.710532 137274321021824 utils.py:1231] [63150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.06375810994945
+I1202 11:56:03.710586 137274321021824 utils.py:1231] [63150] core_hours = 110.06375810994945
+I1202 11:56:03.710650 137274321021824 train.py:125] NOTE: Steps:63150/112603 [56.1%]
+Walltime:4d14h5m (0s eval)
+ETA:3d14h11m
+Total train time:8d4h15m
+I1202 12:01:04.590445 137274321021824 utils.py:1231] [63200] l2_params = 289.66087877157906
+I1202 12:01:04.590690 137274321021824 utils.py:1231] [63200] train/loss = 2.539562165737152
+I1202 12:01:04.590895 137274321021824 utils.py:1231] [63200] l2_grads = 1.628819465637207
+I1202 12:01:04.590978 137274321021824 utils.py:1231] [63200] lr = 0.0004709666436124575
+I1202 12:01:04.591037 137274321021824 utils.py:1231] [63200] uptime = 396653.95339841896
+I1202 12:01:04.591087 137274321021824 utils.py:1231] [63200] examples_seen = 64716800.0
+I1202 12:01:04.591135 137274321021824 utils.py:1231] [63200] progress = 0.5612639094873139
+I1202 12:01:04.591237 137274321021824 utils.py:1231] [63200] epoch = 50.513945488761415
+I1202 12:01:04.591288 137274321021824 utils.py:1231] [63200] img/sec/core = 170.16703689441056
+I1202 12:01:04.591345 137274321021824 utils.py:1231] [63200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.14733611942638
+I1202 12:01:04.591394 137274321021824 utils.py:1231] [63200] core_hours = 110.14733611942638
+I1202 12:01:04.591452 137274321021824 train.py:125] NOTE: Steps:63200/112603 [56.1%]
+Walltime:4d14h10m (0s eval)
+ETA:3d14h6m
+Total train time:8d4h15m
+I1202 12:06:01.610994 137274321021824 utils.py:1231] [63250] l2_params = 289.57347906849543
+I1202 12:06:01.611279 137274321021824 utils.py:1231] [63250] train/loss = 2.1470377892255783
+I1202 12:06:01.611461 137274321021824 utils.py:1231] [63250] l2_grads = 1.7902681827545166
+I1202 12:06:01.611570 137274321021824 utils.py:1231] [63250] lr = 0.0004702024966100105
+I1202 12:06:01.611653 137274321021824 utils.py:1231] [63250] uptime = 396950.974014719
+I1202 12:06:01.611727 137274321021824 utils.py:1231] [63250] examples_seen = 64768000.0
+I1202 12:06:01.611794 137274321021824 utils.py:1231] [63250] progress = 0.5617079473903892
+I1202 12:06:01.611863 137274321021824 utils.py:1231] [63250] epoch = 50.55390905323038
+I1202 12:06:01.611936 137274321021824 utils.py:1231] [63250] img/sec/core = 172.3786067034383
+I1202 12:06:01.611995 137274321021824 utils.py:1231] [63250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.2298418461764
+I1202 12:06:01.612051 137274321021824 utils.py:1231] [63250] core_hours = 110.2298418461764
+I1202 12:06:01.612117 137274321021824 train.py:125] NOTE: Steps:63250/112603 [56.2%]
+Walltime:4d14h15m (0s eval)
+ETA:3d14h0m
+Total train time:8d4h14m
+I1202 12:11:07.727750 137274321021824 utils.py:1231] [63300] l2_params = 289.48741274905706
+I1202 12:11:07.728024 137274321021824 utils.py:1231] [63300] train/loss = 2.0259193778038025
+I1202 12:11:07.728159 137274321021824 utils.py:1231] [63300] l2_grads = 1.671099066734314
+I1202 12:11:07.728278 137274321021824 utils.py:1231] [63300] lr = 0.00046943841944679113
+I1202 12:11:07.728364 137274321021824 utils.py:1231] [63300] uptime = 397257.090720014
+I1202 12:11:07.728442 137274321021824 utils.py:1231] [63300] examples_seen = 64819200.0
+I1202 12:11:07.728499 137274321021824 utils.py:1231] [63300] progress = 0.5621519852934647
+I1202 12:11:07.728564 137274321021824 utils.py:1231] [63300] epoch = 50.59387261769933
+I1202 12:11:07.728634 137274321021824 utils.py:1231] [63300] img/sec/core = 167.25647151683896
+I1202 12:11:07.728709 137274321021824 utils.py:1231] [63300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.31487426431389
+I1202 12:11:07.728762 137274321021824 utils.py:1231] [63300] core_hours = 110.31487426431389
+I1202 12:11:07.728826 137274321021824 train.py:125] NOTE: Steps:63300/112603 [56.2%]
+Walltime:4d14h20m (0s eval)
+ETA:3d13h55m
+Total train time:8d4h14m
+I1202 12:16:14.467622 137274321021824 utils.py:1231] [63350] l2_params = 289.39281696162743
+I1202 12:16:14.467829 137274321021824 utils.py:1231] [63350] train/loss = 2.306752324104309
+I1202 12:16:14.467932 137274321021824 utils.py:1231] [63350] l2_grads = 1.711634635925293
+I1202 12:16:14.468003 137274321021824 utils.py:1231] [63350] lr = 0.00046867441391363924
+I1202 12:16:14.468081 137274321021824 utils.py:1231] [63350] uptime = 397563.830437479
+I1202 12:16:14.468160 137274321021824 utils.py:1231] [63350] examples_seen = 64870400.0
+I1202 12:16:14.468249 137274321021824 utils.py:1231] [63350] progress = 0.56259602319654
+I1202 12:16:14.468316 137274321021824 utils.py:1231] [63350] epoch = 50.63383618216829
+I1202 12:16:14.688475 137274321021824 utils.py:1231] [63350] img/sec/core = 166.91676064362397
+I1202 12:16:14.688825 137274321021824 utils.py:1231] [63350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.4000797413875
+I1202 12:16:14.689071 137274321021824 utils.py:1231] [63350] core_hours = 110.4000797413875
+I1202 12:16:14.689283 137274321021824 train.py:125] NOTE: Steps:63350/112603 [56.3%]
+Walltime:4d14h26m (0s eval)
+ETA:3d13h50m
+Total train time:8d4h14m
+I1202 12:21:22.923034 137274321021824 utils.py:1231] [63400] l2_params = 289.30878476231516
+I1202 12:21:22.923319 137274321021824 utils.py:1231] [63400] train/loss = 2.594213217496872
+I1202 12:21:22.923495 137274321021824 utils.py:1231] [63400] l2_grads = 1.7025972604751587
+I1202 12:21:22.923590 137274321021824 utils.py:1231] [63400] lr = 0.0004679104818012266
+I1202 12:21:22.923704 137274321021824 utils.py:1231] [63400] uptime = 397872.286057845
+I1202 12:21:22.923785 137274321021824 utils.py:1231] [63400] examples_seen = 64921600.0
+I1202 12:21:22.923850 137274321021824 utils.py:1231] [63400] progress = 0.5630400610996155
+I1202 12:21:22.923926 137274321021824 utils.py:1231] [63400] epoch = 50.67379974663724
+I1202 12:21:22.923992 137274321021824 utils.py:1231] [63400] img/sec/core = 165.98822203094443
+I1202 12:21:22.924068 137274321021824 utils.py:1231] [63400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.48576185815584
+I1202 12:21:22.924134 137274321021824 utils.py:1231] [63400] core_hours = 110.48576185815584
+I1202 12:21:22.924213 137274321021824 train.py:125] NOTE: Steps:63400/112603 [56.3%]
+Walltime:4d14h31m (0s eval)
+ETA:3d13h44m
+Total train time:8d4h14m
+I1202 12:26:33.027960 137274321021824 utils.py:1231] [63450] l2_params = 289.20829719045764
+I1202 12:26:33.028172 137274321021824 utils.py:1231] [63450] train/loss = 3.0712006092071533
+I1202 12:26:33.028269 137274321021824 utils.py:1231] [63450] l2_grads = 1.6101499795913696
+I1202 12:26:33.028355 137274321021824 utils.py:1231] [63450] lr = 0.0004671466249000537
+I1202 12:26:33.028416 137274321021824 utils.py:1231] [63450] uptime = 398182.390774285
+I1202 12:26:33.028484 137274321021824 utils.py:1231] [63450] examples_seen = 64972800.0
+I1202 12:26:33.028542 137274321021824 utils.py:1231] [63450] progress = 0.5634840990026909
+I1202 12:26:33.028603 137274321021824 utils.py:1231] [63450] epoch = 50.713763311106206
+I1202 12:26:33.028655 137274321021824 utils.py:1231] [63450] img/sec/core = 165.10551850926984
+I1202 12:26:33.028721 137274321021824 utils.py:1231] [63450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.57190205716694
+I1202 12:26:33.028774 137274321021824 utils.py:1231] [63450] core_hours = 110.57190205716694
+I1202 12:26:33.028837 137274321021824 train.py:125] NOTE: Steps:63450/112603 [56.3%]
+Walltime:4d14h36m (0s eval)
+ETA:3d13h39m
+Total train time:8d4h14m
+I1202 12:31:42.641314 137274321021824 utils.py:1231] [63500] l2_params = 289.10883189250694
+I1202 12:31:42.641534 137274321021824 utils.py:1231] [63500] train/loss = 3.1822389364242554
+I1202 12:31:42.641772 137274321021824 utils.py:1231] [63500] l2_grads = 1.4608631134033203
+I1202 12:31:42.641891 137274321021824 utils.py:1231] [63500] lr = 0.00046638284500044483
+I1202 12:31:42.641984 137274321021824 utils.py:1231] [63500] uptime = 398492.004336164
+I1202 12:31:42.642105 137274321021824 utils.py:1231] [63500] examples_seen = 65024000.0
+I1202 12:31:42.642229 137274321021824 utils.py:1231] [63500] progress = 0.5639281369057663
+I1202 12:31:42.642330 137274321021824 utils.py:1231] [63500] epoch = 50.75372687557516
+I1202 12:31:42.642426 137274321021824 utils.py:1231] [63500] img/sec/core = 165.36743316177592
+I1202 12:31:42.642517 137274321021824 utils.py:1231] [63500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.65790582435555
+I1202 12:31:42.642580 137274321021824 utils.py:1231] [63500] core_hours = 110.65790582435555
+I1202 12:31:42.642663 137274321021824 train.py:125] NOTE: Steps:63500/112603 [56.4%]
+Walltime:4d14h41m (0s eval)
+ETA:3d13h34m
+Total train time:8d4h13m
+I1202 12:36:50.266628 137274321021824 utils.py:1231] [63550] l2_params = 289.0080354965175
+I1202 12:36:50.266844 137274321021824 utils.py:1231] [63550] train/loss = 3.2689031064510345
+I1202 12:36:50.266961 137274321021824 utils.py:1231] [63550] l2_grads = 1.5499322414398193
+I1202 12:36:50.267034 137274321021824 utils.py:1231] [63550] lr = 0.0004656191438925422
+I1202 12:36:50.267088 137274321021824 utils.py:1231] [63550] uptime = 398799.629449722
+I1202 12:36:50.267141 137274321021824 utils.py:1231] [63550] examples_seen = 65075200.0
+I1202 12:36:50.267192 137274321021824 utils.py:1231] [63550] progress = 0.5643721748088417
+I1202 12:36:50.267238 137274321021824 utils.py:1231] [63550] epoch = 50.79369044004412
+I1202 12:36:50.267299 137274321021824 utils.py:1231] [63550] img/sec/core = 166.43634652520873
+I1202 12:36:50.267354 137274321021824 utils.py:1231] [63550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.74335724478833
+I1202 12:36:50.267403 137274321021824 utils.py:1231] [63550] core_hours = 110.74335724478833
+I1202 12:36:50.267462 137274321021824 train.py:125] NOTE: Steps:63550/112603 [56.4%]
+Walltime:4d14h46m (0s eval)
+ETA:3d13h28m
+Total train time:8d4h13m
+I1202 12:41:52.864335 137274321021824 utils.py:1231] [63600] l2_params = 288.90958513518655
+I1202 12:41:52.864532 137274321021824 utils.py:1231] [63600] train/loss = 3.5718747973442078
+I1202 12:41:52.864640 137274321021824 utils.py:1231] [63600] l2_grads = 1.4898499250411987
+I1202 12:41:52.864739 137274321021824 utils.py:1231] [63600] lr = 0.0004648555233663048
+I1202 12:41:52.864803 137274321021824 utils.py:1231] [63600] uptime = 399102.227165574
+I1202 12:41:52.864865 137274321021824 utils.py:1231] [63600] examples_seen = 65126400.0
+I1202 12:41:52.864923 137274321021824 utils.py:1231] [63600] progress = 0.5648162127119171
+I1202 12:41:52.864976 137274321021824 utils.py:1231] [63600] epoch = 50.83365400451307
+I1202 12:41:52.865029 137274321021824 utils.py:1231] [63600] img/sec/core = 169.20154157755564
+I1202 12:41:52.865084 137274321021824 utils.py:1231] [63600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.82741216585832
+I1202 12:41:52.865137 137274321021824 utils.py:1231] [63600] core_hours = 110.82741216585832
+I1202 12:41:52.865199 137274321021824 train.py:125] NOTE: Steps:63600/112603 [56.5%]
+Walltime:4d14h51m (0s eval)
+ETA:3d13h23m
+Total train time:8d4h13m
+I1202 12:47:02.809933 137274321021824 utils.py:1231] [63650] l2_params = 288.8135185561167
+I1202 12:47:02.810220 137274321021824 utils.py:1231] [63650] train/loss = 2.2313798367977142
+I1202 12:47:02.810399 137274321021824 utils.py:1231] [63650] l2_grads = 1.6876834630966187
+I1202 12:47:02.810471 137274321021824 utils.py:1231] [63650] lr = 0.0004640919852115021
+I1202 12:47:02.810530 137274321021824 utils.py:1231] [63650] uptime = 399412.172891494
+I1202 12:47:02.810593 137274321021824 utils.py:1231] [63650] examples_seen = 65177600.0
+I1202 12:47:02.810655 137274321021824 utils.py:1231] [63650] progress = 0.5652602506149925
+I1202 12:47:02.810720 137274321021824 utils.py:1231] [63650] epoch = 50.87361756898203
+I1202 12:47:02.810783 137274321021824 utils.py:1231] [63650] img/sec/core = 165.1902114411172
+I1202 12:47:02.810855 137274321021824 utils.py:1231] [63650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.9135082008361
+I1202 12:47:02.810931 137274321021824 utils.py:1231] [63650] core_hours = 110.9135082008361
+I1202 12:47:02.810995 137274321021824 train.py:125] NOTE: Steps:63650/112603 [56.5%]
+Walltime:4d14h56m (0s eval)
+ETA:3d13h18m
+Total train time:8d4h13m
+I1202 12:52:10.929786 137274321021824 utils.py:1231] [63700] l2_params = 288.7261785592438
+I1202 12:52:10.930069 137274321021824 utils.py:1231] [63700] train/loss = 3.1203157007694244
+I1202 12:52:10.930257 137274321021824 utils.py:1231] [63700] l2_grads = 1.5747970342636108
+I1202 12:52:10.930339 137274321021824 utils.py:1231] [63700] lr = 0.0004633285312177111
+I1202 12:52:10.930409 137274321021824 utils.py:1231] [63700] uptime = 399720.292770167
+I1202 12:52:10.930491 137274321021824 utils.py:1231] [63700] examples_seen = 65228800.0
+I1202 12:52:10.930555 137274321021824 utils.py:1231] [63700] progress = 0.5657042885180679
+I1202 12:52:10.930622 137274321021824 utils.py:1231] [63700] epoch = 50.91358113345099
+I1202 12:52:10.930694 137274321021824 utils.py:1231] [63700] img/sec/core = 166.16909048681723
+I1202 12:52:10.930759 137274321021824 utils.py:1231] [63700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 110.99909705602305
+I1202 12:52:10.930817 137274321021824 utils.py:1231] [63700] core_hours = 110.99909705602305
+I1202 12:52:10.930897 137274321021824 train.py:125] NOTE: Steps:63700/112603 [56.6%]
+Walltime:4d15h2m (0s eval)
+ETA:3d13h13m
+Total train time:8d4h13m
+I1202 12:57:22.707676 137274321021824 utils.py:1231] [63750] l2_params = 288.6358852735565
+I1202 12:57:22.707931 137274321021824 utils.py:1231] [63750] train/loss = 2.462112635374069
+I1202 12:57:22.708056 137274321021824 utils.py:1231] [63750] l2_grads = 1.6273727416992188
+I1202 12:57:22.708152 137274321021824 utils.py:1231] [63750] lr = 0.0004625651631743107
+I1202 12:57:22.708226 137274321021824 utils.py:1231] [63750] uptime = 400032.07058403397
+I1202 12:57:22.708302 137274321021824 utils.py:1231] [63750] examples_seen = 65280000.0
+I1202 12:57:22.708359 137274321021824 utils.py:1231] [63750] progress = 0.5661483264211433
+I1202 12:57:22.708422 137274321021824 utils.py:1231] [63750] epoch = 50.953544697919945
+I1202 12:57:22.708487 137274321021824 utils.py:1231] [63750] img/sec/core = 164.21951057058396
+I1202 12:57:22.708557 137274321021824 utils.py:1231] [63750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.08570200431943
+I1202 12:57:22.708620 137274321021824 utils.py:1231] [63750] core_hours = 111.08570200431943
+I1202 12:57:22.708687 137274321021824 train.py:125] NOTE: Steps:63750/112603 [56.6%]
+Walltime:4d15h7m (0s eval)
+ETA:3d13h7m
+Total train time:8d4h13m
+I1202 13:02:34.494318 137274321021824 utils.py:1231] [63800] l2_params = 288.54545496672114
+I1202 13:02:34.494524 137274321021824 utils.py:1231] [63800] train/loss = 2.2004327476024628
+I1202 13:02:34.494627 137274321021824 utils.py:1231] [63800] l2_grads = 1.7213832139968872
+I1202 13:02:34.494695 137274321021824 utils.py:1231] [63800] lr = 0.00046180188287047943
+I1202 13:02:34.494767 137274321021824 utils.py:1231] [63800] uptime = 400343.857124224
+I1202 13:02:34.494859 137274321021824 utils.py:1231] [63800] examples_seen = 65331200.0
+I1202 13:02:34.494937 137274321021824 utils.py:1231] [63800] progress = 0.5665923643242187
+I1202 13:02:34.495002 137274321021824 utils.py:1231] [63800] epoch = 50.9935082623889
+I1202 13:02:34.495068 137274321021824 utils.py:1231] [63800] img/sec/core = 164.21491437311926
+I1202 13:02:34.495131 137274321021824 utils.py:1231] [63800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.17230937659446
+I1202 13:02:34.495188 137274321021824 utils.py:1231] [63800] core_hours = 111.17230937659446
+I1202 13:02:34.495251 137274321021824 train.py:125] NOTE: Steps:63800/112603 [56.7%]
+Walltime:4d15h12m (0s eval)
+ETA:3d13h2m
+Total train time:8d4h13m
+I1202 13:07:41.710849 137274321021824 utils.py:1231] [63850] l2_params = 288.4443689408733
+I1202 13:07:41.711070 137274321021824 utils.py:1231] [63850] train/loss = 3.2415172159671783
+I1202 13:07:41.711174 137274321021824 utils.py:1231] [63850] l2_grads = 1.4823541641235352
+I1202 13:07:41.711251 137274321021824 utils.py:1231] [63850] lr = 0.0004610386920951889
+I1202 13:07:41.711310 137274321021824 utils.py:1231] [63850] uptime = 400651.073671512
+I1202 13:07:41.711371 137274321021824 utils.py:1231] [63850] examples_seen = 65382400.0
+I1202 13:07:41.711427 137274321021824 utils.py:1231] [63850] progress = 0.5670364022272941
+I1202 13:07:41.711488 137274321021824 utils.py:1231] [63850] epoch = 51.033471826857856
+I1202 13:07:41.711545 137274321021824 utils.py:1231] [63850] img/sec/core = 166.65768967191238
+I1202 13:07:41.711607 137274321021824 utils.py:1231] [63850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.25764730639666
+I1202 13:07:41.711662 137274321021824 utils.py:1231] [63850] core_hours = 111.25764730639666
+I1202 13:07:41.711728 137274321021824 train.py:125] NOTE: Steps:63850/112603 [56.7%]
+Walltime:4d15h17m (0s eval)
+ETA:3d12h57m
+Total train time:8d4h12m
+I1202 13:12:51.027245 137274321021824 utils.py:1231] [63900] l2_params = 288.3460345125555
+I1202 13:12:51.027512 137274321021824 utils.py:1231] [63900] train/loss = 3.595599114894867
+I1202 13:12:51.027748 137274321021824 utils.py:1231] [63900] l2_grads = 1.541563630104065
+I1202 13:12:51.027847 137274321021824 utils.py:1231] [63900] lr = 0.0004602755926372017
+I1202 13:12:51.027929 137274321021824 utils.py:1231] [63900] uptime = 400960.390288677
+I1202 13:12:51.027989 137274321021824 utils.py:1231] [63900] examples_seen = 65433600.0
+I1202 13:12:51.028045 137274321021824 utils.py:1231] [63900] progress = 0.5674804401303696
+I1202 13:12:51.028102 137274321021824 utils.py:1231] [63900] epoch = 51.07343539132681
+I1202 13:12:51.028169 137274321021824 utils.py:1231] [63900] img/sec/core = 165.52618630472486
+I1202 13:12:51.028253 137274321021824 utils.py:1231] [63900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.34356858894249
+I1202 13:12:51.028321 137274321021824 utils.py:1231] [63900] core_hours = 111.34356858894249
+I1202 13:12:51.028395 137274321021824 train.py:125] NOTE: Steps:63900/112603 [56.7%]
+Walltime:4d15h22m (0s eval)
+ETA:3d12h51m
+Total train time:8d4h12m
+I1202 13:17:58.193912 137274321021824 utils.py:1231] [63950] l2_params = 288.2654617869868
+I1202 13:17:58.194186 137274321021824 utils.py:1231] [63950] train/loss = 2.1985377073287964
+I1202 13:17:58.194394 137274321021824 utils.py:1231] [63950] l2_grads = 1.729668140411377
+I1202 13:17:58.194491 137274321021824 utils.py:1231] [63950] lr = 0.0004595125862850661
+I1202 13:17:58.194577 137274321021824 utils.py:1231] [63950] uptime = 401267.556935734
+I1202 13:17:58.194654 137274321021824 utils.py:1231] [63950] examples_seen = 65484800.0
+I1202 13:17:58.194725 137274321021824 utils.py:1231] [63950] progress = 0.5679244780334449
+I1202 13:17:58.194799 137274321021824 utils.py:1231] [63950] epoch = 51.11339895579577
+I1202 13:17:58.194862 137274321021824 utils.py:1231] [63950] img/sec/core = 166.6847637611531
+I1202 13:17:58.194943 137274321021824 utils.py:1231] [63950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.42889265756943
+I1202 13:17:58.195032 137274321021824 utils.py:1231] [63950] core_hours = 111.42889265756943
+I1202 13:17:58.195103 137274321021824 train.py:125] NOTE: Steps:63950/112603 [56.8%]
+Walltime:4d15h27m (0s eval)
+ETA:3d12h46m
+Total train time:8d4h12m
+I1202 13:23:07.032167 137274321021824 utils.py:1231] [64000] l2_params = 288.1813735701622
+I1202 13:23:07.032390 137274321021824 utils.py:1231] [64000] train/loss = 4.420177102088928
+I1202 13:23:07.032486 137274321021824 utils.py:1231] [64000] l2_grads = 1.6958767175674438
+I1202 13:23:07.032552 137274321021824 utils.py:1231] [64000] lr = 0.00045874967482711205
+I1202 13:23:07.032604 137274321021824 utils.py:1231] [64000] uptime = 401576.39496612997
+I1202 13:23:07.032658 137274321021824 utils.py:1231] [64000] examples_seen = 65536000.0
+I1202 13:23:07.032710 137274321021824 utils.py:1231] [64000] progress = 0.5683685159365204
+I1202 13:23:07.032760 137274321021824 utils.py:1231] [64000] epoch = 51.15336252026473
+I1202 13:23:07.032816 137274321021824 utils.py:1231] [64000] img/sec/core = 165.78269177003506
+I1202 13:23:07.032873 137274321021824 utils.py:1231] [64000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.51468099934611
+I1202 13:23:07.032932 137274321021824 utils.py:1231] [64000] core_hours = 111.51468099934611
+I1202 13:23:07.032994 137274321021824 train.py:125] NOTE: Steps:64000/112603 [56.8%]
+Walltime:4d15h32m (0s eval)
+ETA:3d12h41m
+Total train time:8d4h12m
+I1202 13:28:18.178973 137274321021824 utils.py:1231] [64050] l2_params = 288.0988915014938
+I1202 13:28:18.179230 137274321021824 utils.py:1231] [64050] train/loss = 3.3247672617435455
+I1202 13:28:18.179362 137274321021824 utils.py:1231] [64050] l2_grads = 1.5048871040344238
+I1202 13:28:18.179456 137274321021824 utils.py:1231] [64050] lr = 0.0004579868600514482
+I1202 13:28:18.179569 137274321021824 utils.py:1231] [64050] uptime = 401887.541918252
+I1202 13:28:18.179647 137274321021824 utils.py:1231] [64050] examples_seen = 65587200.0
+I1202 13:28:18.179706 137274321021824 utils.py:1231] [64050] progress = 0.5688125538395957
+I1202 13:28:18.179762 137274321021824 utils.py:1231] [64050] epoch = 51.193326084733684
+I1202 13:28:18.179817 137274321021824 utils.py:1231] [64050] img/sec/core = 164.55247159200073
+I1202 13:28:18.179877 137274321021824 utils.py:1231] [64050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.60111070826889
+I1202 13:28:18.179954 137274321021824 utils.py:1231] [64050] core_hours = 111.60111070826889
+I1202 13:28:18.180020 137274321021824 train.py:125] NOTE: Steps:64050/112603 [56.9%]
+Walltime:4d15h38m (0s eval)
+ETA:3d12h36m
+Total train time:8d4h12m
+I1202 13:33:29.965310 137274321021824 utils.py:1231] [64100] l2_params = 288.00128741771323
+I1202 13:33:29.965540 137274321021824 utils.py:1231] [64100] train/loss = 2.098799154162407
+I1202 13:33:29.965645 137274321021824 utils.py:1231] [64100] l2_grads = 1.821623682975769
+I1202 13:33:29.965706 137274321021824 utils.py:1231] [64100] lr = 0.0004572241437459551
+I1202 13:33:29.965760 137274321021824 utils.py:1231] [64100] uptime = 402199.328121598
+I1202 13:33:29.965814 137274321021824 utils.py:1231] [64100] examples_seen = 65638400.0
+I1202 13:33:29.965865 137274321021824 utils.py:1231] [64100] progress = 0.5692565917426712
+I1202 13:33:29.965925 137274321021824 utils.py:1231] [64100] epoch = 51.23328964920264
+I1202 13:33:29.965979 137274321021824 utils.py:1231] [64100] img/sec/core = 164.21509178577884
+I1202 13:33:29.966035 137274321021824 utils.py:1231] [64100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.68771798697611
+I1202 13:33:29.966087 137274321021824 utils.py:1231] [64100] core_hours = 111.68771798697611
+I1202 13:33:29.966149 137274321021824 train.py:125] NOTE: Steps:64100/112603 [56.9%]
+Walltime:4d15h43m (0s eval)
+ETA:3d12h30m
+Total train time:8d4h12m
+I1202 13:38:41.739691 137274321021824 utils.py:1231] [64150] l2_params = 287.92185851598197
+I1202 13:38:41.739911 137274321021824 utils.py:1231] [64150] train/loss = 4.399590015411377
+I1202 13:38:41.740015 137274321021824 utils.py:1231] [64150] l2_grads = 1.6391446590423584
+I1202 13:38:41.740086 137274321021824 utils.py:1231] [64150] lr = 0.0004564615276982831
+I1202 13:38:41.740155 137274321021824 utils.py:1231] [64150] uptime = 402511.102516267
+I1202 13:38:41.740223 137274321021824 utils.py:1231] [64150] examples_seen = 65689600.0
+I1202 13:38:41.740282 137274321021824 utils.py:1231] [64150] progress = 0.5697006296457465
+I1202 13:38:41.740340 137274321021824 utils.py:1231] [64150] epoch = 51.273253213671595
+I1202 13:38:41.740399 137274321021824 utils.py:1231] [64150] img/sec/core = 164.22131154920285
+I1202 13:38:41.740467 137274321021824 utils.py:1231] [64150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.77432198549528
+I1202 13:38:41.740524 137274321021824 utils.py:1231] [64150] core_hours = 111.77432198549528
+I1202 13:38:41.740590 137274321021824 train.py:125] NOTE: Steps:64150/112603 [57.0%]
+Walltime:4d15h48m (0s eval)
+ETA:3d12h25m
+Total train time:8d4h12m
+I1202 13:43:53.521797 137274321021824 utils.py:1231] [64200] l2_params = 287.8392327872263
+I1202 13:43:53.522049 137274321021824 utils.py:1231] [64200] train/loss = 3.1702842116355896
+I1202 13:43:53.522153 137274321021824 utils.py:1231] [64200] l2_grads = 1.5787079334259033
+I1202 13:43:53.522240 137274321021824 utils.py:1231] [64200] lr = 0.00045569901369584795
+I1202 13:43:53.522307 137274321021824 utils.py:1231] [64200] uptime = 402822.884667673
+I1202 13:43:53.522368 137274321021824 utils.py:1231] [64200] examples_seen = 65740800.0
+I1202 13:43:53.522426 137274321021824 utils.py:1231] [64200] progress = 0.570144667548822
+I1202 13:43:53.522484 137274321021824 utils.py:1231] [64200] epoch = 51.31321677814056
+I1202 13:43:53.522542 137274321021824 utils.py:1231] [64200] img/sec/core = 164.21722593519905
+I1202 13:43:53.522605 137274321021824 utils.py:1231] [64200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.86092813866361
+I1202 13:43:53.522662 137274321021824 utils.py:1231] [64200] core_hours = 111.86092813866361
+I1202 13:43:53.522740 137274321021824 train.py:125] NOTE: Steps:64200/112603 [57.0%]
+Walltime:4d15h53m (0s eval)
+ETA:3d12h20m
+Total train time:8d4h12m
+I1202 13:49:05.299463 137274321021824 utils.py:1231] [64250] l2_params = 287.7457211507323
+I1202 13:49:05.299666 137274321021824 utils.py:1231] [64250] train/loss = 3.2687894701957703
+I1202 13:49:05.299759 137274321021824 utils.py:1231] [64250] l2_grads = 1.5616753101348877
+I1202 13:49:05.299816 137274321021824 utils.py:1231] [64250] lr = 0.00045493660352582537
+I1202 13:49:05.299865 137274321021824 utils.py:1231] [64250] uptime = 403134.6622277
+I1202 13:49:05.299923 137274321021824 utils.py:1231] [64250] examples_seen = 65792000.0
+I1202 13:49:05.299970 137274321021824 utils.py:1231] [64250] progress = 0.5705887054518973
+I1202 13:49:05.300015 137274321021824 utils.py:1231] [64250] epoch = 51.35318034260951
+I1202 13:49:05.300062 137274321021824 utils.py:1231] [64250] img/sec/core = 164.2196442731861
+I1202 13:49:05.300116 137274321021824 utils.py:1231] [64250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 111.94753301644889
+I1202 13:49:05.300163 137274321021824 utils.py:1231] [64250] core_hours = 111.94753301644889
+I1202 13:49:05.300234 137274321021824 train.py:125] NOTE: Steps:64250/112603 [57.1%]
+Walltime:4d15h58m (0s eval)
+ETA:3d12h15m
+Total train time:8d4h12m
+I1202 13:54:17.085967 137274321021824 utils.py:1231] [64300] l2_params = 287.64737442782877
+I1202 13:54:17.086200 137274321021824 utils.py:1231] [64300] train/loss = 2.1593145728111267
+I1202 13:54:17.086315 137274321021824 utils.py:1231] [64300] l2_grads = 1.7698098421096802
+I1202 13:54:17.086399 137274321021824 utils.py:1231] [64300] lr = 0.00045417429897514846
+I1202 13:54:17.086459 137274321021824 utils.py:1231] [64300] uptime = 403446.44882086397
+I1202 13:54:17.086520 137274321021824 utils.py:1231] [64300] examples_seen = 65843200.0
+I1202 13:54:17.086592 137274321021824 utils.py:1231] [64300] progress = 0.5710327433549728
+I1202 13:54:17.086644 137274321021824 utils.py:1231] [64300] epoch = 51.39314390707847
+I1202 13:54:17.086704 137274321021824 utils.py:1231] [64300] img/sec/core = 164.2148864722801
+I1202 13:54:17.086764 137274321021824 utils.py:1231] [64300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.03414040343888
+I1202 13:54:17.086819 137274321021824 utils.py:1231] [64300] core_hours = 112.03414040343888
+I1202 13:54:17.086887 137274321021824 train.py:125] NOTE: Steps:64300/112603 [57.1%]
+Walltime:4d16h4m (0s eval)
+ETA:3d12h9m
+Total train time:8d4h12m
+I1202 13:59:28.862428 137274321021824 utils.py:1231] [64350] l2_params = 287.5576888954411
+I1202 13:59:28.862652 137274321021824 utils.py:1231] [64350] train/loss = 4.555555641651154
+I1202 13:59:28.862756 137274321021824 utils.py:1231] [64350] l2_grads = 1.7204738855361938
+I1202 13:59:28.862818 137274321021824 utils.py:1231] [64350] lr = 0.0004534121018305025
+I1202 13:59:28.862872 137274321021824 utils.py:1231] [64350] uptime = 403758.225234002
+I1202 13:59:28.862932 137274321021824 utils.py:1231] [64350] examples_seen = 65894400.0
+I1202 13:59:28.862981 137274321021824 utils.py:1231] [64350] progress = 0.5714767812580482
+I1202 13:59:28.863030 137274321021824 utils.py:1231] [64350] epoch = 51.43310747154742
+I1202 13:59:28.863081 137274321021824 utils.py:1231] [64350] img/sec/core = 164.22024836540018
+I1202 13:59:28.863137 137274321021824 utils.py:1231] [64350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.12074496264388
+I1202 13:59:28.863192 137274321021824 utils.py:1231] [64350] core_hours = 112.12074496264388
+I1202 13:59:28.863254 137274321021824 train.py:125] NOTE: Steps:64350/112603 [57.1%]
+Walltime:4d16h9m (0s eval)
+ETA:3d12h4m
+Total train time:8d4h12m
+I1202 14:04:40.640557 137274321021824 utils.py:1231] [64400] l2_params = 287.47851455701226
+I1202 14:04:40.640771 137274321021824 utils.py:1231] [64400] train/loss = 4.210903912782669
+I1202 14:04:40.640874 137274321021824 utils.py:1231] [64400] l2_grads = 1.5536704063415527
+I1202 14:04:40.640970 137274321021824 utils.py:1231] [64400] lr = 0.0004526500138783212
+I1202 14:04:40.641030 137274321021824 utils.py:1231] [64400] uptime = 404070.00339127897
+I1202 14:04:40.641087 137274321021824 utils.py:1231] [64400] examples_seen = 65945600.0
+I1202 14:04:40.641142 137274321021824 utils.py:1231] [64400] progress = 0.5719208191611236
+I1202 14:04:40.641196 137274321021824 utils.py:1231] [64400] epoch = 51.473071036016385
+I1202 14:04:40.641250 137274321021824 utils.py:1231] [64400] img/sec/core = 164.21932968997385
+I1202 14:04:40.641320 137274321021824 utils.py:1231] [64400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.20735000633194
+I1202 14:04:40.641371 137274321021824 utils.py:1231] [64400] core_hours = 112.20735000633194
+I1202 14:04:40.641451 137274321021824 train.py:125] NOTE: Steps:64400/112603 [57.2%]
+Walltime:4d16h14m (0s eval)
+ETA:3d11h59m
+Total train time:8d4h11m
+I1202 14:09:52.430854 137274321021824 utils.py:1231] [64450] l2_params = 287.3868494138903
+I1202 14:09:52.431079 137274321021824 utils.py:1231] [64450] train/loss = 4.306727588176727
+I1202 14:09:52.431237 137274321021824 utils.py:1231] [64450] l2_grads = 1.5593253374099731
+I1202 14:09:52.431341 137274321021824 utils.py:1231] [64450] lr = 0.0004518880369047819
+I1202 14:09:52.431437 137274321021824 utils.py:1231] [64450] uptime = 404381.79379315797
+I1202 14:09:52.431525 137274321021824 utils.py:1231] [64450] examples_seen = 65996800.0
+I1202 14:09:52.431606 137274321021824 utils.py:1231] [64450] progress = 0.572364857064199
+I1202 14:09:52.431714 137274321021824 utils.py:1231] [64450] epoch = 51.51303460048534
+I1202 14:09:52.431791 137274321021824 utils.py:1231] [64450] img/sec/core = 164.21288048459502
+I1202 14:09:52.431872 137274321021824 utils.py:1231] [64450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.29395845129832
+I1202 14:09:52.431957 137274321021824 utils.py:1231] [64450] core_hours = 112.29395845129832
+I1202 14:09:52.432040 137274321021824 train.py:125] NOTE: Steps:64450/112603 [57.2%]
+Walltime:4d16h19m (0s eval)
+ETA:3d11h54m
+Total train time:8d4h11m
+I1202 14:15:04.203284 137274321021824 utils.py:1231] [64500] l2_params = 287.3062578130632
+I1202 14:15:04.203570 137274321021824 utils.py:1231] [64500] train/loss = 4.547321915626526
+I1202 14:15:04.203732 137274321021824 utils.py:1231] [64500] l2_grads = 1.5310086011886597
+I1202 14:15:04.203829 137274321021824 utils.py:1231] [64500] lr = 0.00045112617269580274
+I1202 14:15:04.203915 137274321021824 utils.py:1231] [64500] uptime = 404693.56627505
+I1202 14:15:04.203976 137274321021824 utils.py:1231] [64500] examples_seen = 66048000.0
+I1202 14:15:04.204033 137274321021824 utils.py:1231] [64500] progress = 0.5728088949672744
+I1202 14:15:04.204091 137274321021824 utils.py:1231] [64500] epoch = 51.552998164954296
+I1202 14:15:04.204149 137274321021824 utils.py:1231] [64500] img/sec/core = 164.22231907475364
+I1202 14:15:04.204209 137274321021824 utils.py:1231] [64500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.38056191849054
+I1202 14:15:04.204264 137274321021824 utils.py:1231] [64500] core_hours = 112.38056191849054
+I1202 14:15:04.204331 137274321021824 train.py:125] NOTE: Steps:64500/112603 [57.3%]
+Walltime:4d16h24m (0s eval)
+ETA:3d11h48m
+Total train time:8d4h11m
+I1202 14:20:15.988697 137274321021824 utils.py:1231] [64550] l2_params = 287.2158171794855
+I1202 14:20:15.988977 137274321021824 utils.py:1231] [64550] train/loss = 2.624923348426819
+I1202 14:20:15.989201 137274321021824 utils.py:1231] [64550] l2_grads = 1.5007908344268799
+I1202 14:20:15.989298 137274321021824 utils.py:1231] [64550] lr = 0.00045036442303703604
+I1202 14:20:15.989362 137274321021824 utils.py:1231] [64550] uptime = 405005.35172399797
+I1202 14:20:15.989422 137274321021824 utils.py:1231] [64550] examples_seen = 66099200.0
+I1202 14:20:15.989474 137274321021824 utils.py:1231] [64550] progress = 0.5732529328703498
+I1202 14:20:15.989532 137274321021824 utils.py:1231] [64550] epoch = 51.59296172942325
+I1202 14:20:15.989592 137274321021824 utils.py:1231] [64550] img/sec/core = 164.2154891216317
+I1202 14:20:15.989651 137274321021824 utils.py:1231] [64550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.46716898764276
+I1202 14:20:15.989705 137274321021824 utils.py:1231] [64550] core_hours = 112.46716898764276
+I1202 14:20:15.989768 137274321021824 train.py:125] NOTE: Steps:64550/112603 [57.3%]
+Walltime:4d16h30m (0s eval)
+ETA:3d11h43m
+Total train time:8d4h11m
+I1202 14:25:27.773202 137274321021824 utils.py:1231] [64600] l2_params = 287.13488281218133
+I1202 14:25:27.773424 137274321021824 utils.py:1231] [64600] train/loss = 4.365931749343872
+I1202 14:25:27.773530 137274321021824 utils.py:1231] [64600] l2_grads = 1.6118942499160767
+I1202 14:25:27.773590 137274321021824 utils.py:1231] [64600] lr = 0.00044960278971386717
+I1202 14:25:27.773643 137274321021824 utils.py:1231] [64600] uptime = 405317.136005113
+I1202 14:25:27.773697 137274321021824 utils.py:1231] [64600] examples_seen = 66150400.0
+I1202 14:25:27.773752 137274321021824 utils.py:1231] [64600] progress = 0.5736969707734252
+I1202 14:25:27.773799 137274321021824 utils.py:1231] [64600] epoch = 51.63292529389221
+I1202 14:25:27.773851 137274321021824 utils.py:1231] [64600] img/sec/core = 164.21610421439857
+I1202 14:25:27.773914 137274321021824 utils.py:1231] [64600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.55377573239694
+I1202 14:25:27.773964 137274321021824 utils.py:1231] [64600] core_hours = 112.55377573239694
+I1202 14:25:27.774026 137274321021824 train.py:125] NOTE: Steps:64600/112603 [57.4%]
+Walltime:4d16h35m (0s eval)
+ETA:3d11h38m
+Total train time:8d4h11m
+I1202 14:30:39.515330 137274321021824 utils.py:1231] [64650] l2_params = 287.0430080573215
+I1202 14:30:39.515563 137274321021824 utils.py:1231] [64650] train/loss = 2.143884599208832
+I1202 14:30:39.515698 137274321021824 utils.py:1231] [64650] l2_grads = 1.7910069227218628
+I1202 14:30:39.515796 137274321021824 utils.py:1231] [64650] lr = 0.00044884127451140775
+I1202 14:30:39.515898 137274321021824 utils.py:1231] [64650] uptime = 405628.87824142596
+I1202 14:30:39.516007 137274321021824 utils.py:1231] [64650] examples_seen = 66201600.0
+I1202 14:30:39.516072 137274321021824 utils.py:1231] [64650] progress = 0.5741410086765006
+I1202 14:30:39.516139 137274321021824 utils.py:1231] [64650] epoch = 51.67288885836117
+I1202 14:30:39.516205 137274321021824 utils.py:1231] [64650] img/sec/core = 164.23825210711118
+I1202 14:30:39.516273 137274321021824 utils.py:1231] [64650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.64037079803943
+I1202 14:30:39.516345 137274321021824 utils.py:1231] [64650] core_hours = 112.64037079803943
+I1202 14:30:39.516453 137274321021824 train.py:125] NOTE: Steps:64650/112603 [57.4%]
+Walltime:4d16h40m (0s eval)
+ETA:3d11h33m
+Total train time:8d4h11m
+I1202 14:35:51.292921 137274321021824 utils.py:1231] [64700] l2_params = 286.95729538369005
+I1202 14:35:51.293136 137274321021824 utils.py:1231] [64700] train/loss = 3.8589162826538086
+I1202 14:35:51.293236 137274321021824 utils.py:1231] [64700] l2_grads = 1.5579763650894165
+I1202 14:35:51.293308 137274321021824 utils.py:1231] [64700] lr = 0.0004480798792144932
+I1202 14:35:51.293372 137274321021824 utils.py:1231] [64700] uptime = 405940.655730266
+I1202 14:35:51.293434 137274321021824 utils.py:1231] [64700] examples_seen = 66252800.0
+I1202 14:35:51.293491 137274321021824 utils.py:1231] [64700] progress = 0.574585046579576
+I1202 14:35:51.293548 137274321021824 utils.py:1231] [64700] epoch = 51.712852422830125
+I1202 14:35:51.293607 137274321021824 utils.py:1231] [64700] img/sec/core = 164.21968176883897
+I1202 14:35:51.293673 137274321021824 utils.py:1231] [64700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.72697565605056
+I1202 14:35:51.293731 137274321021824 utils.py:1231] [64700] core_hours = 112.72697565605056
+I1202 14:35:51.293800 137274321021824 train.py:125] NOTE: Steps:64700/112603 [57.5%]
+Walltime:4d16h45m (0s eval)
+ETA:3d11h27m
+Total train time:8d4h11m
+I1202 14:41:03.073909 137274321021824 utils.py:1231] [64750] l2_params = 286.8571716041953
+I1202 14:41:03.074170 137274321021824 utils.py:1231] [64750] train/loss = 2.087027847766876
+I1202 14:41:03.074306 137274321021824 utils.py:1231] [64750] l2_grads = 1.7363337278366089
+I1202 14:41:03.074393 137274321021824 utils.py:1231] [64750] lr = 0.00044731860560767825
+I1202 14:41:03.074465 137274321021824 utils.py:1231] [64750] uptime = 406252.436827383
+I1202 14:41:03.074543 137274321021824 utils.py:1231] [64750] examples_seen = 66304000.0
+I1202 14:41:03.074605 137274321021824 utils.py:1231] [64750] progress = 0.5750290844826514
+I1202 14:41:03.074666 137274321021824 utils.py:1231] [64750] epoch = 51.75281598729908
+I1202 14:41:03.074729 137274321021824 utils.py:1231] [64750] img/sec/core = 164.21778123638808
+I1202 14:41:03.074788 137274321021824 utils.py:1231] [64750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.81358151636083
+I1202 14:41:03.074842 137274321021824 utils.py:1231] [64750] core_hours = 112.81358151636083
+I1202 14:41:03.074912 137274321021824 train.py:125] NOTE: Steps:64750/112603 [57.5%]
+Walltime:4d16h50m (0s eval)
+ETA:3d11h22m
+Total train time:8d4h11m
+I1202 14:46:14.832310 137274321021824 utils.py:1231] [64800] l2_params = 286.7756285368468
+I1202 14:46:14.832565 137274321021824 utils.py:1231] [64800] train/loss = 3.6461613476276398
+I1202 14:46:14.832701 137274321021824 utils.py:1231] [64800] l2_grads = 1.5652583837509155
+I1202 14:46:14.832801 137274321021824 utils.py:1231] [64800] lr = 0.0004465574554752308
+I1202 14:46:14.832879 137274321021824 utils.py:1231] [64800] uptime = 406564.195239918
+I1202 14:46:14.832957 137274321021824 utils.py:1231] [64800] examples_seen = 66355200.0
+I1202 14:46:14.833020 137274321021824 utils.py:1231] [64800] progress = 0.5754731223857269
+I1202 14:46:14.833077 137274321021824 utils.py:1231] [64800] epoch = 51.792779551768035
+I1202 14:46:14.833135 137274321021824 utils.py:1231] [64800] img/sec/core = 164.22973026990925
+I1202 14:46:14.833195 137274321021824 utils.py:1231] [64800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.90018107539832
+I1202 14:46:14.833253 137274321021824 utils.py:1231] [64800] core_hours = 112.90018107539832
+I1202 14:46:14.833320 137274321021824 train.py:125] NOTE: Steps:64800/112603 [57.5%]
+Walltime:4d16h56m (0s eval)
+ETA:3d11h17m
+Total train time:8d4h11m
+I1202 14:51:26.623014 137274321021824 utils.py:1231] [64850] l2_params = 286.67610493952935
+I1202 14:51:26.623232 137274321021824 utils.py:1231] [64850] train/loss = 2.238484963774681
+I1202 14:51:26.623378 137274321021824 utils.py:1231] [64850] l2_grads = 1.7804865837097168
+I1202 14:51:26.623447 137274321021824 utils.py:1231] [64850] lr = 0.000445796430601131
+I1202 14:51:26.623500 137274321021824 utils.py:1231] [64850] uptime = 406875.985862475
+I1202 14:51:26.623553 137274321021824 utils.py:1231] [64850] examples_seen = 66406400.0
+I1202 14:51:26.623602 137274321021824 utils.py:1231] [64850] progress = 0.5759171602888022
+I1202 14:51:26.623655 137274321021824 utils.py:1231] [64850] epoch = 51.83274311623699
+I1202 14:51:26.623705 137274321021824 utils.py:1231] [64850] img/sec/core = 164.21276425862226
+I1202 14:51:26.623761 137274321021824 utils.py:1231] [64850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 112.98678958166417
+I1202 14:51:26.623811 137274321021824 utils.py:1231] [64850] core_hours = 112.98678958166417
+I1202 14:51:26.623871 137274321021824 train.py:125] NOTE: Steps:64850/112603 [57.6%]
+Walltime:4d17h1m (0s eval)
+ETA:3d11h12m
+Total train time:8d4h11m
+I1202 14:56:38.422685 137274321021824 utils.py:1231] [64900] l2_params = 286.5841636323081
+I1202 14:56:38.422986 137274321021824 utils.py:1231] [64900] train/loss = 3.450512170791626
+I1202 14:56:38.423192 137274321021824 utils.py:1231] [64900] l2_grads = 1.4556670188903809
+I1202 14:56:38.423306 137274321021824 utils.py:1231] [64900] lr = 0.0004450355327690654
+I1202 14:56:38.423389 137274321021824 utils.py:1231] [64900] uptime = 407187.785730362
+I1202 14:56:38.423451 137274321021824 utils.py:1231] [64900] examples_seen = 66457600.0
+I1202 14:56:38.423538 137274321021824 utils.py:1231] [64900] progress = 0.5763611981918777
+I1202 14:56:38.423597 137274321021824 utils.py:1231] [64900] epoch = 51.87270668070595
+I1202 14:56:38.423699 137274321021824 utils.py:1231] [64900] img/sec/core = 164.2078951058339
+I1202 14:56:38.423819 137274321021824 utils.py:1231] [64900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.07340065607721
+I1202 14:56:38.423916 137274321021824 utils.py:1231] [64900] core_hours = 113.07340065607721
+I1202 14:56:38.424020 137274321021824 train.py:125] NOTE: Steps:64900/112603 [57.6%]
+Walltime:4d17h6m (0s eval)
+ETA:3d11h6m
+Total train time:8d4h11m
+I1202 15:01:50.057645 137274321021824 utils.py:1231] [64950] l2_params = 286.49588529642847
+I1202 15:01:50.057940 137274321021824 utils.py:1231] [64950] train/loss = 4.491183221340179
+I1202 15:01:50.058057 137274321021824 utils.py:1231] [64950] l2_grads = 1.8349931240081787
+I1202 15:01:50.058129 137274321021824 utils.py:1231] [64950] lr = 0.00044427476376242145
+I1202 15:01:50.058189 137274321021824 utils.py:1231] [64950] uptime = 407499.420550432
+I1202 15:01:50.058249 137274321021824 utils.py:1231] [64950] examples_seen = 66508800.0
+I1202 15:01:50.058300 137274321021824 utils.py:1231] [64950] progress = 0.576805236094953
+I1202 15:01:50.058358 137274321021824 utils.py:1231] [64950] epoch = 51.91267024517491
+I1202 15:01:50.058413 137274321021824 utils.py:1231] [64950] img/sec/core = 164.29486277720554
+I1202 15:01:50.058476 137274321021824 utils.py:1231] [64950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.15996588387445
+I1202 15:01:50.058532 137274321021824 utils.py:1231] [64950] core_hours = 113.15996588387445
+I1202 15:01:50.058600 137274321021824 train.py:125] NOTE: Steps:64950/112603 [57.7%]
+Walltime:4d17h11m (0s eval)
+ETA:3d11h1m
+Total train time:8d4h11m
+I1202 15:07:01.957366 137274321021824 utils.py:1231] [65000] l2_params = 286.40257317679743
+I1202 15:07:01.957577 137274321021824 utils.py:1231] [65000] train/loss = 2.8223826587200165
+I1202 15:07:01.957684 137274321021824 utils.py:1231] [65000] l2_grads = 1.6245851516723633
+I1202 15:07:01.957756 137274321021824 utils.py:1231] [65000] lr = 0.0004435141253642857
+I1202 15:07:01.957824 137274321021824 utils.py:1231] [65000] uptime = 407811.320185368
+I1202 15:07:01.957890 137274321021824 utils.py:1231] [65000] examples_seen = 66560000.0
+I1202 15:07:01.957951 137274321021824 utils.py:1231] [65000] progress = 0.5772492739980285
+I1202 15:07:01.958015 137274321021824 utils.py:1231] [65000] epoch = 51.952633809643864
+I1202 15:07:01.958074 137274321021824 utils.py:1231] [65000] img/sec/core = 164.15537007763533
+I1202 15:07:01.958137 137274321021824 utils.py:1231] [65000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.24660467135666
+I1202 15:07:01.958193 137274321021824 utils.py:1231] [65000] core_hours = 113.24660467135666
+I1202 15:07:01.958260 137274321021824 train.py:125] NOTE: Steps:65000/112603 [57.7%]
+Walltime:4d17h16m (0s eval)
+ETA:3d10h56m
+Total train time:8d4h11m
+I1202 15:07:02.333657 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:65000/112603 [57.7%]
+Walltime:4d17h16m (0s eval)
+ETA:3d10h56m
+Total train time:8d4h11m
+I1202 15:08:40.163440 137274321021824 utils.py:1231] [65000] val/acc@1 = 0.6830955038265306
+I1202 15:08:40.163717 137274321021824 utils.py:1231] [65000] val/loss = 1.2901717313394254
+I1202 15:08:40.163903 137274321021824 utils.py:1231] [65000] z/secs/eval/val = 97.82998157897964
+I1202 15:08:40.163976 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.82998157897964
+I1202 15:13:51.606774 137274321021824 utils.py:1231] [65050] l2_params = 286.3107537326882
+I1202 15:13:51.606983 137274321021824 utils.py:1231] [65050] train/loss = 2.2702121138572693
+I1202 15:13:51.607093 137274321021824 utils.py:1231] [65050] l2_grads = 1.8118841648101807
+I1202 15:13:51.607163 137274321021824 utils.py:1231] [65050] lr = 0.0004427536193574383
+I1202 15:13:51.607250 137274321021824 utils.py:1231] [65050] uptime = 408220.969606217
+I1202 15:13:51.607327 137274321021824 utils.py:1231] [65050] examples_seen = 66611200.0
+I1202 15:13:51.607392 137274321021824 utils.py:1231] [65050] progress = 0.5776933119011038
+I1202 15:13:51.607450 137274321021824 utils.py:1231] [65050] epoch = 51.99259737411282
+I1202 15:13:51.607510 137274321021824 utils.py:1231] [65050] img/sec/core = 124.98491977332503
+I1202 15:13:51.607574 137274321021824 utils.py:1231] [65050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.36039617714805
+I1202 15:13:51.607637 137274321021824 utils.py:1231] [65050] core_hours = 113.36039617714805
+I1202 15:13:51.607704 137274321021824 train.py:125] NOTE: Steps:65050/112603 [57.8%]
+Walltime:4d17h23m (0s eval)
+ETA:3d10h52m
+Total train time:8d4h14m
+I1202 15:19:03.301081 137274321021824 utils.py:1231] [65100] l2_params = 286.22642416671783
+I1202 15:19:03.301325 137274321021824 utils.py:1231] [65100] train/loss = 2.3760098814964294
+I1202 15:19:03.301444 137274321021824 utils.py:1231] [65100] l2_grads = 1.8414989709854126
+I1202 15:19:03.301515 137274321021824 utils.py:1231] [65100] lr = 0.0004419932475243492
+I1202 15:19:03.301576 137274321021824 utils.py:1231] [65100] uptime = 408532.6639371
+I1202 15:19:03.301638 137274321021824 utils.py:1231] [65100] examples_seen = 66662400.0
+I1202 15:19:03.301711 137274321021824 utils.py:1231] [65100] progress = 0.5781373498041793
+I1202 15:19:03.301784 137274321021824 utils.py:1231] [65100] epoch = 52.032560938581774
+I1202 15:19:03.301852 137274321021824 utils.py:1231] [65100] img/sec/core = 164.2634944785768
+I1202 15:19:03.301927 137274321021824 utils.py:1231] [65100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.44697793572665
+I1202 15:19:03.301982 137274321021824 utils.py:1231] [65100] core_hours = 113.44697793572665
+I1202 15:19:03.302050 137274321021824 train.py:125] NOTE: Steps:65100/112603 [57.8%]
+Walltime:4d17h28m (0s eval)
+ETA:3d10h47m
+Total train time:8d4h14m
+I1202 15:24:15.029952 137274321021824 utils.py:1231] [65150] l2_params = 286.1334282920903
+I1202 15:24:15.030177 137274321021824 utils.py:1231] [65150] train/loss = 4.148439168930054
+I1202 15:24:15.030287 137274321021824 utils.py:1231] [65150] l2_grads = 1.6109766960144043
+I1202 15:24:15.030357 137274321021824 utils.py:1231] [65150] lr = 0.00044123301164717397
+I1202 15:24:15.030416 137274321021824 utils.py:1231] [65150] uptime = 408844.39277439
+I1202 15:24:15.030467 137274321021824 utils.py:1231] [65150] examples_seen = 66713600.0
+I1202 15:24:15.030515 137274321021824 utils.py:1231] [65150] progress = 0.5785813877072546
+I1202 15:24:15.030562 137274321021824 utils.py:1231] [65150] epoch = 52.07252450305074
+I1202 15:24:15.030613 137274321021824 utils.py:1231] [65150] img/sec/core = 164.24531155058085
+I1202 15:24:15.030673 137274321021824 utils.py:1231] [65150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.53356927941833
+I1202 15:24:15.030724 137274321021824 utils.py:1231] [65150] core_hours = 113.53356927941833
+I1202 15:24:15.030782 137274321021824 train.py:125] NOTE: Steps:65150/112603 [57.9%]
+Walltime:4d17h34m (0s eval)
+ETA:3d10h41m
+Total train time:8d4h14m
+I1202 15:29:26.813228 137274321021824 utils.py:1231] [65200] l2_params = 286.02760408912513
+I1202 15:29:26.813496 137274321021824 utils.py:1231] [65200] train/loss = 2.1629963517189026
+I1202 15:29:26.813625 137274321021824 utils.py:1231] [65200] l2_grads = 1.705582857131958
+I1202 15:29:26.813705 137274321021824 utils.py:1231] [65200] lr = 0.0004404729135077492
+I1202 15:29:26.813768 137274321021824 utils.py:1231] [65200] uptime = 409156.176129782
+I1202 15:29:26.813827 137274321021824 utils.py:1231] [65200] examples_seen = 66764800.0
+I1202 15:29:26.813877 137274321021824 utils.py:1231] [65200] progress = 0.5790254256103301
+I1202 15:29:26.813939 137274321021824 utils.py:1231] [65200] epoch = 52.11248806751969
+I1202 15:29:26.813991 137274321021824 utils.py:1231] [65200] img/sec/core = 164.2165917921707
+I1202 15:29:26.814049 137274321021824 utils.py:1231] [65200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.6201757670272
+I1202 15:29:26.814103 137274321021824 utils.py:1231] [65200] core_hours = 113.6201757670272
+I1202 15:29:26.814165 137274321021824 train.py:125] NOTE: Steps:65200/112603 [57.9%]
+Walltime:4d17h39m (0s eval)
+ETA:3d10h36m
+Total train time:8d4h13m
+I1202 15:34:38.604668 137274321021824 utils.py:1231] [65250] l2_params = 285.948645519808
+I1202 15:34:38.604943 137274321021824 utils.py:1231] [65250] train/loss = 4.3375661969184875
+I1202 15:34:38.605113 137274321021824 utils.py:1231] [65250] l2_grads = 1.6358094215393066
+I1202 15:34:38.605189 137274321021824 utils.py:1231] [65250] lr = 0.00043971295488758825
+I1202 15:34:38.605257 137274321021824 utils.py:1231] [65250] uptime = 409467.967611989
+I1202 15:34:38.605328 137274321021824 utils.py:1231] [65250] examples_seen = 66816000.0
+I1202 15:34:38.605400 137274321021824 utils.py:1231] [65250] progress = 0.5794694635134056
+I1202 15:34:38.605458 137274321021824 utils.py:1231] [65250] epoch = 52.15245163198865
+I1202 15:34:38.605517 137274321021824 utils.py:1231] [65250] img/sec/core = 164.21231150248488
+I1202 15:34:38.605580 137274321021824 utils.py:1231] [65250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.70678451208472
+I1202 15:34:38.605640 137274321021824 utils.py:1231] [65250] core_hours = 113.70678451208472
+I1202 15:34:38.605709 137274321021824 train.py:125] NOTE: Steps:65250/112603 [57.9%]
+Walltime:4d17h44m (0s eval)
+ETA:3d10h31m
+Total train time:8d4h13m
+I1202 15:39:50.403491 137274321021824 utils.py:1231] [65300] l2_params = 285.8472320726773
+I1202 15:39:50.403757 137274321021824 utils.py:1231] [65300] train/loss = 2.0533299893140793
+I1202 15:39:50.403854 137274321021824 utils.py:1231] [65300] l2_grads = 1.6967780590057373
+I1202 15:39:50.403920 137274321021824 utils.py:1231] [65300] lr = 0.0004389531375678787
+I1202 15:39:50.403971 137274321021824 utils.py:1231] [65300] uptime = 409779.766333438
+I1202 15:39:50.404024 137274321021824 utils.py:1231] [65300] examples_seen = 66867200.0
+I1202 15:39:50.404073 137274321021824 utils.py:1231] [65300] progress = 0.5799135014164809
+I1202 15:39:50.404121 137274321021824 utils.py:1231] [65300] epoch = 52.1924151964576
+I1202 15:39:50.404170 137274321021824 utils.py:1231] [65300] img/sec/core = 164.2084988740958
+I1202 15:39:50.404226 137274321021824 utils.py:1231] [65300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.79339526804277
+I1202 15:39:50.404275 137274321021824 utils.py:1231] [65300] core_hours = 113.79339526804277
+I1202 15:39:50.404337 137274321021824 train.py:125] NOTE: Steps:65300/112603 [58.0%]
+Walltime:4d17h49m (0s eval)
+ETA:3d10h26m
+Total train time:8d4h13m
+I1202 15:45:02.196799 137274321021824 utils.py:1231] [65350] l2_params = 285.7697460153365
+I1202 15:45:02.197102 137274321021824 utils.py:1231] [65350] train/loss = 3.6899674832820892
+I1202 15:45:02.197292 137274321021824 utils.py:1231] [65350] l2_grads = 1.5473049879074097
+I1202 15:45:02.197388 137274321021824 utils.py:1231] [65350] lr = 0.00043819346332947594
+I1202 15:45:02.197460 137274321021824 utils.py:1231] [65350] uptime = 410091.55981782096
+I1202 15:45:02.197531 137274321021824 utils.py:1231] [65350] examples_seen = 66918400.0
+I1202 15:45:02.197603 137274321021824 utils.py:1231] [65350] progress = 0.5803575393195564
+I1202 15:45:02.197667 137274321021824 utils.py:1231] [65350] epoch = 52.23237876092656
+I1202 15:45:02.197733 137274321021824 utils.py:1231] [65350] img/sec/core = 164.21125701623623
+I1202 15:45:02.197795 137274321021824 utils.py:1231] [65350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.88000456926027
+I1202 15:45:02.197852 137274321021824 utils.py:1231] [65350] core_hours = 113.88000456926027
+I1202 15:45:02.197948 137274321021824 train.py:125] NOTE: Steps:65350/112603 [58.0%]
+Walltime:4d17h54m (0s eval)
+ETA:3d10h20m
+Total train time:8d4h13m
+I1202 15:50:14.029696 137274321021824 utils.py:1231] [65400] l2_params = 285.6753310646674
+I1202 15:50:14.029930 137274321021824 utils.py:1231] [65400] train/loss = 2.152885854244232
+I1202 15:50:14.030084 137274321021824 utils.py:1231] [65400] l2_grads = 1.6960151195526123
+I1202 15:50:14.030192 137274321021824 utils.py:1231] [65400] lr = 0.0004374339339529008
+I1202 15:50:14.030283 137274321021824 utils.py:1231] [65400] uptime = 410403.392638584
+I1202 15:50:14.030378 137274321021824 utils.py:1231] [65400] examples_seen = 66969600.0
+I1202 15:50:14.030457 137274321021824 utils.py:1231] [65400] progress = 0.5808015772226317
+I1202 15:50:14.030540 137274321021824 utils.py:1231] [65400] epoch = 52.27234232539552
+I1202 15:50:14.030618 137274321021824 utils.py:1231] [65400] img/sec/core = 164.19054246667693
+I1202 15:50:14.030695 137274321021824 utils.py:1231] [65400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 113.96662479725
+I1202 15:50:14.030765 137274321021824 utils.py:1231] [65400] core_hours = 113.96662479725
+I1202 15:50:14.030896 137274321021824 train.py:125] NOTE: Steps:65400/112603 [58.1%]
+Walltime:4d18h0m (0s eval)
+ETA:3d10h15m
+Total train time:8d4h13m
+I1202 15:55:25.825958 137274321021824 utils.py:1231] [65450] l2_params = 285.5819726898593
+I1202 15:55:25.826206 137274321021824 utils.py:1231] [65450] train/loss = 4.171525776386261
+I1202 15:55:25.826326 137274321021824 utils.py:1231] [65450] l2_grads = 1.5625851154327393
+I1202 15:55:25.826397 137274321021824 utils.py:1231] [65450] lr = 0.0004366745512183338
+I1202 15:55:25.826458 137274321021824 utils.py:1231] [65450] uptime = 410715.188816965
+I1202 15:55:25.826525 137274321021824 utils.py:1231] [65450] examples_seen = 67020800.0
+I1202 15:55:25.826573 137274321021824 utils.py:1231] [65450] progress = 0.5812456151257072
+I1202 15:55:25.826648 137274321021824 utils.py:1231] [65450] epoch = 52.312305889864476
+I1202 15:55:25.826698 137274321021824 utils.py:1231] [65450] img/sec/core = 164.20983818936622
+I1202 15:55:25.826754 137274321021824 utils.py:1231] [65450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.05323484680027
+I1202 15:55:25.826804 137274321021824 utils.py:1231] [65450] core_hours = 114.05323484680027
+I1202 15:55:25.826862 137274321021824 train.py:125] NOTE: Steps:65450/112603 [58.1%]
+Walltime:4d18h5m (0s eval)
+ETA:3d10h10m
+Total train time:8d4h13m
+I1202 16:00:37.616121 137274321021824 utils.py:1231] [65500] l2_params = 285.48757735410464
+I1202 16:00:37.616362 137274321021824 utils.py:1231] [65500] train/loss = 2.62446466088295
+I1202 16:00:37.616489 137274321021824 utils.py:1231] [65500] l2_grads = 1.5753381252288818
+I1202 16:00:37.616572 137274321021824 utils.py:1231] [65500] lr = 0.0004359153169056123
+I1202 16:00:37.616645 137274321021824 utils.py:1231] [65500] uptime = 411026.979006189
+I1202 16:00:37.616730 137274321021824 utils.py:1231] [65500] examples_seen = 67072000.0
+I1202 16:00:37.616796 137274321021824 utils.py:1231] [65500] progress = 0.5816896530287825
+I1202 16:00:37.616854 137274321021824 utils.py:1231] [65500] epoch = 52.35226945433343
+I1202 16:00:37.616921 137274321021824 utils.py:1231] [65500] img/sec/core = 164.21299248519838
+I1202 16:00:37.616984 137274321021824 utils.py:1231] [65500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.13984323269582
+I1202 16:00:37.617036 137274321021824 utils.py:1231] [65500] core_hours = 114.13984323269582
+I1202 16:00:37.617101 137274321021824 train.py:125] NOTE: Steps:65500/112603 [58.2%]
+Walltime:4d18h10m (0s eval)
+ETA:3d10h5m
+Total train time:8d4h13m
+I1202 16:05:49.402126 137274321021824 utils.py:1231] [65550] l2_params = 285.4044636635287
+I1202 16:05:49.402376 137274321021824 utils.py:1231] [65550] train/loss = 3.924117147922516
+I1202 16:05:49.402485 137274321021824 utils.py:1231] [65550] l2_grads = 1.750503659248352
+I1202 16:05:49.402560 137274321021824 utils.py:1231] [65550] lr = 0.00043515623279422527
+I1202 16:05:49.402614 137274321021824 utils.py:1231] [65550] uptime = 411338.76497685496
+I1202 16:05:49.402674 137274321021824 utils.py:1231] [65550] examples_seen = 67123200.0
+I1202 16:05:49.402730 137274321021824 utils.py:1231] [65550] progress = 0.582133690931858
+I1202 16:05:49.402777 137274321021824 utils.py:1231] [65550] epoch = 52.392233018802386
+I1202 16:05:49.402826 137274321021824 utils.py:1231] [65550] img/sec/core = 164.21521433641138
+I1202 16:05:49.402892 137274321021824 utils.py:1231] [65550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.2264504467697
+I1202 16:05:49.402942 137274321021824 utils.py:1231] [65550] core_hours = 114.2264504467697
+I1202 16:05:49.403001 137274321021824 train.py:125] NOTE: Steps:65550/112603 [58.2%]
+Walltime:4d18h15m (0s eval)
+ETA:3d9h59m
+Total train time:8d4h13m
+I1202 16:11:01.176722 137274321021824 utils.py:1231] [65600] l2_params = 285.3091165597901
+I1202 16:11:01.176988 137274321021824 utils.py:1231] [65600] train/loss = 2.3003524243831635
+I1202 16:11:01.177118 137274321021824 utils.py:1231] [65600] l2_grads = 1.9388668537139893
+I1202 16:11:01.177217 137274321021824 utils.py:1231] [65600] lr = 0.00043439730066331027
+I1202 16:11:01.177324 137274321021824 utils.py:1231] [65600] uptime = 411650.539668243
+I1202 16:11:01.177386 137274321021824 utils.py:1231] [65600] examples_seen = 67174400.0
+I1202 16:11:01.177455 137274321021824 utils.py:1231] [65600] progress = 0.5825777288349333
+I1202 16:11:01.177517 137274321021824 utils.py:1231] [65600] epoch = 52.43219658327135
+I1202 16:11:01.177590 137274321021824 utils.py:1231] [65600] img/sec/core = 164.22115525817452
+I1202 16:11:01.177662 137274321021824 utils.py:1231] [65600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.31305452771082
+I1202 16:11:01.177726 137274321021824 utils.py:1231] [65600] core_hours = 114.31305452771082
+I1202 16:11:01.177796 137274321021824 train.py:125] NOTE: Steps:65600/112603 [58.3%]
+Walltime:4d18h20m (0s eval)
+ETA:3d9h54m
+Total train time:8d4h13m
+I1202 16:16:12.959743 137274321021824 utils.py:1231] [65650] l2_params = 285.2118979098349
+I1202 16:16:12.959990 137274321021824 utils.py:1231] [65650] train/loss = 2.3127986043691635
+I1202 16:16:12.960099 137274321021824 utils.py:1231] [65650] l2_grads = 1.7105003595352173
+I1202 16:16:12.960174 137274321021824 utils.py:1231] [65650] lr = 0.0004336385222916484
+I1202 16:16:12.960237 137274321021824 utils.py:1231] [65650] uptime = 411962.32259815
+I1202 16:16:12.960299 137274321021824 utils.py:1231] [65650] examples_seen = 67225600.0
+I1202 16:16:12.960357 137274321021824 utils.py:1231] [65650] progress = 0.5830217667380088
+I1202 16:16:12.960414 137274321021824 utils.py:1231] [65650] epoch = 52.472160147740304
+I1202 16:16:12.960476 137274321021824 utils.py:1231] [65650] img/sec/core = 164.2168158958138
+I1202 16:16:12.960540 137274321021824 utils.py:1231] [65650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.39966089712945
+I1202 16:16:12.960599 137274321021824 utils.py:1231] [65650] core_hours = 114.39966089712945
+I1202 16:16:12.960668 137274321021824 train.py:125] NOTE: Steps:65650/112603 [58.3%]
+Walltime:4d18h26m (0s eval)
+ETA:3d9h49m
+Total train time:8d4h13m
+I1202 16:21:24.747037 137274321021824 utils.py:1231] [65700] l2_params = 285.13286770184254
+I1202 16:21:24.747286 137274321021824 utils.py:1231] [65700] train/loss = 3.3671757876873016
+I1202 16:21:24.747446 137274321021824 utils.py:1231] [65700] l2_grads = 1.4981038570404053
+I1202 16:21:24.747539 137274321021824 utils.py:1231] [65700] lr = 0.00043287989945765994
+I1202 16:21:24.747599 137274321021824 utils.py:1231] [65700] uptime = 412274.109960437
+I1202 16:21:24.747659 137274321021824 utils.py:1231] [65700] examples_seen = 67276800.0
+I1202 16:21:24.747715 137274321021824 utils.py:1231] [65700] progress = 0.5834658046410842
+I1202 16:21:24.747771 137274321021824 utils.py:1231] [65700] epoch = 52.51212371220926
+I1202 16:21:24.747835 137274321021824 utils.py:1231] [65700] img/sec/core = 164.2144813838579
+I1202 16:21:24.747914 137274321021824 utils.py:1231] [65700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.48626849776473
+I1202 16:21:24.747971 137274321021824 utils.py:1231] [65700] core_hours = 114.48626849776473
+I1202 16:21:24.748038 137274321021824 train.py:125] NOTE: Steps:65700/112603 [58.3%]
+Walltime:4d18h31m (0s eval)
+ETA:3d9h44m
+Total train time:8d4h13m
+I1202 16:26:36.529048 137274321021824 utils.py:1231] [65750] l2_params = 285.03944901603137
+I1202 16:26:36.529291 137274321021824 utils.py:1231] [65750] train/loss = 2.066580906510353
+I1202 16:26:36.529420 137274321021824 utils.py:1231] [65750] l2_grads = 1.749709129333496
+I1202 16:26:36.529494 137274321021824 utils.py:1231] [65750] lr = 0.00043212143393940106
+I1202 16:26:36.529561 137274321021824 utils.py:1231] [65750] uptime = 412585.89192242397
+I1202 16:26:36.529628 137274321021824 utils.py:1231] [65750] examples_seen = 67328000.0
+I1202 16:26:36.529678 137274321021824 utils.py:1231] [65750] progress = 0.5839098425441596
+I1202 16:26:36.529731 137274321021824 utils.py:1231] [65750] epoch = 52.552087276678215
+I1202 16:26:36.529784 137274321021824 utils.py:1231] [65750] img/sec/core = 164.21732570320987
+I1202 16:26:36.529846 137274321021824 utils.py:1231] [65750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.57287459831666
+I1202 16:26:36.529904 137274321021824 utils.py:1231] [65750] core_hours = 114.57287459831666
+I1202 16:26:36.529976 137274321021824 train.py:125] NOTE: Steps:65750/112603 [58.4%]
+Walltime:4d18h36m (0s eval)
+ETA:3d9h38m
+Total train time:8d4h13m
+I1202 16:31:48.304953 137274321021824 utils.py:1231] [65800] l2_params = 284.95631323845004
+I1202 16:31:48.305173 137274321021824 utils.py:1231] [65800] train/loss = 2.176051437854767
+I1202 16:31:48.305275 137274321021824 utils.py:1231] [65800] l2_grads = 1.8093867301940918
+I1202 16:31:48.305343 137274321021824 utils.py:1231] [65800] lr = 0.00043136312751455906
+I1202 16:31:48.305400 137274321021824 utils.py:1231] [65800] uptime = 412897.66776127
+I1202 16:31:48.305461 137274321021824 utils.py:1231] [65800] examples_seen = 67379200.0
+I1202 16:31:48.305516 137274321021824 utils.py:1231] [65800] progress = 0.584353880447235
+I1202 16:31:48.305571 137274321021824 utils.py:1231] [65800] epoch = 52.59205084114717
+I1202 16:31:48.305627 137274321021824 utils.py:1231] [65800] img/sec/core = 164.22055085956097
+I1202 16:31:48.305689 137274321021824 utils.py:1231] [65800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.6594789979961
+I1202 16:31:48.305745 137274321021824 utils.py:1231] [65800] core_hours = 114.6594789979961
+I1202 16:31:48.305810 137274321021824 train.py:125] NOTE: Steps:65800/112603 [58.4%]
+Walltime:4d18h41m (0s eval)
+ETA:3d9h33m
+Total train time:8d4h13m
+I1202 16:37:00.099044 137274321021824 utils.py:1231] [65850] l2_params = 284.87388349940306
+I1202 16:37:00.099261 137274321021824 utils.py:1231] [65850] train/loss = 4.443312644958496
+I1202 16:37:00.099354 137274321021824 utils.py:1231] [65850] l2_grads = 1.7775630950927734
+I1202 16:37:00.099412 137274321021824 utils.py:1231] [65850] lr = 0.00043060498196044895
+I1202 16:37:00.099462 137274321021824 utils.py:1231] [65850] uptime = 413209.46182446
+I1202 16:37:00.099521 137274321021824 utils.py:1231] [65850] examples_seen = 67430400.0
+I1202 16:37:00.099568 137274321021824 utils.py:1231] [65850] progress = 0.5847979183503104
+I1202 16:37:00.099615 137274321021824 utils.py:1231] [65850] epoch = 52.63201440561613
+I1202 16:37:00.099667 137274321021824 utils.py:1231] [65850] img/sec/core = 164.2109521783941
+I1202 16:37:00.099722 137274321021824 utils.py:1231] [65850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.74608845999335
+I1202 16:37:00.099779 137274321021824 utils.py:1231] [65850] core_hours = 114.74608845999335
+I1202 16:37:00.099838 137274321021824 train.py:125] NOTE: Steps:65850/112603 [58.5%]
+Walltime:4d18h46m (0s eval)
+ETA:3d9h28m
+Total train time:8d4h13m
+I1202 16:42:11.889092 137274321021824 utils.py:1231] [65900] l2_params = 284.7783087646858
+I1202 16:42:11.889359 137274321021824 utils.py:1231] [65900] train/loss = 4.591119289398193
+I1202 16:42:11.889496 137274321021824 utils.py:1231] [65900] l2_grads = 1.707527756690979
+I1202 16:42:11.889569 137274321021824 utils.py:1231] [65900] lr = 0.0004298469990540078
+I1202 16:42:11.889629 137274321021824 utils.py:1231] [65900] uptime = 413521.251990618
+I1202 16:42:11.889690 137274321021824 utils.py:1231] [65900] examples_seen = 67481600.0
+I1202 16:42:11.889746 137274321021824 utils.py:1231] [65900] progress = 0.5852419562533858
+I1202 16:42:11.889805 137274321021824 utils.py:1231] [65900] epoch = 52.67197797008509
+I1202 16:42:11.889869 137274321021824 utils.py:1231] [65900] img/sec/core = 164.21300463355524
+I1202 16:42:11.889948 137274321021824 utils.py:1231] [65900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.83269683948167
+I1202 16:42:11.890006 137274321021824 utils.py:1231] [65900] core_hours = 114.83269683948167
+I1202 16:42:11.890070 137274321021824 train.py:125] NOTE: Steps:65900/112603 [58.5%]
+Walltime:4d18h52m (0s eval)
+ETA:3d9h23m
+Total train time:8d4h13m
+I1202 16:47:23.683911 137274321021824 utils.py:1231] [65950] l2_params = 284.6894107966987
+I1202 16:47:23.684131 137274321021824 utils.py:1231] [65950] train/loss = 2.186076432466507
+I1202 16:47:23.684247 137274321021824 utils.py:1231] [65950] l2_grads = 1.689918041229248
+I1202 16:47:23.684337 137274321021824 utils.py:1231] [65950] lr = 0.000429089180571792
+I1202 16:47:23.684440 137274321021824 utils.py:1231] [65950] uptime = 413833.046794396
+I1202 16:47:23.684543 137274321021824 utils.py:1231] [65950] examples_seen = 67532800.0
+I1202 16:47:23.684624 137274321021824 utils.py:1231] [65950] progress = 0.5856859941564612
+I1202 16:47:23.684703 137274321021824 utils.py:1231] [65950] epoch = 52.71194153455404
+I1202 16:47:23.684775 137274321021824 utils.py:1231] [65950] img/sec/core = 164.21056213770208
+I1202 16:47:23.684871 137274321021824 utils.py:1231] [65950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 114.91930650719777
+I1202 16:47:23.684975 137274321021824 utils.py:1231] [65950] core_hours = 114.91930650719777
+I1202 16:47:23.685072 137274321021824 train.py:125] NOTE: Steps:65950/112603 [58.6%]
+Walltime:4d18h57m (0s eval)
+ETA:3d9h17m
+Total train time:8d4h13m
+I1202 16:52:35.468276 137274321021824 utils.py:1231] [66000] l2_params = 284.5992361969133
+I1202 16:52:35.468468 137274321021824 utils.py:1231] [66000] train/loss = 3.980801433324814
+I1202 16:52:35.468571 137274321021824 utils.py:1231] [66000] l2_grads = 1.497768759727478
+I1202 16:52:35.468634 137274321021824 utils.py:1231] [66000] lr = 0.00042833152828997264
+I1202 16:52:35.468687 137274321021824 utils.py:1231] [66000] uptime = 414144.831049342
+I1202 16:52:35.468750 137274321021824 utils.py:1231] [66000] examples_seen = 67584000.0
+I1202 16:52:35.468799 137274321021824 utils.py:1231] [66000] progress = 0.5861300320595366
+I1202 16:52:35.468847 137274321021824 utils.py:1231] [66000] epoch = 52.751905099023
+I1202 16:52:35.468904 137274321021824 utils.py:1231] [66000] img/sec/core = 164.216117997586
+I1202 16:52:35.468963 137274321021824 utils.py:1231] [66000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.00591324468277
+I1202 16:52:35.469016 137274321021824 utils.py:1231] [66000] core_hours = 115.00591324468277
+I1202 16:52:35.469078 137274321021824 train.py:125] NOTE: Steps:66000/112603 [58.6%]
+Walltime:4d19h2m (0s eval)
+ETA:3d9h12m
+Total train time:8d4h13m
+I1202 16:57:47.619279 137274321021824 utils.py:1231] [66050] l2_params = 284.50856426514804
+I1202 16:57:47.619524 137274321021824 utils.py:1231] [66050] train/loss = 2.1219318360090256
+I1202 16:57:47.619647 137274321021824 utils.py:1231] [66050] l2_grads = 1.7291014194488525
+I1202 16:57:47.619716 137274321021824 utils.py:1231] [66050] lr = 0.00042757404398433073
+I1202 16:57:47.619775 137274321021824 utils.py:1231] [66050] uptime = 414456.982137092
+I1202 16:57:47.619838 137274321021824 utils.py:1231] [66050] examples_seen = 67635200.0
+I1202 16:57:47.619899 137274321021824 utils.py:1231] [66050] progress = 0.586574069962612
+I1202 16:57:47.619950 137274321021824 utils.py:1231] [66050] epoch = 52.791868663491954
+I1202 16:57:47.620003 137274321021824 utils.py:1231] [66050] img/sec/core = 164.02313497945806
+I1202 16:57:47.620060 137274321021824 utils.py:1231] [66050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.09262188016889
+I1202 16:57:47.620114 137274321021824 utils.py:1231] [66050] core_hours = 115.09262188016889
+I1202 16:57:47.620176 137274321021824 train.py:125] NOTE: Steps:66050/112603 [58.7%]
+Walltime:4d19h7m (0s eval)
+ETA:3d9h7m
+Total train time:8d4h13m
+I1202 17:02:59.333658 137274321021824 utils.py:1231] [66100] l2_params = 284.4171546300607
+I1202 17:02:59.333874 137274321021824 utils.py:1231] [66100] train/loss = 2.425181567668915
+I1202 17:02:59.333976 137274321021824 utils.py:1231] [66100] l2_grads = 1.6462223529815674
+I1202 17:02:59.334043 137274321021824 utils.py:1231] [66100] lr = 0.0004268167294302539
+I1202 17:02:59.334096 137274321021824 utils.py:1231] [66100] uptime = 414768.696458177
+I1202 17:02:59.334149 137274321021824 utils.py:1231] [66100] examples_seen = 67686400.0
+I1202 17:02:59.334200 137274321021824 utils.py:1231] [66100] progress = 0.5870181078656874
+I1202 17:02:59.334249 137274321021824 utils.py:1231] [66100] epoch = 52.831832227960916
+I1202 17:02:59.334300 137274321021824 utils.py:1231] [66100] img/sec/core = 164.25296028036956
+I1202 17:02:59.334357 137274321021824 utils.py:1231] [66100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.17920919158138
+I1202 17:02:59.334408 137274321021824 utils.py:1231] [66100] core_hours = 115.17920919158138
+I1202 17:02:59.334469 137274321021824 train.py:125] NOTE: Steps:66100/112603 [58.7%]
+Walltime:4d19h12m (0s eval)
+ETA:3d9h2m
+Total train time:8d4h12m
+I1202 17:08:11.114329 137274321021824 utils.py:1231] [66150] l2_params = 284.3303108962503
+I1202 17:08:11.114534 137274321021824 utils.py:1231] [66150] train/loss = 2.33352467417717
+I1202 17:08:11.114639 137274321021824 utils.py:1231] [66150] l2_grads = 1.6846082210540771
+I1202 17:08:11.114709 137274321021824 utils.py:1231] [66150] lr = 0.0004260595864027321
+I1202 17:08:11.114768 137274321021824 utils.py:1231] [66150] uptime = 415080.477129676
+I1202 17:08:11.114829 137274321021824 utils.py:1231] [66150] examples_seen = 67737600.0
+I1202 17:08:11.114894 137274321021824 utils.py:1231] [66150] progress = 0.5874621457687628
+I1202 17:08:11.114963 137274321021824 utils.py:1231] [66150] epoch = 52.87179579242987
+I1202 17:08:11.115038 137274321021824 utils.py:1231] [66150] img/sec/core = 164.21800541336307
+I1202 17:08:11.115096 137274321021824 utils.py:1231] [66150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.26581493366443
+I1202 17:08:11.115150 137274321021824 utils.py:1231] [66150] core_hours = 115.26581493366443
+I1202 17:08:11.115217 137274321021824 train.py:125] NOTE: Steps:66150/112603 [58.7%]
+Walltime:4d19h18m (0s eval)
+ETA:3d8h56m
+Total train time:8d4h12m
+I1202 17:13:22.898744 137274321021824 utils.py:1231] [66200] l2_params = 284.2316381724182
+I1202 17:13:22.899112 137274321021824 utils.py:1231] [66200] train/loss = 2.730591803789139
+I1202 17:13:22.899316 137274321021824 utils.py:1231] [66200] l2_grads = 1.643641710281372
+I1202 17:13:22.899435 137274321021824 utils.py:1231] [66200] lr = 0.00042530261667635274
+I1202 17:13:22.899529 137274321021824 utils.py:1231] [66200] uptime = 415392.261887107
+I1202 17:13:22.899603 137274321021824 utils.py:1231] [66200] examples_seen = 67788800.0
+I1202 17:13:22.899695 137274321021824 utils.py:1231] [66200] progress = 0.5879061836718382
+I1202 17:13:22.899796 137274321021824 utils.py:1231] [66200] epoch = 52.91175935689883
+I1202 17:13:22.899870 137274321021824 utils.py:1231] [66200] img/sec/core = 164.21585334019582
+I1202 17:13:22.899957 137274321021824 utils.py:1231] [66200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.3524218107286
+I1202 17:13:22.900018 137274321021824 utils.py:1231] [66200] core_hours = 115.3524218107286
+I1202 17:13:22.900091 137274321021824 train.py:125] NOTE: Steps:66200/112603 [58.8%]
+Walltime:4d19h23m (0s eval)
+ETA:3d8h51m
+Total train time:8d4h12m
+I1202 17:18:34.681633 137274321021824 utils.py:1231] [66250] l2_params = 284.14858162694765
+I1202 17:18:34.681862 137274321021824 utils.py:1231] [66250] train/loss = 3.9752522706985474
+I1202 17:18:34.681977 137274321021824 utils.py:1231] [66250] l2_grads = 1.6206151247024536
+I1202 17:18:34.682062 137274321021824 utils.py:1231] [66250] lr = 0.00042454582202529773
+I1202 17:18:34.682124 137274321021824 utils.py:1231] [66250] uptime = 415704.04448511096
+I1202 17:18:34.682187 137274321021824 utils.py:1231] [66250] examples_seen = 67840000.0
+I1202 17:18:34.682246 137274321021824 utils.py:1231] [66250] progress = 0.5883502215749137
+I1202 17:18:34.682305 137274321021824 utils.py:1231] [66250] epoch = 52.95172292136778
+I1202 17:18:34.682371 137274321021824 utils.py:1231] [66250] img/sec/core = 164.2169907101318
+I1202 17:18:34.682435 137274321021824 utils.py:1231] [66250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.43902808795194
+I1202 17:18:34.682495 137274321021824 utils.py:1231] [66250] core_hours = 115.43902808795194
+I1202 17:18:34.682573 137274321021824 train.py:125] NOTE: Steps:66250/112603 [58.8%]
+Walltime:4d19h28m (0s eval)
+ETA:3d8h46m
+Total train time:8d4h12m
+I1202 17:23:46.453981 137274321021824 utils.py:1231] [66300] l2_params = 284.0544248745646
+I1202 17:23:46.454217 137274321021824 utils.py:1231] [66300] train/loss = 2.6047529578208923
+I1202 17:23:46.454364 137274321021824 utils.py:1231] [66300] l2_grads = 1.6836541891098022
+I1202 17:23:46.454446 137274321021824 utils.py:1231] [66300] lr = 0.00042378920422333816
+I1202 17:23:46.454499 137274321021824 utils.py:1231] [66300] uptime = 416015.816860765
+I1202 17:23:46.454553 137274321021824 utils.py:1231] [66300] examples_seen = 67891200.0
+I1202 17:23:46.454602 137274321021824 utils.py:1231] [66300] progress = 0.588794259477989
+I1202 17:23:46.454652 137274321021824 utils.py:1231] [66300] epoch = 52.99168648583674
+I1202 17:23:46.454708 137274321021824 utils.py:1231] [66300] img/sec/core = 164.22237503433394
+I1202 17:23:46.454765 137274321021824 utils.py:1231] [66300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.5256315256336
+I1202 17:23:46.454817 137274321021824 utils.py:1231] [66300] core_hours = 115.5256315256336
+I1202 17:23:46.454877 137274321021824 train.py:125] NOTE: Steps:66300/112603 [58.9%]
+Walltime:4d19h33m (0s eval)
+ETA:3d8h41m
+Total train time:8d4h12m
+I1202 17:28:58.236856 137274321021824 utils.py:1231] [66350] l2_params = 283.9559392781743
+I1202 17:28:58.237077 137274321021824 utils.py:1231] [66350] train/loss = 2.151275023818016
+I1202 17:28:58.237168 137274321021824 utils.py:1231] [66350] l2_grads = 1.8107812404632568
+I1202 17:28:58.237230 137274321021824 utils.py:1231] [66350] lr = 0.0004230327650438306
+I1202 17:28:58.237281 137274321021824 utils.py:1231] [66350] uptime = 416327.599643383
+I1202 17:28:58.237336 137274321021824 utils.py:1231] [66350] examples_seen = 67942400.0
+I1202 17:28:58.237385 137274321021824 utils.py:1231] [66350] progress = 0.5892382973810645
+I1202 17:28:58.237433 137274321021824 utils.py:1231] [66350] epoch = 53.0316500503057
+I1202 17:28:58.237484 137274321021824 utils.py:1231] [66350] img/sec/core = 164.21689347332708
+I1202 17:28:58.237540 137274321021824 utils.py:1231] [66350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.61223785413861
+I1202 17:28:58.237595 137274321021824 utils.py:1231] [66350] core_hours = 115.61223785413861
+I1202 17:28:58.237656 137274321021824 train.py:125] NOTE: Steps:66350/112603 [58.9%]
+Walltime:4d19h38m (0s eval)
+ETA:3d8h35m
+Total train time:8d4h12m
+I1202 17:34:10.027859 137274321021824 utils.py:1231] [66400] l2_params = 283.86295592530837
+I1202 17:34:10.028152 137274321021824 utils.py:1231] [66400] train/loss = 3.2986644208431244
+I1202 17:34:10.028369 137274321021824 utils.py:1231] [66400] l2_grads = 1.5862128734588623
+I1202 17:34:10.028462 137274321021824 utils.py:1231] [66400] lr = 0.0004222765062597134
+I1202 17:34:10.028523 137274321021824 utils.py:1231] [66400] uptime = 416639.390884967
+I1202 17:34:10.028589 137274321021824 utils.py:1231] [66400] examples_seen = 67993600.0
+I1202 17:34:10.028652 137274321021824 utils.py:1231] [66400] progress = 0.5896823352841398
+I1202 17:34:10.028735 137274321021824 utils.py:1231] [66400] epoch = 53.071613614774655
+I1202 17:34:10.028804 137274321021824 utils.py:1231] [66400] img/sec/core = 164.21243823235085
+I1202 17:34:10.028877 137274321021824 utils.py:1231] [66400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.69884653235638
+I1202 17:34:10.028980 137274321021824 utils.py:1231] [66400] core_hours = 115.69884653235638
+I1202 17:34:10.029049 137274321021824 train.py:125] NOTE: Steps:66400/112603 [59.0%]
+Walltime:4d19h43m (0s eval)
+ETA:3d8h30m
+Total train time:8d4h12m
+I1202 17:39:21.812233 137274321021824 utils.py:1231] [66450] l2_params = 283.78186059507397
+I1202 17:39:21.812491 137274321021824 utils.py:1231] [66450] train/loss = 2.098992183804512
+I1202 17:39:21.812621 137274321021824 utils.py:1231] [66450] l2_grads = 1.793619155883789
+I1202 17:39:21.812713 137274321021824 utils.py:1231] [66450] lr = 0.00042152042964350197
+I1202 17:39:21.812781 137274321021824 utils.py:1231] [66450] uptime = 416951.175141537
+I1202 17:39:21.812855 137274321021824 utils.py:1231] [66450] examples_seen = 68044800.0
+I1202 17:39:21.812921 137274321021824 utils.py:1231] [66450] progress = 0.5901263731872153
+I1202 17:39:21.812979 137274321021824 utils.py:1231] [66450] epoch = 53.11157717924361
+I1202 17:39:21.813037 137274321021824 utils.py:1231] [66450] img/sec/core = 164.21611714223187
+I1202 17:39:21.813093 137274321021824 utils.py:1231] [66450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.78545327029249
+I1202 17:39:21.813148 137274321021824 utils.py:1231] [66450] core_hours = 115.78545327029249
+I1202 17:39:21.813212 137274321021824 train.py:125] NOTE: Steps:66450/112603 [59.0%]
+Walltime:4d19h49m (0s eval)
+ETA:3d8h25m
+Total train time:8d4h12m
+I1202 17:44:33.599047 137274321021824 utils.py:1231] [66500] l2_params = 283.69684521502137
+I1202 17:44:33.599268 137274321021824 utils.py:1231] [66500] train/loss = 1.98407281935215
+I1202 17:44:33.599365 137274321021824 utils.py:1231] [66500] l2_grads = 1.7308297157287598
+I1202 17:44:33.599436 137274321021824 utils.py:1231] [66500] lr = 0.00042076453696728407
+I1202 17:44:33.599494 137274321021824 utils.py:1231] [66500] uptime = 417262.961856017
+I1202 17:44:33.599551 137274321021824 utils.py:1231] [66500] examples_seen = 68096000.0
+I1202 17:44:33.599607 137274321021824 utils.py:1231] [66500] progress = 0.5905704110902906
+I1202 17:44:33.599661 137274321021824 utils.py:1231] [66500] epoch = 53.151540743712566
+I1202 17:44:33.599716 137274321021824 utils.py:1231] [66500] img/sec/core = 164.21482257634835
+I1202 17:44:33.599783 137274321021824 utils.py:1231] [66500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.8720606909814
+I1202 17:44:33.599848 137274321021824 utils.py:1231] [66500] core_hours = 115.8720606909814
+I1202 17:44:33.599933 137274321021824 train.py:125] NOTE: Steps:66500/112603 [59.1%]
+Walltime:4d19h54m (0s eval)
+ETA:3d8h20m
+Total train time:8d4h12m
+I1202 17:49:45.383658 137274321021824 utils.py:1231] [66550] l2_params = 283.6013608626874
+I1202 17:49:45.383872 137274321021824 utils.py:1231] [66550] train/loss = 2.023175448179245
+I1202 17:49:45.384017 137274321021824 utils.py:1231] [66550] l2_grads = 1.7036312818527222
+I1202 17:49:45.384104 137274321021824 utils.py:1231] [66550] lr = 0.0004200088300027172
+I1202 17:49:45.384184 137274321021824 utils.py:1231] [66550] uptime = 417574.746541864
+I1202 17:49:45.384260 137274321021824 utils.py:1231] [66550] examples_seen = 68147200.0
+I1202 17:49:45.384332 137274321021824 utils.py:1231] [66550] progress = 0.5910144489933661
+I1202 17:49:45.384405 137274321021824 utils.py:1231] [66550] epoch = 53.19150430818153
+I1202 17:49:45.384482 137274321021824 utils.py:1231] [66550] img/sec/core = 164.21589104323422
+I1202 17:49:45.384554 137274321021824 utils.py:1231] [66550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 115.95866754816112
+I1202 17:49:45.384618 137274321021824 utils.py:1231] [66550] core_hours = 115.95866754816112
+I1202 17:49:45.384695 137274321021824 train.py:125] NOTE: Steps:66550/112603 [59.1%]
+Walltime:4d19h59m (0s eval)
+ETA:3d8h14m
+Total train time:8d4h12m
+I1202 17:54:57.177633 137274321021824 utils.py:1231] [66600] l2_params = 283.5131835549615
+I1202 17:54:57.177849 137274321021824 utils.py:1231] [66600] train/loss = 3.416750729084015
+I1202 17:54:57.177962 137274321021824 utils.py:1231] [66600] l2_grads = 1.6298929452896118
+I1202 17:54:57.178021 137274321021824 utils.py:1231] [66600] lr = 0.00041925331052102287
+I1202 17:54:57.178072 137274321021824 utils.py:1231] [66600] uptime = 417886.540434729
+I1202 17:54:57.178123 137274321021824 utils.py:1231] [66600] examples_seen = 68198400.0
+I1202 17:54:57.178172 137274321021824 utils.py:1231] [66600] progress = 0.5914584868964414
+I1202 17:54:57.178221 137274321021824 utils.py:1231] [66600] epoch = 53.231467872650484
+I1202 17:54:57.178271 137274321021824 utils.py:1231] [66600] img/sec/core = 164.21104188261535
+I1202 17:54:57.178333 137274321021824 utils.py:1231] [66600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.04527696284585
+I1202 17:54:57.178395 137274321021824 utils.py:1231] [66600] core_hours = 116.04527696284585
+I1202 17:54:57.178456 137274321021824 train.py:125] NOTE: Steps:66600/112603 [59.1%]
+Walltime:4d20h4m (0s eval)
+ETA:3d8h9m
+Total train time:8d4h12m
+I1202 18:00:08.977675 137274321021824 utils.py:1231] [66650] l2_params = 283.41505353079094
+I1202 18:00:08.977896 137274321021824 utils.py:1231] [66650] train/loss = 2.0345636010169983
+I1202 18:00:08.978001 137274321021824 utils.py:1231] [66650] l2_grads = 1.7690412998199463
+I1202 18:00:08.978069 137274321021824 utils.py:1231] [66650] lr = 0.0004184979802929841
+I1202 18:00:08.978131 137274321021824 utils.py:1231] [66650] uptime = 418198.340492528
+I1202 18:00:08.978195 137274321021824 utils.py:1231] [66650] examples_seen = 68249600.0
+I1202 18:00:08.978250 137274321021824 utils.py:1231] [66650] progress = 0.5919025247995169
+I1202 18:00:08.978321 137274321021824 utils.py:1231] [66650] epoch = 53.27143143711944
+I1202 18:00:08.978391 137274321021824 utils.py:1231] [66650] img/sec/core = 164.20779508966658
+I1202 18:00:08.978468 137274321021824 utils.py:1231] [66650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.13188809001221
+I1202 18:00:08.978532 137274321021824 utils.py:1231] [66650] core_hours = 116.13188809001221
+I1202 18:00:08.978609 137274321021824 train.py:125] NOTE: Steps:66650/112603 [59.2%]
+Walltime:4d20h9m (0s eval)
+ETA:3d8h4m
+Total train time:8d4h12m
+I1202 18:05:20.773798 137274321021824 utils.py:1231] [66700] l2_params = 283.33407965440256
+I1202 18:05:20.774047 137274321021824 utils.py:1231] [66700] train/loss = 2.9759441614151
+I1202 18:05:20.774205 137274321021824 utils.py:1231] [66700] l2_grads = 1.5645947456359863
+I1202 18:05:20.774302 137274321021824 utils.py:1231] [66700] lr = 0.00041774284108893984
+I1202 18:05:20.774390 137274321021824 utils.py:1231] [66700] uptime = 418510.136751303
+I1202 18:05:20.774466 137274321021824 utils.py:1231] [66700] examples_seen = 68300800.0
+I1202 18:05:20.774542 137274321021824 utils.py:1231] [66700] progress = 0.5923465627025923
+I1202 18:05:20.774609 137274321021824 utils.py:1231] [66700] epoch = 53.311395001588394
+I1202 18:05:20.774689 137274321021824 utils.py:1231] [66700] img/sec/core = 164.20979584924487
+I1202 18:05:20.774768 137274321021824 utils.py:1231] [66700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.21849816189416
+I1202 18:05:20.774836 137274321021824 utils.py:1231] [66700] core_hours = 116.21849816189416
+I1202 18:05:20.774923 137274321021824 train.py:125] NOTE: Steps:66700/112603 [59.2%]
+Walltime:4d20h15m (0s eval)
+ETA:3d7h59m
+Total train time:8d4h12m
+I1202 18:10:32.566304 137274321021824 utils.py:1231] [66750] l2_params = 283.2406552558587
+I1202 18:10:32.566559 137274321021824 utils.py:1231] [66750] train/loss = 4.489552438259125
+I1202 18:10:32.566686 137274321021824 utils.py:1231] [66750] l2_grads = 1.5939708948135376
+I1202 18:10:32.566773 137274321021824 utils.py:1231] [66750] lr = 0.000416987894678781
+I1202 18:10:32.566838 137274321021824 utils.py:1231] [66750] uptime = 418821.929199747
+I1202 18:10:32.566917 137274321021824 utils.py:1231] [66750] examples_seen = 68352000.0
+I1202 18:10:32.566976 137274321021824 utils.py:1231] [66750] progress = 0.5927906006056677
+I1202 18:10:32.567029 137274321021824 utils.py:1231] [66750] epoch = 53.35135856605735
+I1202 18:10:32.567085 137274321021824 utils.py:1231] [66750] img/sec/core = 164.21180261264638
+I1202 18:10:32.567143 137274321021824 utils.py:1231] [66750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.30510717535081
+I1202 18:10:32.567214 137274321021824 utils.py:1231] [66750] core_hours = 116.30510717535081
+I1202 18:10:32.567301 137274321021824 train.py:125] NOTE: Steps:66750/112603 [59.3%]
+Walltime:4d20h20m (0s eval)
+ETA:3d7h53m
+Total train time:8d4h12m
+I1202 18:15:44.362025 137274321021824 utils.py:1231] [66800] l2_params = 283.14851317199316
+I1202 18:15:44.362278 137274321021824 utils.py:1231] [66800] train/loss = 2.922287940979004
+I1202 18:15:44.362380 137274321021824 utils.py:1231] [66800] l2_grads = 1.6026602983474731
+I1202 18:15:44.362452 137274321021824 utils.py:1231] [66800] lr = 0.0004162331428319469
+I1202 18:15:44.362514 137274321021824 utils.py:1231] [66800] uptime = 419133.724874715
+I1202 18:15:44.362577 137274321021824 utils.py:1231] [66800] examples_seen = 68403200.0
+I1202 18:15:44.362638 137274321021824 utils.py:1231] [66800] progress = 0.5932346385087431
+I1202 18:15:44.362699 137274321021824 utils.py:1231] [66800] epoch = 53.39132213052631
+I1202 18:15:44.362758 137274321021824 utils.py:1231] [66800] img/sec/core = 164.21010331607985
+I1202 18:15:44.362821 137274321021824 utils.py:1231] [66800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.39171708506416
+I1202 18:15:44.362879 137274321021824 utils.py:1231] [66800] core_hours = 116.39171708506416
+I1202 18:15:44.362959 137274321021824 train.py:125] NOTE: Steps:66800/112603 [59.3%]
+Walltime:4d20h25m (0s eval)
+ETA:3d7h48m
+Total train time:8d4h12m
+I1202 18:20:56.151360 137274321021824 utils.py:1231] [66850] l2_params = 283.0632999772728
+I1202 18:20:56.151563 137274321021824 utils.py:1231] [66850] train/loss = 4.332266092300415
+I1202 18:20:56.151651 137274321021824 utils.py:1231] [66850] l2_grads = 1.5632243156433105
+I1202 18:20:56.151707 137274321021824 utils.py:1231] [66850] lr = 0.0004154785873174209
+I1202 18:20:56.151755 137274321021824 utils.py:1231] [66850] uptime = 419445.514117816
+I1202 18:20:56.151808 137274321021824 utils.py:1231] [66850] examples_seen = 68454400.0
+I1202 18:20:56.151854 137274321021824 utils.py:1231] [66850] progress = 0.5936786764118185
+I1202 18:20:56.151906 137274321021824 utils.py:1231] [66850] epoch = 53.43128569499527
+I1202 18:20:56.151955 137274321021824 utils.py:1231] [66850] img/sec/core = 164.21349078874468
+I1202 18:20:56.152009 137274321021824 utils.py:1231] [66850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.47832520814777
+I1202 18:20:56.152058 137274321021824 utils.py:1231] [66850] core_hours = 116.47832520814777
+I1202 18:20:56.152115 137274321021824 train.py:125] NOTE: Steps:66850/112603 [59.4%]
+Walltime:4d20h30m (0s eval)
+ETA:3d7h43m
+Total train time:8d4h12m
+I1202 18:26:07.870569 137274321021824 utils.py:1231] [66900] l2_params = 282.97267645213384
+I1202 18:26:07.870787 137274321021824 utils.py:1231] [66900] train/loss = 2.0464664548635483
+I1202 18:26:07.870893 137274321021824 utils.py:1231] [66900] l2_grads = 1.7519556283950806
+I1202 18:26:07.870978 137274321021824 utils.py:1231] [66900] lr = 0.00041472422990372604
+I1202 18:26:07.871038 137274321021824 utils.py:1231] [66900] uptime = 419757.23339912
+I1202 18:26:07.871106 137274321021824 utils.py:1231] [66900] examples_seen = 68505600.0
+I1202 18:26:07.871170 137274321021824 utils.py:1231] [66900] progress = 0.5941227143148939
+I1202 18:26:07.871229 137274321021824 utils.py:1231] [66900] epoch = 53.47124925946422
+I1202 18:26:07.871304 137274321021824 utils.py:1231] [66900] img/sec/core = 164.25034661256635
+I1202 18:26:07.871379 137274321021824 utils.py:1231] [66900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.56491389739888
+I1202 18:26:07.871450 137274321021824 utils.py:1231] [66900] core_hours = 116.56491389739888
+I1202 18:26:07.871524 137274321021824 train.py:125] NOTE: Steps:66900/112603 [59.4%]
+Walltime:4d20h35m (0s eval)
+ETA:3d7h38m
+Total train time:8d4h12m
+I1202 18:31:19.603696 137274321021824 utils.py:1231] [66950] l2_params = 282.87820493755333
+I1202 18:31:19.603977 137274321021824 utils.py:1231] [66950] train/loss = 2.162678450345993
+I1202 18:31:19.604096 137274321021824 utils.py:1231] [66950] l2_grads = 1.8209506273269653
+I1202 18:31:19.604168 137274321021824 utils.py:1231] [66950] lr = 0.00041397007235892126
+I1202 18:31:19.604228 137274321021824 utils.py:1231] [66950] uptime = 420068.966584879
+I1202 18:31:19.604287 137274321021824 utils.py:1231] [66950] examples_seen = 68556800.0
+I1202 18:31:19.604342 137274321021824 utils.py:1231] [66950] progress = 0.5945667522179693
+I1202 18:31:19.604394 137274321021824 utils.py:1231] [66950] epoch = 53.51121282393318
+I1202 18:31:19.604448 137274321021824 utils.py:1231] [66950] img/sec/core = 164.2430204385888
+I1202 18:31:19.604503 137274321021824 utils.py:1231] [66950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.6515064489986
+I1202 18:31:19.604553 137274321021824 utils.py:1231] [66950] core_hours = 116.6515064489986
+I1202 18:31:19.604629 137274321021824 train.py:125] NOTE: Steps:66950/112603 [59.5%]
+Walltime:4d20h41m (0s eval)
+ETA:3d7h32m
+Total train time:8d4h12m
+I1202 18:36:31.627784 137274321021824 utils.py:1231] [67000] l2_params = 282.7829744090597
+I1202 18:36:31.627995 137274321021824 utils.py:1231] [67000] train/loss = 3.4151527881622314
+I1202 18:36:31.628096 137274321021824 utils.py:1231] [67000] l2_grads = 1.5739569664001465
+I1202 18:36:31.628177 137274321021824 utils.py:1231] [67000] lr = 0.0004132161164505971
+I1202 18:36:31.628236 137274321021824 utils.py:1231] [67000] uptime = 420380.990597435
+I1202 18:36:31.628305 137274321021824 utils.py:1231] [67000] examples_seen = 68608000.0
+I1202 18:36:31.628362 137274321021824 utils.py:1231] [67000] progress = 0.5950107901210447
+I1202 18:36:31.628421 137274321021824 utils.py:1231] [67000] epoch = 53.55117638840213
+I1202 18:36:31.628479 137274321021824 utils.py:1231] [67000] img/sec/core = 164.0899351962842
+I1202 18:36:31.628548 137274321021824 utils.py:1231] [67000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.73817978581971
+I1202 18:36:31.628607 137274321021824 utils.py:1231] [67000] core_hours = 116.73817978581971
+I1202 18:36:31.628673 137274321021824 train.py:125] NOTE: Steps:67000/112603 [59.5%]
+Walltime:4d20h46m (0s eval)
+ETA:3d7h27m
+Total train time:8d4h12m
+I1202 18:41:43.771574 137274321021824 utils.py:1231] [67050] l2_params = 282.7006989167658
+I1202 18:41:43.771792 137274321021824 utils.py:1231] [67050] train/loss = 3.1183921098709106
+I1202 18:41:43.771905 137274321021824 utils.py:1231] [67050] l2_grads = 1.5161205530166626
+I1202 18:41:43.771982 137274321021824 utils.py:1231] [67050] lr = 0.00041246236394587105
+I1202 18:41:43.772045 137274321021824 utils.py:1231] [67050] uptime = 420693.13440558297
+I1202 18:41:43.772109 137274321021824 utils.py:1231] [67050] examples_seen = 68659200.0
+I1202 18:41:43.772171 137274321021824 utils.py:1231] [67050] progress = 0.5954548280241201
+I1202 18:41:43.772240 137274321021824 utils.py:1231] [67050] epoch = 53.591139952871096
+I1202 18:41:43.772314 137274321021824 utils.py:1231] [67050] img/sec/core = 164.0269602135636
+I1202 18:41:43.772380 137274321021824 utils.py:1231] [67050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.82488639919416
+I1202 18:41:43.772449 137274321021824 utils.py:1231] [67050] core_hours = 116.82488639919416
+I1202 18:41:43.772538 137274321021824 train.py:125] NOTE: Steps:67050/112603 [59.5%]
+Walltime:4d20h51m (0s eval)
+ETA:3d7h22m
+Total train time:8d4h12m
+I1202 18:46:55.562352 137274321021824 utils.py:1231] [67100] l2_params = 282.609997409661
+I1202 18:46:55.562554 137274321021824 utils.py:1231] [67100] train/loss = 2.1922096610069275
+I1202 18:46:55.562650 137274321021824 utils.py:1231] [67100] l2_grads = 1.7504452466964722
+I1202 18:46:55.562712 137274321021824 utils.py:1231] [67100] lr = 0.00041170881661138436
+I1202 18:46:55.562765 137274321021824 utils.py:1231] [67100] uptime = 421004.92512696097
+I1202 18:46:55.562819 137274321021824 utils.py:1231] [67100] examples_seen = 68710400.0
+I1202 18:46:55.562872 137274321021824 utils.py:1231] [67100] progress = 0.5958988659271955
+I1202 18:46:55.562928 137274321021824 utils.py:1231] [67100] epoch = 53.63110351734005
+I1202 18:46:55.562980 137274321021824 utils.py:1231] [67100] img/sec/core = 164.21271221194522
+I1202 18:46:55.563038 137274321021824 utils.py:1231] [67100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.91149493291027
+I1202 18:46:55.563092 137274321021824 utils.py:1231] [67100] core_hours = 116.91149493291027
+I1202 18:46:55.563158 137274321021824 train.py:125] NOTE: Steps:67100/112603 [59.6%]
+Walltime:4d20h56m (0s eval)
+ETA:3d7h17m
+Total train time:8d4h11m
+I1202 18:52:07.352008 137274321021824 utils.py:1231] [67150] l2_params = 282.5228119611085
+I1202 18:52:07.352279 137274321021824 utils.py:1231] [67150] train/loss = 2.162777602672577
+I1202 18:52:07.352408 137274321021824 utils.py:1231] [67150] l2_grads = 1.8164591789245605
+I1202 18:52:07.352482 137274321021824 utils.py:1231] [67150] lr = 0.0004109554762132973
+I1202 18:52:07.352540 137274321021824 utils.py:1231] [67150] uptime = 421316.714901259
+I1202 18:52:07.352602 137274321021824 utils.py:1231] [67150] examples_seen = 68761600.0
+I1202 18:52:07.352651 137274321021824 utils.py:1231] [67150] progress = 0.596342903830271
+I1202 18:52:07.352700 137274321021824 utils.py:1231] [67150] epoch = 53.67106708180901
+I1202 18:52:07.352751 137274321021824 utils.py:1231] [67150] img/sec/core = 164.21321101781845
+I1202 18:52:07.352808 137274321021824 utils.py:1231] [67150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 116.99810320354861
+I1202 18:52:07.352865 137274321021824 utils.py:1231] [67150] core_hours = 116.99810320354861
+I1202 18:52:07.352933 137274321021824 train.py:125] NOTE: Steps:67150/112603 [59.6%]
+Walltime:4d21h1m (0s eval)
+ETA:3d7h11m
+Total train time:8d4h11m
+I1202 18:57:19.142336 137274321021824 utils.py:1231] [67200] l2_params = 282.4442236085891
+I1202 18:57:19.142621 137274321021824 utils.py:1231] [67200] train/loss = 2.259476214647293
+I1202 18:57:19.142794 137274321021824 utils.py:1231] [67200] l2_grads = 1.8690334558486938
+I1202 18:57:19.142867 137274321021824 utils.py:1231] [67200] lr = 0.0004102023445172849
+I1202 18:57:19.142935 137274321021824 utils.py:1231] [67200] uptime = 421628.505296229
+I1202 18:57:19.142988 137274321021824 utils.py:1231] [67200] examples_seen = 68812800.0
+I1202 18:57:19.143037 137274321021824 utils.py:1231] [67200] progress = 0.5967869417333463
+I1202 18:57:19.143088 137274321021824 utils.py:1231] [67200] epoch = 53.71103064627796
+I1202 18:57:19.143140 137274321021824 utils.py:1231] [67200] img/sec/core = 164.21288412341622
+I1202 18:57:19.143196 137274321021824 utils.py:1231] [67200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.08471164659584
+I1202 18:57:19.143247 137274321021824 utils.py:1231] [67200] core_hours = 117.08471164659584
+I1202 18:57:19.143307 137274321021824 train.py:125] NOTE: Steps:67200/112603 [59.7%]
+Walltime:4d21h7m (0s eval)
+ETA:3d7h6m
+Total train time:8d4h11m
+I1202 19:02:30.940964 137274321021824 utils.py:1231] [67250] l2_params = 282.34870512973043
+I1202 19:02:30.941170 137274321021824 utils.py:1231] [67250] train/loss = 2.9819257855415344
+I1202 19:02:30.941271 137274321021824 utils.py:1231] [67250] l2_grads = 1.6174589395523071
+I1202 19:02:30.941340 137274321021824 utils.py:1231] [67250] lr = 0.0004094494232885333
+I1202 19:02:30.941402 137274321021824 utils.py:1231] [67250] uptime = 421940.30376255
+I1202 19:02:30.941463 137274321021824 utils.py:1231] [67250] examples_seen = 68864000.0
+I1202 19:02:30.941523 137274321021824 utils.py:1231] [67250] progress = 0.5972309796364218
+I1202 19:02:30.941578 137274321021824 utils.py:1231] [67250] epoch = 53.75099421074692
+I1202 19:02:30.941637 137274321021824 utils.py:1231] [67250] img/sec/core = 164.2086332371029
+I1202 19:02:30.941700 137274321021824 utils.py:1231] [67250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.17132233168499
+I1202 19:02:30.941754 137274321021824 utils.py:1231] [67250] core_hours = 117.17132233168499
+I1202 19:02:30.941820 137274321021824 train.py:125] NOTE: Steps:67250/112603 [59.7%]
+Walltime:4d21h12m (0s eval)
+ETA:3d7h1m
+Total train time:8d4h11m
+I1202 19:07:42.725383 137274321021824 utils.py:1231] [67300] l2_params = 282.2637546977804
+I1202 19:07:42.725579 137274321021824 utils.py:1231] [67300] train/loss = 2.137403592467308
+I1202 19:07:42.725689 137274321021824 utils.py:1231] [67300] l2_grads = 1.8113594055175781
+I1202 19:07:42.725764 137274321021824 utils.py:1231] [67300] lr = 0.00040869671429173487
+I1202 19:07:42.725826 137274321021824 utils.py:1231] [67300] uptime = 422252.088187504
+I1202 19:07:42.725897 137274321021824 utils.py:1231] [67300] examples_seen = 68915200.0
+I1202 19:07:42.725957 137274321021824 utils.py:1231] [67300] progress = 0.5976750175394971
+I1202 19:07:42.726014 137274321021824 utils.py:1231] [67300] epoch = 53.79095777521588
+I1202 19:07:42.726073 137274321021824 utils.py:1231] [67300] img/sec/core = 164.21602845478128
+I1202 19:07:42.726137 137274321021824 utils.py:1231] [67300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.25792911639444
+I1202 19:07:42.726199 137274321021824 utils.py:1231] [67300] core_hours = 117.25792911639444
+I1202 19:07:42.726300 137274321021824 train.py:125] NOTE: Steps:67300/112603 [59.8%]
+Walltime:4d21h17m (0s eval)
+ETA:3d6h56m
+Total train time:8d4h11m
+I1202 19:12:54.487369 137274321021824 utils.py:1231] [67350] l2_params = 282.1750887608509
+I1202 19:12:54.487627 137274321021824 utils.py:1231] [67350] train/loss = 3.6008733808994293
+I1202 19:12:54.487781 137274321021824 utils.py:1231] [67350] l2_grads = 1.645617961883545
+I1202 19:12:54.487875 137274321021824 utils.py:1231] [67350] lr = 0.0004079442192910856
+I1202 19:12:54.487956 137274321021824 utils.py:1231] [67350] uptime = 422563.85031671
+I1202 19:12:54.488019 137274321021824 utils.py:1231] [67350] examples_seen = 68966400.0
+I1202 19:12:54.488079 137274321021824 utils.py:1231] [67350] progress = 0.5981190554425726
+I1202 19:12:54.488140 137274321021824 utils.py:1231] [67350] epoch = 53.830921339684835
+I1202 19:12:54.488201 137274321021824 utils.py:1231] [67350] img/sec/core = 164.22777240581138
+I1202 19:12:54.709678 137274321021824 utils.py:1231] [67350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.34452970784056
+I1202 19:12:54.710107 137274321021824 utils.py:1231] [67350] core_hours = 117.34452970784056
+I1202 19:12:54.710190 137274321021824 train.py:125] NOTE: Steps:67350/112603 [59.8%]
+Walltime:4d21h22m (0s eval)
+ETA:3d6h50m
+Total train time:8d4h11m
+I1202 19:18:06.506547 137274321021824 utils.py:1231] [67400] l2_params = 282.0786491922296
+I1202 19:18:06.506752 137274321021824 utils.py:1231] [67400] train/loss = 2.1766377091407776
+I1202 19:18:06.506846 137274321021824 utils.py:1231] [67400] l2_grads = 1.808812141418457
+I1202 19:18:06.506911 137274321021824 utils.py:1231] [67400] lr = 0.00040719194005027856
+I1202 19:18:06.506964 137274321021824 utils.py:1231] [67400] uptime = 422875.869326231
+I1202 19:18:06.507016 137274321021824 utils.py:1231] [67400] examples_seen = 69017600.0
+I1202 19:18:06.507066 137274321021824 utils.py:1231] [67400] progress = 0.598563093345648
+I1202 19:18:06.507114 137274321021824 utils.py:1231] [67400] epoch = 53.87088490415379
+I1202 19:18:06.507166 137274321021824 utils.py:1231] [67400] img/sec/core = 164.09256627857587
+I1202 19:18:06.507221 137274321021824 utils.py:1231] [67400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.43120165492972
+I1202 19:18:06.507272 137274321021824 utils.py:1231] [67400] core_hours = 117.43120165492972
+I1202 19:18:06.507332 137274321021824 train.py:125] NOTE: Steps:67400/112603 [59.9%]
+Walltime:4d21h27m (0s eval)
+ETA:3d6h45m
+Total train time:8d4h11m
+I1202 19:23:18.305989 137274321021824 utils.py:1231] [67450] l2_params = 281.9820567541095
+I1202 19:23:18.306221 137274321021824 utils.py:1231] [67450] train/loss = 2.013116553425789
+I1202 19:23:18.306363 137274321021824 utils.py:1231] [67450] l2_grads = 1.9599350690841675
+I1202 19:23:18.306442 137274321021824 utils.py:1231] [67450] lr = 0.00040643987833250176
+I1202 19:23:18.306496 137274321021824 utils.py:1231] [67450] uptime = 423187.668857637
+I1202 19:23:18.306550 137274321021824 utils.py:1231] [67450] examples_seen = 69068800.0
+I1202 19:23:18.306600 137274321021824 utils.py:1231] [67450] progress = 0.5990071312487234
+I1202 19:23:18.306650 137274321021824 utils.py:1231] [67450] epoch = 53.910848468622746
+I1202 19:23:18.306702 137274321021824 utils.py:1231] [67450] img/sec/core = 164.2080723121327
+I1202 19:23:18.306762 137274321021824 utils.py:1231] [67450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.51781263587583
+I1202 19:23:18.306815 137274321021824 utils.py:1231] [67450] core_hours = 117.51781263587583
+I1202 19:23:18.306877 137274321021824 train.py:125] NOTE: Steps:67450/112603 [59.9%]
+Walltime:4d21h33m (0s eval)
+ETA:3d6h40m
+Total train time:8d4h11m
+I1202 19:28:30.103055 137274321021824 utils.py:1231] [67500] l2_params = 281.8889605692551
+I1202 19:28:30.103336 137274321021824 utils.py:1231] [67500] train/loss = 2.0880991518497467
+I1202 19:28:30.103436 137274321021824 utils.py:1231] [67500] l2_grads = 1.8603954315185547
+I1202 19:28:30.103498 137274321021824 utils.py:1231] [67500] lr = 0.00040568803590043374
+I1202 19:28:30.103553 137274321021824 utils.py:1231] [67500] uptime = 423499.46591485
+I1202 19:28:30.103608 137274321021824 utils.py:1231] [67500] examples_seen = 69120000.0
+I1202 19:28:30.103658 137274321021824 utils.py:1231] [67500] progress = 0.5994511691517987
+I1202 19:28:30.103707 137274321021824 utils.py:1231] [67500] epoch = 53.95081203309171
+I1202 19:28:30.103761 137274321021824 utils.py:1231] [67500] img/sec/core = 164.20937534705038
+I1202 19:28:30.103819 137274321021824 utils.py:1231] [67500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.60442292954612
+I1202 19:28:30.103870 137274321021824 utils.py:1231] [67500] core_hours = 117.60442292954612
+I1202 19:28:30.103938 137274321021824 train.py:125] NOTE: Steps:67500/112603 [59.9%]
+Walltime:4d21h38m (0s eval)
+ETA:3d6h35m
+Total train time:8d4h11m
+I1202 19:28:30.104046 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:67500/112603 [59.9%]
+Walltime:4d21h38m (0s eval)
+ETA:3d6h35m
+Total train time:8d4h11m
+I1202 19:30:08.336146 137274321021824 utils.py:1231] [67500] val/acc@1 = 0.6925023915816326
+I1202 19:30:08.336395 137274321021824 utils.py:1231] [67500] val/loss = 1.238312049179661
+I1202 19:30:08.336593 137274321021824 utils.py:1231] [67500] z/secs/eval/val = 98.23246661102166
+I1202 19:30:08.336685 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.23246661102166
+I1202 19:35:20.116031 137274321021824 utils.py:1231] [67550] l2_params = 281.8064018307987
+I1202 19:35:20.116256 137274321021824 utils.py:1231] [67550] train/loss = 2.1439166218042374
+I1202 19:35:20.116349 137274321021824 utils.py:1231] [67550] l2_grads = 1.7969096899032593
+I1202 19:35:20.116406 137274321021824 utils.py:1231] [67550] lr = 0.00040493641451623877
+I1202 19:35:20.116460 137274321021824 utils.py:1231] [67550] uptime = 423909.47882227297
+I1202 19:35:20.116509 137274321021824 utils.py:1231] [67550] examples_seen = 69171200.0
+I1202 19:35:20.116562 137274321021824 utils.py:1231] [67550] progress = 0.5998952070548742
+I1202 19:35:20.116609 137274321021824 utils.py:1231] [67550] epoch = 53.99077559756066
+I1202 19:35:20.116657 137274321021824 utils.py:1231] [67550] img/sec/core = 124.87411755353087
+I1202 19:35:20.116709 137274321021824 utils.py:1231] [67550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.71831540383026
+I1202 19:35:20.116758 137274321021824 utils.py:1231] [67550] core_hours = 117.71831540383026
+I1202 19:35:20.116815 137274321021824 train.py:125] NOTE: Steps:67550/112603 [60.0%]
+Walltime:4d21h45m (0s eval)
+ETA:3d6h30m
+Total train time:8d4h14m
+I1202 19:40:31.907580 137274321021824 utils.py:1231] [67600] l2_params = 281.7135457509742
+I1202 19:40:31.907840 137274321021824 utils.py:1231] [67600] train/loss = 2.1064426004886627
+I1202 19:40:31.907960 137274321021824 utils.py:1231] [67600] l2_grads = 1.7578712701797485
+I1202 19:40:31.908037 137274321021824 utils.py:1231] [67600] lr = 0.00040418501594156246
+I1202 19:40:31.908100 137274321021824 utils.py:1231] [67600] uptime = 424221.270461627
+I1202 19:40:31.908169 137274321021824 utils.py:1231] [67600] examples_seen = 69222400.0
+I1202 19:40:31.908234 137274321021824 utils.py:1231] [67600] progress = 0.6003392449579497
+I1202 19:40:31.908305 137274321021824 utils.py:1231] [67600] epoch = 54.03073916202962
+I1202 19:40:31.908372 137274321021824 utils.py:1231] [67600] img/sec/core = 164.21222873736295
+I1202 19:40:31.908442 137274321021824 utils.py:1231] [67600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.80492419253973
+I1202 19:40:31.908500 137274321021824 utils.py:1231] [67600] core_hours = 117.80492419253973
+I1202 19:40:31.908568 137274321021824 train.py:125] NOTE: Steps:67600/112603 [60.0%]
+Walltime:4d21h50m (0s eval)
+ETA:3d6h25m
+Total train time:8d4h14m
+I1202 19:45:43.652746 137274321021824 utils.py:1231] [67650] l2_params = 281.6207096592756
+I1202 19:45:43.652957 137274321021824 utils.py:1231] [67650] train/loss = 2.0736412405967712
+I1202 19:45:43.653067 137274321021824 utils.py:1231] [67650] l2_grads = 1.8564143180847168
+I1202 19:45:43.653137 137274321021824 utils.py:1231] [67650] lr = 0.0004034338419375295
+I1202 19:45:43.653200 137274321021824 utils.py:1231] [67650] uptime = 424533.015561205
+I1202 19:45:43.653265 137274321021824 utils.py:1231] [67650] examples_seen = 69273600.0
+I1202 19:45:43.653324 137274321021824 utils.py:1231] [67650] progress = 0.600783282861025
+I1202 19:45:43.653383 137274321021824 utils.py:1231] [67650] epoch = 54.070702726498574
+I1202 19:45:43.653443 137274321021824 utils.py:1231] [67650] img/sec/core = 164.2367436386612
+I1202 19:45:43.653507 137274321021824 utils.py:1231] [67650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.89152005353361
+I1202 19:45:43.653566 137274321021824 utils.py:1231] [67650] core_hours = 117.89152005353361
+I1202 19:45:43.653634 137274321021824 train.py:125] NOTE: Steps:67650/112603 [60.1%]
+Walltime:4d21h55m (0s eval)
+ETA:3d6h20m
+Total train time:8d4h14m
+I1202 19:50:55.452611 137274321021824 utils.py:1231] [67700] l2_params = 281.53681275220725
+I1202 19:50:55.452809 137274321021824 utils.py:1231] [67700] train/loss = 2.143089607357979
+I1202 19:50:55.452910 137274321021824 utils.py:1231] [67700] l2_grads = 1.8732054233551025
+I1202 19:50:55.452974 137274321021824 utils.py:1231] [67700] lr = 0.0004026828942647367
+I1202 19:50:55.453028 137274321021824 utils.py:1231] [67700] uptime = 424844.815389235
+I1202 19:50:55.453080 137274321021824 utils.py:1231] [67700] examples_seen = 69324800.0
+I1202 19:50:55.453131 137274321021824 utils.py:1231] [67700] progress = 0.6012273207641005
+I1202 19:50:55.453180 137274321021824 utils.py:1231] [67700] epoch = 54.11066629096753
+I1202 19:50:55.453230 137274321021824 utils.py:1231] [67700] img/sec/core = 164.2079160963318
+I1202 19:50:55.453288 137274321021824 utils.py:1231] [67700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 117.97813111687528
+I1202 19:50:55.453341 137274321021824 utils.py:1231] [67700] core_hours = 117.97813111687528
+I1202 19:50:55.453409 137274321021824 train.py:125] NOTE: Steps:67700/112603 [60.1%]
+Walltime:4d22h0m (0s eval)
+ETA:3d6h15m
+Total train time:8d4h14m
+I1202 19:56:07.245297 137274321021824 utils.py:1231] [67750] l2_params = 281.4447650397414
+I1202 19:56:07.245518 137274321021824 utils.py:1231] [67750] train/loss = 2.0283312052488327
+I1202 19:56:07.245614 137274321021824 utils.py:1231] [67750] l2_grads = 1.762516736984253
+I1202 19:56:07.245676 137274321021824 utils.py:1231] [67750] lr = 0.00040193217468325194
+I1202 19:56:07.245729 137274321021824 utils.py:1231] [67750] uptime = 425156.608091172
+I1202 19:56:07.245783 137274321021824 utils.py:1231] [67750] examples_seen = 69376000.0
+I1202 19:56:07.245832 137274321021824 utils.py:1231] [67750] progress = 0.6016713586671758
+I1202 19:56:07.245885 137274321021824 utils.py:1231] [67750] epoch = 54.15062985543649
+I1202 19:56:07.245937 137274321021824 utils.py:1231] [67750] img/sec/core = 164.21166910552373
+I1202 19:56:07.245994 137274321021824 utils.py:1231] [67750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.06474020074667
+I1202 19:56:07.246045 137274321021824 utils.py:1231] [67750] core_hours = 118.06474020074667
+I1202 19:56:07.246105 137274321021824 train.py:125] NOTE: Steps:67750/112603 [60.2%]
+Walltime:4d22h5m (0s eval)
+ETA:3d6h9m
+Total train time:8d4h14m
+I1202 20:01:18.926362 137274321021824 utils.py:1231] [67800] l2_params = 281.36461082029643
+I1202 20:01:18.926618 137274321021824 utils.py:1231] [67800] train/loss = 2.198395848274231
+I1202 20:01:18.926753 137274321021824 utils.py:1231] [67800] l2_grads = 1.8074359893798828
+I1202 20:01:18.926826 137274321021824 utils.py:1231] [67800] lr = 0.000401181684952607
+I1202 20:01:18.926898 137274321021824 utils.py:1231] [67800] uptime = 425468.289247845
+I1202 20:01:18.926951 137274321021824 utils.py:1231] [67800] examples_seen = 69427200.0
+I1202 20:01:18.927002 137274321021824 utils.py:1231] [67800] progress = 0.6021153965702513
+I1202 20:01:18.927051 137274321021824 utils.py:1231] [67800] epoch = 54.19059341990545
+I1202 20:01:18.927102 137274321021824 utils.py:1231] [67800] img/sec/core = 164.27043760529912
+I1202 20:01:18.927158 137274321021824 utils.py:1231] [67800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.1513182998225
+I1202 20:01:18.927210 137274321021824 utils.py:1231] [67800] core_hours = 118.1513182998225
+I1202 20:01:18.927270 137274321021824 train.py:125] NOTE: Steps:67800/112603 [60.2%]
+Walltime:4d22h11m (0s eval)
+ETA:3d6h4m
+Total train time:8d4h13m
+I1202 20:06:30.721221 137274321021824 utils.py:1231] [67850] l2_params = 281.2602846315609
+I1202 20:06:30.721486 137274321021824 utils.py:1231] [67850] train/loss = 2.1853821873664856
+I1202 20:06:30.721618 137274321021824 utils.py:1231] [67850] l2_grads = 1.8126004934310913
+I1202 20:06:30.721694 137274321021824 utils.py:1231] [67850] lr = 0.000400431426831796
+I1202 20:06:30.721765 137274321021824 utils.py:1231] [67850] uptime = 425780.084126883
+I1202 20:06:30.721841 137274321021824 utils.py:1231] [67850] examples_seen = 69478400.0
+I1202 20:06:30.721907 137274321021824 utils.py:1231] [67850] progress = 0.6025594344733266
+I1202 20:06:30.721963 137274321021824 utils.py:1231] [67850] epoch = 54.2305569843744
+I1202 20:06:30.722020 137274321021824 utils.py:1231] [67850] img/sec/core = 164.21052250110708
+I1202 20:06:30.722082 137274321021824 utils.py:1231] [67850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.23792798844417
+I1202 20:06:30.722135 137274321021824 utils.py:1231] [67850] core_hours = 118.23792798844417
+I1202 20:06:30.722200 137274321021824 train.py:125] NOTE: Steps:67850/112603 [60.3%]
+Walltime:4d22h16m (0s eval)
+ETA:3d5h59m
+Total train time:8d4h13m
+I1202 20:11:42.480893 137274321021824 utils.py:1231] [67900] l2_params = 281.16976986091976
+I1202 20:11:42.481114 137274321021824 utils.py:1231] [67900] train/loss = 2.1403636038303375
+I1202 20:11:42.481211 137274321021824 utils.py:1231] [67900] l2_grads = 1.7883143424987793
+I1202 20:11:42.481289 137274321021824 utils.py:1231] [67900] lr = 0.00039968140207927014
+I1202 20:11:42.481352 137274321021824 utils.py:1231] [67900] uptime = 426091.84371359396
+I1202 20:11:42.481417 137274321021824 utils.py:1231] [67900] examples_seen = 69529600.0
+I1202 20:11:42.481471 137274321021824 utils.py:1231] [67900] progress = 0.6030034723764021
+I1202 20:11:42.481527 137274321021824 utils.py:1231] [67900] epoch = 54.27052054884336
+I1202 20:11:42.481582 137274321021824 utils.py:1231] [67900] img/sec/core = 164.22911173368558
+I1202 20:11:42.481646 137274321021824 utils.py:1231] [67900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.32452787364166
+I1202 20:11:42.481711 137274321021824 utils.py:1231] [67900] core_hours = 118.32452787364166
+I1202 20:11:42.481791 137274321021824 train.py:125] NOTE: Steps:67900/112603 [60.3%]
+Walltime:4d22h21m (0s eval)
+ETA:3d5h54m
+Total train time:8d4h13m
+I1202 20:16:54.251147 137274321021824 utils.py:1231] [67950] l2_params = 281.08347217330385
+I1202 20:16:54.251350 137274321021824 utils.py:1231] [67950] train/loss = 2.1297967582941055
+I1202 20:16:54.251455 137274321021824 utils.py:1231] [67950] l2_grads = 1.7087098360061646
+I1202 20:16:54.251527 137274321021824 utils.py:1231] [67950] lr = 0.0003989316124529332
+I1202 20:16:54.251591 137274321021824 utils.py:1231] [67950] uptime = 426403.61394937796
+I1202 20:16:54.251677 137274321021824 utils.py:1231] [67950] examples_seen = 69580800.0
+I1202 20:16:54.251749 137274321021824 utils.py:1231] [67950] progress = 0.6034475102794774
+I1202 20:16:54.251850 137274321021824 utils.py:1231] [67950] epoch = 54.31048411331231
+I1202 20:16:54.251953 137274321021824 utils.py:1231] [67950] img/sec/core = 164.22350219304403
+I1202 20:16:54.252049 137274321021824 utils.py:1231] [67950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.411130716915
+I1202 20:16:54.252136 137274321021824 utils.py:1231] [67950] core_hours = 118.411130716915
+I1202 20:16:54.252235 137274321021824 train.py:125] NOTE: Steps:67950/112603 [60.3%]
+Walltime:4d22h26m (0s eval)
+ETA:3d5h48m
+Total train time:8d4h13m
+I1202 20:22:06.017127 137274321021824 utils.py:1231] [68000] l2_params = 280.99775656819776
+I1202 20:22:06.017336 137274321021824 utils.py:1231] [68000] train/loss = 4.205857515335083
+I1202 20:22:06.017429 137274321021824 utils.py:1231] [68000] l2_grads = 1.632387638092041
+I1202 20:22:06.017495 137274321021824 utils.py:1231] [68000] lr = 0.0003981820597101377
+I1202 20:22:06.017546 137274321021824 utils.py:1231] [68000] uptime = 426715.37990836
+I1202 20:22:06.017599 137274321021824 utils.py:1231] [68000] examples_seen = 69632000.0
+I1202 20:22:06.017648 137274321021824 utils.py:1231] [68000] progress = 0.6038915481825529
+I1202 20:22:06.017697 137274321021824 utils.py:1231] [68000] epoch = 54.350447677781276
+I1202 20:22:06.017750 137274321021824 utils.py:1231] [68000] img/sec/core = 164.22575500921565
+I1202 20:22:06.017805 137274321021824 utils.py:1231] [68000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.49773237218778
+I1202 20:22:06.017854 137274321021824 utils.py:1231] [68000] core_hours = 118.49773237218778
+I1202 20:22:06.017919 137274321021824 train.py:125] NOTE: Steps:68000/112603 [60.4%]
+Walltime:4d22h31m (0s eval)
+ETA:3d5h43m
+Total train time:8d4h13m
+I1202 20:27:18.162539 137274321021824 utils.py:1231] [68050] l2_params = 280.90597576921976
+I1202 20:27:18.162793 137274321021824 utils.py:1231] [68050] train/loss = 2.14795283973217
+I1202 20:27:18.162907 137274321021824 utils.py:1231] [68050] l2_grads = 1.8054841756820679
+I1202 20:27:18.162970 137274321021824 utils.py:1231] [68050] lr = 0.0003974327456076817
+I1202 20:27:18.163023 137274321021824 utils.py:1231] [68050] uptime = 427027.525384693
+I1202 20:27:18.163077 137274321021824 utils.py:1231] [68050] examples_seen = 69683200.0
+I1202 20:27:18.163127 137274321021824 utils.py:1231] [68050] progress = 0.6043355860856283
+I1202 20:27:18.163177 137274321021824 utils.py:1231] [68050] epoch = 54.39041124225023
+I1202 20:27:18.163230 137274321021824 utils.py:1231] [68050] img/sec/core = 164.02608361167552
+I1202 20:27:18.163287 137274321021824 utils.py:1231] [68050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.58443944894695
+I1202 20:27:18.163339 137274321021824 utils.py:1231] [68050] core_hours = 118.58443944894695
+I1202 20:27:18.163411 137274321021824 train.py:125] NOTE: Steps:68050/112603 [60.4%]
+Walltime:4d22h37m (0s eval)
+ETA:3d5h38m
+Total train time:8d4h13m
+I1202 20:32:29.937142 137274321021824 utils.py:1231] [68100] l2_params = 280.8221511567315
+I1202 20:32:29.937435 137274321021824 utils.py:1231] [68100] train/loss = 4.322499752044678
+I1202 20:32:29.937595 137274321021824 utils.py:1231] [68100] l2_grads = 1.7750651836395264
+I1202 20:32:29.937677 137274321021824 utils.py:1231] [68100] lr = 0.0003966836719018034
+I1202 20:32:29.937733 137274321021824 utils.py:1231] [68100] uptime = 427339.30009166896
+I1202 20:32:29.937788 137274321021824 utils.py:1231] [68100] examples_seen = 69734400.0
+I1202 20:32:29.937843 137274321021824 utils.py:1231] [68100] progress = 0.6047796239887037
+I1202 20:32:29.937900 137274321021824 utils.py:1231] [68100] epoch = 54.430374806719186
+I1202 20:32:29.937954 137274321021824 utils.py:1231] [68100] img/sec/core = 164.22114704752866
+I1202 20:32:29.938031 137274321021824 utils.py:1231] [68100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.67104353421804
+I1202 20:32:29.938091 137274321021824 utils.py:1231] [68100] core_hours = 118.67104353421804
+I1202 20:32:29.938161 137274321021824 train.py:125] NOTE: Steps:68100/112603 [60.5%]
+Walltime:4d22h42m (0s eval)
+ETA:3d5h33m
+Total train time:8d4h13m
+I1202 20:37:41.736907 137274321021824 utils.py:1231] [68150] l2_params = 280.7313810649486
+I1202 20:37:41.737118 137274321021824 utils.py:1231] [68150] train/loss = 3.423952639102936
+I1202 20:37:41.737225 137274321021824 utils.py:1231] [68150] l2_grads = 1.5249394178390503
+I1202 20:37:41.737300 137274321021824 utils.py:1231] [68150] lr = 0.0003959348403481776
+I1202 20:37:41.737367 137274321021824 utils.py:1231] [68150] uptime = 427651.099724815
+I1202 20:37:41.737431 137274321021824 utils.py:1231] [68150] examples_seen = 69785600.0
+I1202 20:37:41.737492 137274321021824 utils.py:1231] [68150] progress = 0.6052236618917791
+I1202 20:37:41.737552 137274321021824 utils.py:1231] [68150] epoch = 54.47033837118814
+I1202 20:37:41.737614 137274321021824 utils.py:1231] [68150] img/sec/core = 164.20801873112072
+I1202 20:37:41.737680 137274321021824 utils.py:1231] [68150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.75765454342528
+I1202 20:37:41.737739 137274321021824 utils.py:1231] [68150] core_hours = 118.75765454342528
+I1202 20:37:41.737808 137274321021824 train.py:125] NOTE: Steps:68150/112603 [60.5%]
+Walltime:4d22h47m (0s eval)
+ETA:3d5h27m
+Total train time:8d4h13m
+I1202 20:42:53.522102 137274321021824 utils.py:1231] [68200] l2_params = 280.6517006068184
+I1202 20:42:53.522338 137274321021824 utils.py:1231] [68200] train/loss = 4.270714521408081
+I1202 20:42:53.522511 137274321021824 utils.py:1231] [68200] l2_grads = 1.7611764669418335
+I1202 20:42:53.522676 137274321021824 utils.py:1231] [68200] lr = 0.000395186252701912
+I1202 20:42:53.522791 137274321021824 utils.py:1231] [68200] uptime = 427962.885144863
+I1202 20:42:53.522861 137274321021824 utils.py:1231] [68200] examples_seen = 69836800.0
+I1202 20:42:53.522943 137274321021824 utils.py:1231] [68200] progress = 0.6056676997948545
+I1202 20:42:53.523024 137274321021824 utils.py:1231] [68200] epoch = 54.5103019356571
+I1202 20:42:53.523082 137274321021824 utils.py:1231] [68200] img/sec/core = 164.21550434307963
+I1202 20:42:53.523148 137274321021824 utils.py:1231] [68200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.84426160454971
+I1202 20:42:53.523230 137274321021824 utils.py:1231] [68200] core_hours = 118.84426160454971
+I1202 20:42:53.523308 137274321021824 train.py:125] NOTE: Steps:68200/112603 [60.6%]
+Walltime:4d22h52m (0s eval)
+ETA:3d5h22m
+Total train time:8d4h13m
+I1202 20:48:05.205037 137274321021824 utils.py:1231] [68250] l2_params = 280.5529097654534
+I1202 20:48:05.205267 137274321021824 utils.py:1231] [68250] train/loss = 2.9232660830020905
+I1202 20:48:05.205378 137274321021824 utils.py:1231] [68250] l2_grads = 1.6211352348327637
+I1202 20:48:05.205452 137274321021824 utils.py:1231] [68250] lr = 0.00039443791071754226
+I1202 20:48:05.205516 137274321021824 utils.py:1231] [68250] uptime = 428274.567875818
+I1202 20:48:05.205596 137274321021824 utils.py:1231] [68250] examples_seen = 69888000.0
+I1202 20:48:05.205671 137274321021824 utils.py:1231] [68250] progress = 0.6061117376979299
+I1202 20:48:05.205774 137274321021824 utils.py:1231] [68250] epoch = 54.55026550012606
+I1202 20:48:05.205860 137274321021824 utils.py:1231] [68250] img/sec/core = 164.26960788979326
+I1202 20:48:05.205937 137274321021824 utils.py:1231] [68250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 118.93084014092611
+I1202 20:48:05.206000 137274321021824 utils.py:1231] [68250] core_hours = 118.93084014092611
+I1202 20:48:05.206068 137274321021824 train.py:125] NOTE: Steps:68250/112603 [60.6%]
+Walltime:4d22h57m (0s eval)
+ETA:3d5h17m
+Total train time:8d4h13m
+I1202 20:53:17.000810 137274321021824 utils.py:1231] [68300] l2_params = 280.4792619828554
+I1202 20:53:17.001061 137274321021824 utils.py:1231] [68300] train/loss = 4.129646509885788
+I1202 20:53:17.001158 137274321021824 utils.py:1231] [68300] l2_grads = 1.6169133186340332
+I1202 20:53:17.001218 137274321021824 utils.py:1231] [68300] lr = 0.0003936898161490278
+I1202 20:53:17.001270 137274321021824 utils.py:1231] [68300] uptime = 428586.363632178
+I1202 20:53:17.001325 137274321021824 utils.py:1231] [68300] examples_seen = 69939200.0
+I1202 20:53:17.001376 137274321021824 utils.py:1231] [68300] progress = 0.6065557756010053
+I1202 20:53:17.001426 137274321021824 utils.py:1231] [68300] epoch = 54.590229064595015
+I1202 20:53:17.001479 137274321021824 utils.py:1231] [68300] img/sec/core = 164.21006045023165
+I1202 20:53:17.001537 137274321021824 utils.py:1231] [68300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.01745007324833
+I1202 20:53:17.001588 137274321021824 utils.py:1231] [68300] core_hours = 119.01745007324833
+I1202 20:53:17.001650 137274321021824 train.py:125] NOTE: Steps:68300/112603 [60.7%]
+Walltime:4d23h3m (0s eval)
+ETA:3d5h12m
+Total train time:8d4h13m
+I1202 20:58:28.788834 137274321021824 utils.py:1231] [68350] l2_params = 280.37397995704566
+I1202 20:58:28.789058 137274321021824 utils.py:1231] [68350] train/loss = 2.017934560775757
+I1202 20:58:28.789156 137274321021824 utils.py:1231] [68350] l2_grads = 1.7349004745483398
+I1202 20:58:28.789217 137274321021824 utils.py:1231] [68350] lr = 0.00039294197074974904
+I1202 20:58:28.789275 137274321021824 utils.py:1231] [68350] uptime = 428898.15163751796
+I1202 20:58:28.789328 137274321021824 utils.py:1231] [68350] examples_seen = 69990400.0
+I1202 20:58:28.789394 137274321021824 utils.py:1231] [68350] progress = 0.6069998135040807
+I1202 20:58:28.789446 137274321021824 utils.py:1231] [68350] epoch = 54.63019262906397
+I1202 20:58:28.789505 137274321021824 utils.py:1231] [68350] img/sec/core = 164.21414269664385
+I1202 20:58:28.789560 137274321021824 utils.py:1231] [68350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.10405785250943
+I1202 20:58:28.789609 137274321021824 utils.py:1231] [68350] core_hours = 119.10405785250943
+I1202 20:58:28.789668 137274321021824 train.py:125] NOTE: Steps:68350/112603 [60.7%]
+Walltime:4d23h8m (0s eval)
+ETA:3d5h6m
+Total train time:8d4h13m
+I1202 21:03:40.576536 137274321021824 utils.py:1231] [68400] l2_params = 280.2922473072302
+I1202 21:03:40.576790 137274321021824 utils.py:1231] [68400] train/loss = 2.0956820100545883
+I1202 21:03:40.576937 137274321021824 utils.py:1231] [68400] l2_grads = 1.8878425359725952
+I1202 21:03:40.577033 137274321021824 utils.py:1231] [68400] lr = 0.000392194376272502
+I1202 21:03:40.577111 137274321021824 utils.py:1231] [68400] uptime = 429209.93947279
+I1202 21:03:40.577183 137274321021824 utils.py:1231] [68400] examples_seen = 70041600.0
+I1202 21:03:40.577256 137274321021824 utils.py:1231] [68400] progress = 0.6074438514071561
+I1202 21:03:40.577319 137274321021824 utils.py:1231] [68400] epoch = 54.670156193532925
+I1202 21:03:40.577375 137274321021824 utils.py:1231] [68400] img/sec/core = 164.2142322689689
+I1202 21:03:40.577437 137274321021824 utils.py:1231] [68400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.19066558452944
+I1202 21:03:40.577494 137274321021824 utils.py:1231] [68400] core_hours = 119.19066558452944
+I1202 21:03:40.577559 137274321021824 train.py:125] NOTE: Steps:68400/112603 [60.7%]
+Walltime:4d23h13m (0s eval)
+ETA:3d5h1m
+Total train time:8d4h13m
+I1202 21:08:52.373572 137274321021824 utils.py:1231] [68450] l2_params = 280.19434314625516
+I1202 21:08:52.373800 137274321021824 utils.py:1231] [68450] train/loss = 2.1689046174287796
+I1202 21:08:52.373906 137274321021824 utils.py:1231] [68450] l2_grads = 1.815629482269287
+I1202 21:08:52.374001 137274321021824 utils.py:1231] [68450] lr = 0.00039144703446949465
+I1202 21:08:52.374058 137274321021824 utils.py:1231] [68450] uptime = 429521.736419396
+I1202 21:08:52.374116 137274321021824 utils.py:1231] [68450] examples_seen = 70092800.0
+I1202 21:08:52.374172 137274321021824 utils.py:1231] [68450] progress = 0.6078878893102315
+I1202 21:08:52.374227 137274321021824 utils.py:1231] [68450] epoch = 54.71011975800189
+I1202 21:08:52.374283 137274321021824 utils.py:1231] [68450] img/sec/core = 164.2094335987702
+I1202 21:08:52.374343 137274321021824 utils.py:1231] [68450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.27727584747555
+I1202 21:08:52.374398 137274321021824 utils.py:1231] [68450] core_hours = 119.27727584747555
+I1202 21:08:52.374461 137274321021824 train.py:125] NOTE: Steps:68450/112603 [60.8%]
+Walltime:4d23h18m (0s eval)
+ETA:3d4h56m
+Total train time:8d4h13m
+I1202 21:14:04.166129 137274321021824 utils.py:1231] [68500] l2_params = 280.0924683606125
+I1202 21:14:04.166397 137274321021824 utils.py:1231] [68500] train/loss = 3.475274533033371
+I1202 21:14:04.166538 137274321021824 utils.py:1231] [68500] l2_grads = 1.61563241481781
+I1202 21:14:04.166630 137274321021824 utils.py:1231] [68500] lr = 0.0003906999470923425
+I1202 21:14:04.166710 137274321021824 utils.py:1231] [68500] uptime = 429833.529067574
+I1202 21:14:04.166785 137274321021824 utils.py:1231] [68500] examples_seen = 70144000.0
+I1202 21:14:04.166858 137274321021824 utils.py:1231] [68500] progress = 0.608331927213307
+I1202 21:14:04.166922 137274321021824 utils.py:1231] [68500] epoch = 54.75008332247084
+I1202 21:14:04.166982 137274321021824 utils.py:1231] [68500] img/sec/core = 164.21169741876045
+I1202 21:14:04.167047 137274321021824 utils.py:1231] [68500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.36388491641388
+I1202 21:14:04.167104 137274321021824 utils.py:1231] [68500] core_hours = 119.36388491641388
+I1202 21:14:04.167184 137274321021824 train.py:125] NOTE: Steps:68500/112603 [60.8%]
+Walltime:4d23h23m (0s eval)
+ETA:3d4h51m
+Total train time:8d4h13m
+I1202 21:19:15.915194 137274321021824 utils.py:1231] [68550] l2_params = 279.99883723134616
+I1202 21:19:15.915440 137274321021824 utils.py:1231] [68550] train/loss = 3.0320151448249817
+I1202 21:19:15.915579 137274321021824 utils.py:1231] [68550] l2_grads = 1.5274215936660767
+I1202 21:19:15.915647 137274321021824 utils.py:1231] [68550] lr = 0.000389953115892065
+I1202 21:19:15.915701 137274321021824 utils.py:1231] [68550] uptime = 430145.278063267
+I1202 21:19:15.915754 137274321021824 utils.py:1231] [68550] examples_seen = 70195200.0
+I1202 21:19:15.915803 137274321021824 utils.py:1231] [68550] progress = 0.6087759651163823
+I1202 21:19:15.915852 137274321021824 utils.py:1231] [68550] epoch = 54.7900468869398
+I1202 21:19:15.915915 137274321021824 utils.py:1231] [68550] img/sec/core = 164.23469107313883
+I1202 21:19:15.915973 137274321021824 utils.py:1231] [68550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.45048185966195
+I1202 21:19:15.916022 137274321021824 utils.py:1231] [68550] core_hours = 119.45048185966195
+I1202 21:19:15.916081 137274321021824 train.py:125] NOTE: Steps:68550/112603 [60.9%]
+Walltime:4d23h29m (0s eval)
+ETA:3d4h45m
+Total train time:8d4h13m
+I1202 21:24:27.605536 137274321021824 utils.py:1231] [68600] l2_params = 279.90773593243694
+I1202 21:24:27.605778 137274321021824 utils.py:1231] [68600] train/loss = 2.6605813205242157
+I1202 21:24:27.605874 137274321021824 utils.py:1231] [68600] l2_grads = 1.7651907205581665
+I1202 21:24:27.605942 137274321021824 utils.py:1231] [68600] lr = 0.0003892065426190808
+I1202 21:24:27.605996 137274321021824 utils.py:1231] [68600] uptime = 430456.968357591
+I1202 21:24:27.606048 137274321021824 utils.py:1231] [68600] examples_seen = 70246400.0
+I1202 21:24:27.606096 137274321021824 utils.py:1231] [68600] progress = 0.6092200030194578
+I1202 21:24:27.606144 137274321021824 utils.py:1231] [68600] epoch = 54.830010451408754
+I1202 21:24:27.606194 137274321021824 utils.py:1231] [68600] img/sec/core = 164.26562178024002
+I1202 21:24:27.606248 137274321021824 utils.py:1231] [68600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.53706249697417
+I1202 21:24:27.606298 137274321021824 utils.py:1231] [68600] core_hours = 119.53706249697417
+I1202 21:24:27.606359 137274321021824 train.py:125] NOTE: Steps:68600/112603 [60.9%]
+Walltime:4d23h34m (0s eval)
+ETA:3d4h40m
+Total train time:8d4h13m
+I1202 21:29:39.413107 137274321021824 utils.py:1231] [68650] l2_params = 279.82156191169184
+I1202 21:29:39.413357 137274321021824 utils.py:1231] [68650] train/loss = 2.167933374643326
+I1202 21:29:39.413487 137274321021824 utils.py:1231] [68650] l2_grads = 1.945890188217163
+I1202 21:29:39.413575 137274321021824 utils.py:1231] [68650] lr = 0.0003884602290232045
+I1202 21:29:39.413640 137274321021824 utils.py:1231] [68650] uptime = 430768.776001432
+I1202 21:29:39.413704 137274321021824 utils.py:1231] [68650] examples_seen = 70297600.0
+I1202 21:29:39.413763 137274321021824 utils.py:1231] [68650] progress = 0.6096640409225331
+I1202 21:29:39.413822 137274321021824 utils.py:1231] [68650] epoch = 54.86997401587771
+I1202 21:29:39.413889 137274321021824 utils.py:1231] [68650] img/sec/core = 164.2038000393427
+I1202 21:29:39.413986 137274321021824 utils.py:1231] [68650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.62367573137445
+I1202 21:29:39.414104 137274321021824 utils.py:1231] [68650] core_hours = 119.62367573137445
+I1202 21:29:39.414202 137274321021824 train.py:125] NOTE: Steps:68650/112603 [61.0%]
+Walltime:4d23h39m (0s eval)
+ETA:3d4h35m
+Total train time:8d4h13m
+I1202 21:34:51.201027 137274321021824 utils.py:1231] [68700] l2_params = 279.72997142907724
+I1202 21:34:51.201275 137274321021824 utils.py:1231] [68700] train/loss = 2.199833944439888
+I1202 21:34:51.201413 137274321021824 utils.py:1231] [68700] l2_grads = 1.8105969429016113
+I1202 21:34:51.201500 137274321021824 utils.py:1231] [68700] lr = 0.0003877141768536419
+I1202 21:34:51.201583 137274321021824 utils.py:1231] [68700] uptime = 431080.563932842
+I1202 21:34:51.201657 137274321021824 utils.py:1231] [68700] examples_seen = 70348800.0
+I1202 21:34:51.201727 137274321021824 utils.py:1231] [68700] progress = 0.6101080788256086
+I1202 21:34:51.201800 137274321021824 utils.py:1231] [68700] epoch = 54.90993758034667
+I1202 21:34:51.201871 137274321021824 utils.py:1231] [68700] img/sec/core = 164.2141816344711
+I1202 21:34:51.201942 137274321021824 utils.py:1231] [68700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.71028349009944
+I1202 21:34:51.202006 137274321021824 utils.py:1231] [68700] core_hours = 119.71028349009944
+I1202 21:34:51.202114 137274321021824 train.py:125] NOTE: Steps:68700/112603 [61.0%]
+Walltime:4d23h44m (0s eval)
+ETA:3d4h30m
+Total train time:8d4h13m
+I1202 21:40:02.993794 137274321021824 utils.py:1231] [68750] l2_params = 279.6441897165973
+I1202 21:40:02.994079 137274321021824 utils.py:1231] [68750] train/loss = 2.1276078671216965
+I1202 21:40:02.994209 137274321021824 utils.py:1231] [68750] l2_grads = 1.866782307624817
+I1202 21:40:02.994319 137274321021824 utils.py:1231] [68750] lr = 0.00038696838785898637
+I1202 21:40:02.994413 137274321021824 utils.py:1231] [68750] uptime = 431392.356772174
+I1202 21:40:02.994517 137274321021824 utils.py:1231] [68750] examples_seen = 70400000.0
+I1202 21:40:02.994596 137274321021824 utils.py:1231] [68750] progress = 0.6105521167286839
+I1202 21:40:02.994676 137274321021824 utils.py:1231] [68750] epoch = 54.94990114481563
+I1202 21:40:02.994775 137274321021824 utils.py:1231] [68750] img/sec/core = 164.21159674381997
+I1202 21:40:02.994853 137274321021824 utils.py:1231] [68750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.79689261213612
+I1202 21:40:02.994951 137274321021824 utils.py:1231] [68750] core_hours = 119.79689261213612
+I1202 21:40:02.995021 137274321021824 train.py:125] NOTE: Steps:68750/112603 [61.1%]
+Walltime:4d23h49m (0s eval)
+ETA:3d4h24m
+Total train time:8d4h12m
+I1202 21:45:14.766098 137274321021824 utils.py:1231] [68800] l2_params = 279.5611842003177
+I1202 21:45:14.766356 137274321021824 utils.py:1231] [68800] train/loss = 1.9987930953502655
+I1202 21:45:14.766499 137274321021824 utils.py:1231] [68800] l2_grads = 1.896294355392456
+I1202 21:45:14.766596 137274321021824 utils.py:1231] [68800] lr = 0.00038622286378721377
+I1202 21:45:14.766673 137274321021824 utils.py:1231] [68800] uptime = 431704.12903430697
+I1202 21:45:14.766747 137274321021824 utils.py:1231] [68800] examples_seen = 70451200.0
+I1202 21:45:14.766823 137274321021824 utils.py:1231] [68800] progress = 0.6109961546317594
+I1202 21:45:14.766895 137274321021824 utils.py:1231] [68800] epoch = 54.98986470928458
+I1202 21:45:14.766964 137274321021824 utils.py:1231] [68800] img/sec/core = 164.22243483021936
+I1202 21:45:14.767029 137274321021824 utils.py:1231] [68800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.88349601828416
+I1202 21:45:14.767090 137274321021824 utils.py:1231] [68800] core_hours = 119.88349601828416
+I1202 21:45:14.767162 137274321021824 train.py:125] NOTE: Steps:68800/112603 [61.1%]
+Walltime:4d23h55m (0s eval)
+ETA:3d4h19m
+Total train time:8d4h12m
+I1202 21:50:26.548392 137274321021824 utils.py:1231] [68850] l2_params = 279.46849087975386
+I1202 21:50:26.548604 137274321021824 utils.py:1231] [68850] train/loss = 4.513080656528473
+I1202 21:50:26.548703 137274321021824 utils.py:1231] [68850] l2_grads = 1.8556915521621704
+I1202 21:50:26.548769 137274321021824 utils.py:1231] [68850] lr = 0.00038547760638567947
+I1202 21:50:26.548827 137274321021824 utils.py:1231] [68850] uptime = 432015.911187984
+I1202 21:50:26.548886 137274321021824 utils.py:1231] [68850] examples_seen = 70502400.0
+I1202 21:50:26.548941 137274321021824 utils.py:1231] [68850] progress = 0.6114401925348347
+I1202 21:50:26.548991 137274321021824 utils.py:1231] [68850] epoch = 55.02982827375354
+I1202 21:50:26.549043 137274321021824 utils.py:1231] [68850] img/sec/core = 164.2172247390361
+I1202 21:50:26.549102 137274321021824 utils.py:1231] [68850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 119.97010217208332
+I1202 21:50:26.549159 137274321021824 utils.py:1231] [68850] core_hours = 119.97010217208332
+I1202 21:50:26.549220 137274321021824 train.py:125] NOTE: Steps:68850/112603 [61.1%]
+Walltime:5d0h0m (0s eval)
+ETA:3d4h14m
+Total train time:8d4h12m
+I1202 21:55:38.331041 137274321021824 utils.py:1231] [68900] l2_params = 279.3750123821913
+I1202 21:55:38.331253 137274321021824 utils.py:1231] [68900] train/loss = 2.028164505958557
+I1202 21:55:38.331359 137274321021824 utils.py:1231] [68900] l2_grads = 1.8855031728744507
+I1202 21:55:38.331442 137274321021824 utils.py:1231] [68900] lr = 0.00038473261740111384
+I1202 21:55:38.331527 137274321021824 utils.py:1231] [68900] uptime = 432327.693882429
+I1202 21:55:38.331624 137274321021824 utils.py:1231] [68900] examples_seen = 70553600.0
+I1202 21:55:38.331709 137274321021824 utils.py:1231] [68900] progress = 0.6118842304379102
+I1202 21:55:38.331799 137274321021824 utils.py:1231] [68900] epoch = 55.06979183822249
+I1202 21:55:38.331865 137274321021824 utils.py:1231] [68900] img/sec/core = 164.21693991432647
+I1202 21:55:38.331952 137274321021824 utils.py:1231] [68900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.05670847609582
+I1202 21:55:38.332035 137274321021824 utils.py:1231] [68900] core_hours = 120.05670847609582
+I1202 21:55:38.332105 137274321021824 train.py:125] NOTE: Steps:68900/112603 [61.2%]
+Walltime:5d0h5m (0s eval)
+ETA:3d4h9m
+Total train time:8d4h12m
+I1202 22:00:50.112977 137274321021824 utils.py:1231] [68950] l2_params = 279.28460955298533
+I1202 22:00:50.113207 137274321021824 utils.py:1231] [68950] train/loss = 1.9119713455438614
+I1202 22:00:50.113302 137274321021824 utils.py:1231] [68950] l2_grads = 1.8735630512237549
+I1202 22:00:50.113364 137274321021824 utils.py:1231] [68950] lr = 0.00038398789857961775
+I1202 22:00:50.113415 137274321021824 utils.py:1231] [68950] uptime = 432639.475777192
+I1202 22:00:50.113472 137274321021824 utils.py:1231] [68950] examples_seen = 70604800.0
+I1202 22:00:50.113521 137274321021824 utils.py:1231] [68950] progress = 0.6123282683409856
+I1202 22:00:50.113571 137274321021824 utils.py:1231] [68950] epoch = 55.109755402691455
+I1202 22:00:50.113622 137274321021824 utils.py:1231] [68950] img/sec/core = 164.21736111046303
+I1202 22:00:50.113690 137274321021824 utils.py:1231] [68950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.14331455797443
+I1202 22:00:50.113750 137274321021824 utils.py:1231] [68950] core_hours = 120.14331455797443
+I1202 22:00:50.113811 137274321021824 train.py:125] NOTE: Steps:68950/112603 [61.2%]
+Walltime:5d0h10m (0s eval)
+ETA:3d4h3m
+Total train time:8d4h12m
+I1202 22:06:01.894410 137274321021824 utils.py:1231] [69000] l2_params = 279.19171794175685
+I1202 22:06:01.894613 137274321021824 utils.py:1231] [69000] train/loss = 3.7500581443309784
+I1202 22:06:01.894726 137274321021824 utils.py:1231] [69000] l2_grads = 1.6790683269500732
+I1202 22:06:01.894800 137274321021824 utils.py:1231] [69000] lr = 0.00038324345166665944
+I1202 22:06:01.894865 137274321021824 utils.py:1231] [69000] uptime = 432951.257226749
+I1202 22:06:01.894938 137274321021824 utils.py:1231] [69000] examples_seen = 70656000.0
+I1202 22:06:01.895001 137274321021824 utils.py:1231] [69000] progress = 0.612772306244061
+I1202 22:06:01.895060 137274321021824 utils.py:1231] [69000] epoch = 55.14971896716041
+I1202 22:06:01.895116 137274321021824 utils.py:1231] [69000] img/sec/core = 164.2175956034263
+I1202 22:06:01.895179 137274321021824 utils.py:1231] [69000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.2299205161847
+I1202 22:06:01.895236 137274321021824 utils.py:1231] [69000] core_hours = 120.2299205161847
+I1202 22:06:01.895302 137274321021824 train.py:125] NOTE: Steps:69000/112603 [61.3%]
+Walltime:5d0h15m (0s eval)
+ETA:3d3h58m
+Total train time:8d4h12m
+I1202 22:11:14.059836 137274321021824 utils.py:1231] [69050] l2_params = 279.1028427831438
+I1202 22:11:14.060055 137274321021824 utils.py:1231] [69050] train/loss = 2.075592190027237
+I1202 22:11:14.060160 137274321021824 utils.py:1231] [69050] l2_grads = 1.7950719594955444
+I1202 22:11:14.060240 137274321021824 utils.py:1231] [69050] lr = 0.0003824992784070699
+I1202 22:11:14.060300 137274321021824 utils.py:1231] [69050] uptime = 433263.422661398
+I1202 22:11:14.060359 137274321021824 utils.py:1231] [69050] examples_seen = 70707200.0
+I1202 22:11:14.060416 137274321021824 utils.py:1231] [69050] progress = 0.6132163441471364
+I1202 22:11:14.060475 137274321021824 utils.py:1231] [69050] epoch = 55.189682531629366
+I1202 22:11:14.060531 137274321021824 utils.py:1231] [69050] img/sec/core = 164.01559659403992
+I1202 22:11:14.060593 137274321021824 utils.py:1231] [69050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.31663313692056
+I1202 22:11:14.060647 137274321021824 utils.py:1231] [69050] core_hours = 120.31663313692056
+I1202 22:11:14.060719 137274321021824 train.py:125] NOTE: Steps:69050/112603 [61.3%]
+Walltime:5d0h21m (0s eval)
+ETA:3d3h53m
+Total train time:8d4h12m
+I1202 22:16:25.792549 137274321021824 utils.py:1231] [69100] l2_params = 279.0089105589759
+I1202 22:16:25.792824 137274321021824 utils.py:1231] [69100] train/loss = 3.210145264863968
+I1202 22:16:25.793042 137274321021824 utils.py:1231] [69100] l2_grads = 1.6130424737930298
+I1202 22:16:25.793120 137274321021824 utils.py:1231] [69100] lr = 0.00038175538054503765
+I1202 22:16:25.793183 137274321021824 utils.py:1231] [69100] uptime = 433575.155541214
+I1202 22:16:25.793236 137274321021824 utils.py:1231] [69100] examples_seen = 70758400.0
+I1202 22:16:25.793286 137274321021824 utils.py:1231] [69100] progress = 0.6136603820502118
+I1202 22:16:25.793335 137274321021824 utils.py:1231] [69100] epoch = 55.22964609609832
+I1202 22:16:25.793385 137274321021824 utils.py:1231] [69100] img/sec/core = 164.243181631098
+I1202 22:16:25.793443 137274321021824 utils.py:1231] [69100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.40322560353611
+I1202 22:16:25.793495 137274321021824 utils.py:1231] [69100] core_hours = 120.40322560353611
+I1202 22:16:25.793555 137274321021824 train.py:125] NOTE: Steps:69100/112603 [61.4%]
+Walltime:5d0h26m (0s eval)
+ETA:3d3h48m
+Total train time:8d4h12m
+I1202 22:21:37.577181 137274321021824 utils.py:1231] [69150] l2_params = 278.9201672604151
+I1202 22:21:37.577395 137274321021824 utils.py:1231] [69150] train/loss = 4.159113585948944
+I1202 22:21:37.577492 137274321021824 utils.py:1231] [69150] l2_grads = 1.6849702596664429
+I1202 22:21:37.577554 137274321021824 utils.py:1231] [69150] lr = 0.00038101175982410685
+I1202 22:21:37.577606 137274321021824 utils.py:1231] [69150] uptime = 433886.939967963
+I1202 22:21:37.577668 137274321021824 utils.py:1231] [69150] examples_seen = 70809600.0
+I1202 22:21:37.577718 137274321021824 utils.py:1231] [69150] progress = 0.6141044199532872
+I1202 22:21:37.577769 137274321021824 utils.py:1231] [69150] epoch = 55.269609660567276
+I1202 22:21:37.577826 137274321021824 utils.py:1231] [69150] img/sec/core = 164.21602750935543
+I1202 22:21:37.577901 137274321021824 utils.py:1231] [69150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.48983238874416
+I1202 22:21:37.577955 137274321021824 utils.py:1231] [69150] core_hours = 120.48983238874416
+I1202 22:21:37.578020 137274321021824 train.py:125] NOTE: Steps:69150/112603 [61.4%]
+Walltime:5d0h31m (0s eval)
+ETA:3d3h42m
+Total train time:8d4h12m
+I1202 22:26:49.354332 137274321021824 utils.py:1231] [69200] l2_params = 278.8282934265825
+I1202 22:26:49.354600 137274321021824 utils.py:1231] [69200] train/loss = 3.1832350492477417
+I1202 22:26:49.354732 137274321021824 utils.py:1231] [69200] l2_grads = 1.631007432937622
+I1202 22:26:49.354818 137274321021824 utils.py:1231] [69200] lr = 0.0003802684179871713
+I1202 22:26:49.354899 137274321021824 utils.py:1231] [69200] uptime = 434198.717249929
+I1202 22:26:49.354976 137274321021824 utils.py:1231] [69200] examples_seen = 70860800.0
+I1202 22:26:49.355064 137274321021824 utils.py:1231] [69200] progress = 0.6145484578563626
+I1202 22:26:49.355150 137274321021824 utils.py:1231] [69200] epoch = 55.30957322503624
+I1202 22:26:49.355225 137274321021824 utils.py:1231] [69200] img/sec/core = 164.21979073376846
+I1202 22:26:49.355302 137274321021824 utils.py:1231] [69200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.57643718929027
+I1202 22:26:49.355376 137274321021824 utils.py:1231] [69200] core_hours = 120.57643718929027
+I1202 22:26:49.355461 137274321021824 train.py:125] NOTE: Steps:69200/112603 [61.5%]
+Walltime:5d0h36m (0s eval)
+ETA:3d3h37m
+Total train time:8d4h12m
+I1202 22:32:01.152377 137274321021824 utils.py:1231] [69250] l2_params = 278.73684439053625
+I1202 22:32:01.152615 137274321021824 utils.py:1231] [69250] train/loss = 2.0318160504102707
+I1202 22:32:01.152707 137274321021824 utils.py:1231] [69250] l2_grads = 1.8248876333236694
+I1202 22:32:01.152768 137274321021824 utils.py:1231] [69250] lr = 0.00037952535677647275
+I1202 22:32:01.152821 137274321021824 utils.py:1231] [69250] uptime = 434510.515182861
+I1202 22:32:01.152875 137274321021824 utils.py:1231] [69250] examples_seen = 70912000.0
+I1202 22:32:01.152934 137274321021824 utils.py:1231] [69250] progress = 0.614992495759438
+I1202 22:32:01.152982 137274321021824 utils.py:1231] [69250] epoch = 55.349536789505194
+I1202 22:32:01.153031 137274321021824 utils.py:1231] [69250] img/sec/core = 164.2089141468431
+I1202 22:32:01.153086 137274321021824 utils.py:1231] [69250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.66304772621584
+I1202 22:32:01.153135 137274321021824 utils.py:1231] [69250] core_hours = 120.66304772621584
+I1202 22:32:01.153195 137274321021824 train.py:125] NOTE: Steps:69250/112603 [61.5%]
+Walltime:5d0h41m (0s eval)
+ETA:3d3h32m
+Total train time:8d4h12m
+I1202 22:37:12.945857 137274321021824 utils.py:1231] [69300] l2_params = 278.650082179086
+I1202 22:37:12.946098 137274321021824 utils.py:1231] [69300] train/loss = 2.0036265552043915
+I1202 22:37:12.946209 137274321021824 utils.py:1231] [69300] l2_grads = 1.8710678815841675
+I1202 22:37:12.946293 137274321021824 utils.py:1231] [69300] lr = 0.00037878257793359327
+I1202 22:37:12.946374 137274321021824 utils.py:1231] [69300] uptime = 434822.308732472
+I1202 22:37:12.946438 137274321021824 utils.py:1231] [69300] examples_seen = 70963200.0
+I1202 22:37:12.946497 137274321021824 utils.py:1231] [69300] progress = 0.6154365336625134
+I1202 22:37:12.946557 137274321021824 utils.py:1231] [69300] epoch = 55.38950035397415
+I1202 22:37:12.946619 137274321021824 utils.py:1231] [69300] img/sec/core = 164.2112226628104
+I1202 22:37:12.946682 137274321021824 utils.py:1231] [69300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.74965704555221
+I1202 22:37:12.946741 137274321021824 utils.py:1231] [69300] core_hours = 120.74965704555221
+I1202 22:37:12.946808 137274321021824 train.py:125] NOTE: Steps:69300/112603 [61.5%]
+Walltime:5d0h47m (0s eval)
+ETA:3d3h27m
+Total train time:8d4h12m
+I1202 22:42:24.741824 137274321021824 utils.py:1231] [69350] l2_params = 278.5589694013792
+I1202 22:42:24.742121 137274321021824 utils.py:1231] [69350] train/loss = 2.028263598680496
+I1202 22:42:24.742234 137274321021824 utils.py:1231] [69350] l2_grads = 1.8897299766540527
+I1202 22:42:24.742304 137274321021824 utils.py:1231] [69350] lr = 0.0003780400831994539
+I1202 22:42:24.742363 137274321021824 utils.py:1231] [69350] uptime = 435134.104724655
+I1202 22:42:24.742425 137274321021824 utils.py:1231] [69350] examples_seen = 71014400.0
+I1202 22:42:24.742479 137274321021824 utils.py:1231] [69350] progress = 0.6158805715655888
+I1202 22:42:24.742536 137274321021824 utils.py:1231] [69350] epoch = 55.429463918443105
+I1202 22:42:24.742590 137274321021824 utils.py:1231] [69350] img/sec/core = 164.20993625201461
+I1202 22:42:24.742652 137274321021824 utils.py:1231] [69350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.83626704338083
+I1202 22:42:24.742707 137274321021824 utils.py:1231] [69350] core_hours = 120.83626704338083
+I1202 22:42:24.742770 137274321021824 train.py:125] NOTE: Steps:69350/112603 [61.6%]
+Walltime:5d0h52m (0s eval)
+ETA:3d3h22m
+Total train time:8d4h12m
+I1202 22:47:36.526951 137274321021824 utils.py:1231] [69400] l2_params = 278.4750575506132
+I1202 22:47:36.527164 137274321021824 utils.py:1231] [69400] train/loss = 2.1073948740959167
+I1202 22:47:36.527257 137274321021824 utils.py:1231] [69400] l2_grads = 2.1013739109039307
+I1202 22:47:36.527318 137274321021824 utils.py:1231] [69400] lr = 0.0003772978743143094
+I1202 22:47:36.527373 137274321021824 utils.py:1231] [69400] uptime = 435445.889735398
+I1202 22:47:36.527425 137274321021824 utils.py:1231] [69400] examples_seen = 71065600.0
+I1202 22:47:36.527474 137274321021824 utils.py:1231] [69400] progress = 0.6163246094686643
+I1202 22:47:36.527522 137274321021824 utils.py:1231] [69400] epoch = 55.46942748291207
+I1202 22:47:36.527573 137274321021824 utils.py:1231] [69400] img/sec/core = 164.21571992183192
+I1202 22:47:36.527630 137274321021824 utils.py:1231] [69400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 120.92287399080944
+I1202 22:47:36.527680 137274321021824 utils.py:1231] [69400] core_hours = 120.92287399080944
+I1202 22:47:36.527740 137274321021824 train.py:125] NOTE: Steps:69400/112603 [61.6%]
+Walltime:5d0h57m (0s eval)
+ETA:3d3h16m
+Total train time:8d4h12m
+I1202 22:52:48.315809 137274321021824 utils.py:1231] [69450] l2_params = 278.383901378063
+I1202 22:52:48.316072 137274321021824 utils.py:1231] [69450] train/loss = 3.2505763471126556
+I1202 22:52:48.316184 137274321021824 utils.py:1231] [69450] l2_grads = 1.6687060594558716
+I1202 22:52:48.316256 137274321021824 utils.py:1231] [69450] lr = 0.0003765559530177456
+I1202 22:52:48.316314 137274321021824 utils.py:1231] [69450] uptime = 435757.678675504
+I1202 22:52:48.316368 137274321021824 utils.py:1231] [69450] examples_seen = 71116800.0
+I1202 22:52:48.316418 137274321021824 utils.py:1231] [69450] progress = 0.6167686473717396
+I1202 22:52:48.316468 137274321021824 utils.py:1231] [69450] epoch = 55.50939104738102
+I1202 22:52:48.316520 137274321021824 utils.py:1231] [69450] img/sec/core = 164.21365037064567
+I1202 22:52:48.316579 137274321021824 utils.py:1231] [69450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.00948202972776
+I1202 22:52:48.316630 137274321021824 utils.py:1231] [69450] core_hours = 121.00948202972776
+I1202 22:52:48.316691 137274321021824 train.py:125] NOTE: Steps:69450/112603 [61.7%]
+Walltime:5d1h2m (0s eval)
+ETA:3d3h11m
+Total train time:8d4h12m
+I1202 22:58:00.021627 137274321021824 utils.py:1231] [69500] l2_params = 278.28650641148954
+I1202 22:58:00.021852 137274321021824 utils.py:1231] [69500] train/loss = 2.1802430152893066
+I1202 22:58:00.021950 137274321021824 utils.py:1231] [69500] l2_grads = 1.9015930891036987
+I1202 22:58:00.022009 137274321021824 utils.py:1231] [69500] lr = 0.00037581432104867325
+I1202 22:58:00.022064 137274321021824 utils.py:1231] [69500] uptime = 436069.38442655397
+I1202 22:58:00.022116 137274321021824 utils.py:1231] [69500] examples_seen = 71168000.0
+I1202 22:58:00.022165 137274321021824 utils.py:1231] [69500] progress = 0.6172126852748151
+I1202 22:58:00.022212 137274321021824 utils.py:1231] [69500] epoch = 55.54935461184998
+I1202 22:58:00.022263 137274321021824 utils.py:1231] [69500] img/sec/core = 164.25747624974562
+I1202 22:58:00.022318 137274321021824 utils.py:1231] [69500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.09606696057499
+I1202 22:58:00.022367 137274321021824 utils.py:1231] [69500] core_hours = 121.09606696057499
+I1202 22:58:00.022425 137274321021824 train.py:125] NOTE: Steps:69500/112603 [61.7%]
+Walltime:5d1h7m (0s eval)
+ETA:3d3h6m
+Total train time:8d4h12m
+I1202 23:03:11.730335 137274321021824 utils.py:1231] [69550] l2_params = 278.2104611633333
+I1202 23:03:11.730618 137274321021824 utils.py:1231] [69550] train/loss = 1.9983685165643692
+I1202 23:03:11.730737 137274321021824 utils.py:1231] [69550] l2_grads = 1.810840129852295
+I1202 23:03:11.730821 137274321021824 utils.py:1231] [69550] lr = 0.00037507298014532607
+I1202 23:03:11.730896 137274321021824 utils.py:1231] [69550] uptime = 436381.093256944
+I1202 23:03:11.730957 137274321021824 utils.py:1231] [69550] examples_seen = 71219200.0
+I1202 23:03:11.731015 137274321021824 utils.py:1231] [69550] progress = 0.6176567231778904
+I1202 23:03:11.731071 137274321021824 utils.py:1231] [69550] epoch = 55.58931817631893
+I1202 23:03:11.731128 137274321021824 utils.py:1231] [69550] img/sec/core = 164.25585356673636
+I1202 23:03:11.731195 137274321021824 utils.py:1231] [69550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.18265274679443
+I1202 23:03:11.731267 137274321021824 utils.py:1231] [69550] core_hours = 121.18265274679443
+I1202 23:03:11.731345 137274321021824 train.py:125] NOTE: Steps:69550/112603 [61.8%]
+Walltime:5d1h13m (0s eval)
+ETA:3d3h1m
+Total train time:8d4h12m
+I1202 23:08:23.496629 137274321021824 utils.py:1231] [69600] l2_params = 278.11729322417693
+I1202 23:08:23.496890 137274321021824 utils.py:1231] [69600] train/loss = 1.9776168167591095
+I1202 23:08:23.497012 137274321021824 utils.py:1231] [69600] l2_grads = 1.9010443687438965
+I1202 23:08:23.497100 137274321021824 utils.py:1231] [69600] lr = 0.0003743319320452542
+I1202 23:08:23.497180 137274321021824 utils.py:1231] [69600] uptime = 436692.85954068496
+I1202 23:08:23.497251 137274321021824 utils.py:1231] [69600] examples_seen = 71270400.0
+I1202 23:08:23.497308 137274321021824 utils.py:1231] [69600] progress = 0.6181007610809659
+I1202 23:08:23.497363 137274321021824 utils.py:1231] [69600] epoch = 55.62928174078789
+I1202 23:08:23.497420 137274321021824 utils.py:1231] [69600] img/sec/core = 164.2255839394628
+I1202 23:08:23.497482 137274321021824 utils.py:1231] [69600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.26925449227805
+I1202 23:08:23.497546 137274321021824 utils.py:1231] [69600] core_hours = 121.26925449227805
+I1202 23:08:23.497614 137274321021824 train.py:125] NOTE: Steps:69600/112603 [61.8%]
+Walltime:5d1h18m (0s eval)
+ETA:3d2h55m
+Total train time:8d4h12m
+I1202 23:13:35.287838 137274321021824 utils.py:1231] [69650] l2_params = 278.0217098815171
+I1202 23:13:35.288042 137274321021824 utils.py:1231] [69650] train/loss = 2.2884572446346283
+I1202 23:13:35.288144 137274321021824 utils.py:1231] [69650] l2_grads = 1.7707393169403076
+I1202 23:13:35.288210 137274321021824 utils.py:1231] [69650] lr = 0.00037359117848532195
+I1202 23:13:35.288266 137274321021824 utils.py:1231] [69650] uptime = 437004.650628466
+I1202 23:13:35.288328 137274321021824 utils.py:1231] [69650] examples_seen = 71321600.0
+I1202 23:13:35.288408 137274321021824 utils.py:1231] [69650] progress = 0.6185447989840412
+I1202 23:13:35.288472 137274321021824 utils.py:1231] [69650] epoch = 55.66924530525685
+I1202 23:13:35.288539 137274321021824 utils.py:1231] [69650] img/sec/core = 164.21251923646741
+I1202 23:13:35.288614 137274321021824 utils.py:1231] [69650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.35586312777278
+I1202 23:13:35.288686 137274321021824 utils.py:1231] [69650] core_hours = 121.35586312777278
+I1202 23:13:35.288767 137274321021824 train.py:125] NOTE: Steps:69650/112603 [61.9%]
+Walltime:5d1h23m (0s eval)
+ETA:3d2h50m
+Total train time:8d4h12m
+I1202 23:18:47.073107 137274321021824 utils.py:1231] [69700] l2_params = 277.93034583876295
+I1202 23:18:47.073345 137274321021824 utils.py:1231] [69700] train/loss = 4.373535752296448
+I1202 23:18:47.073444 137274321021824 utils.py:1231] [69700] l2_grads = 1.7174826860427856
+I1202 23:18:47.073503 137274321021824 utils.py:1231] [69700] lr = 0.00037285072120170424
+I1202 23:18:47.073554 137274321021824 utils.py:1231] [69700] uptime = 437316.43591659697
+I1202 23:18:47.073607 137274321021824 utils.py:1231] [69700] examples_seen = 71372800.0
+I1202 23:18:47.073656 137274321021824 utils.py:1231] [69700] progress = 0.6189888368871167
+I1202 23:18:47.073704 137274321021824 utils.py:1231] [69700] epoch = 55.709208869725806
+I1202 23:18:47.073756 137274321021824 utils.py:1231] [69700] img/sec/core = 164.21557382300506
+I1202 23:18:47.073820 137274321021824 utils.py:1231] [69700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.4424701522536
+I1202 23:18:47.073878 137274321021824 utils.py:1231] [69700] core_hours = 121.4424701522536
+I1202 23:18:47.073954 137274321021824 train.py:125] NOTE: Steps:69700/112603 [61.9%]
+Walltime:5d1h28m (0s eval)
+ETA:3d2h45m
+Total train time:8d4h12m
+I1202 23:23:58.856264 137274321021824 utils.py:1231] [69750] l2_params = 277.84898015965723
+I1202 23:23:58.856527 137274321021824 utils.py:1231] [69750] train/loss = 2.2093786746263504
+I1202 23:23:58.856651 137274321021824 utils.py:1231] [69750] l2_grads = 1.859903335571289
+I1202 23:23:58.856722 137274321021824 utils.py:1231] [69750] lr = 0.0003721105619298805
+I1202 23:23:58.856789 137274321021824 utils.py:1231] [69750] uptime = 437628.21914733597
+I1202 23:23:58.856845 137274321021824 utils.py:1231] [69750] examples_seen = 71424000.0
+I1202 23:23:58.856906 137274321021824 utils.py:1231] [69750] progress = 0.619432874790192
+I1202 23:23:58.856957 137274321021824 utils.py:1231] [69750] epoch = 55.74917243419476
+I1202 23:23:58.857009 137274321021824 utils.py:1231] [69750] img/sec/core = 164.21665744704424
+I1202 23:23:58.857068 137274321021824 utils.py:1231] [69750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.52907660523665
+I1202 23:23:58.857120 137274321021824 utils.py:1231] [69750] core_hours = 121.52907660523665
+I1202 23:23:58.857183 137274321021824 train.py:125] NOTE: Steps:69750/112603 [61.9%]
+Walltime:5d1h33m (0s eval)
+ETA:3d2h40m
+Total train time:8d4h11m
+I1202 23:29:10.638767 137274321021824 utils.py:1231] [69800] l2_params = 277.75437049298483
+I1202 23:29:10.639005 137274321021824 utils.py:1231] [69800] train/loss = 2.2031882107257843
+I1202 23:29:10.639111 137274321021824 utils.py:1231] [69800] l2_grads = 1.8312355279922485
+I1202 23:29:10.639181 137274321021824 utils.py:1231] [69800] lr = 0.00037137070240463184
+I1202 23:29:10.639255 137274321021824 utils.py:1231] [69800] uptime = 437940.001616337
+I1202 23:29:10.639315 137274321021824 utils.py:1231] [69800] examples_seen = 71475200.0
+I1202 23:29:10.639371 137274321021824 utils.py:1231] [69800] progress = 0.6198769126932675
+I1202 23:29:10.639426 137274321021824 utils.py:1231] [69800] epoch = 55.78913599866372
+I1202 23:29:10.639497 137274321021824 utils.py:1231] [69800] img/sec/core = 164.21705865647328
+I1202 23:29:10.639584 137274321021824 utils.py:1231] [69800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.61568284662582
+I1202 23:29:10.639666 137274321021824 utils.py:1231] [69800] core_hours = 121.61568284662582
+I1202 23:29:10.639770 137274321021824 train.py:125] NOTE: Steps:69800/112603 [62.0%]
+Walltime:5d1h39m (0s eval)
+ETA:3d2h34m
+Total train time:8d4h11m
+I1202 23:34:22.443035 137274321021824 utils.py:1231] [69850] l2_params = 277.66278868797053
+I1202 23:34:22.443306 137274321021824 utils.py:1231] [69850] train/loss = 1.977570280432701
+I1202 23:34:22.443433 137274321021824 utils.py:1231] [69850] l2_grads = 1.7821303606033325
+I1202 23:34:22.443531 137274321021824 utils.py:1231] [69850] lr = 0.0003706311443600378
+I1202 23:34:22.443603 137274321021824 utils.py:1231] [69850] uptime = 438251.80596409
+I1202 23:34:22.443677 137274321021824 utils.py:1231] [69850] examples_seen = 71526400.0
+I1202 23:34:22.443745 137274321021824 utils.py:1231] [69850] progress = 0.620320950596343
+I1202 23:34:22.443808 137274321021824 utils.py:1231] [69850] epoch = 55.82909956313267
+I1202 23:34:22.443868 137274321021824 utils.py:1231] [69850] img/sec/core = 164.20553583992555
+I1202 23:34:22.443948 137274321021824 utils.py:1231] [69850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.7022951654461
+I1202 23:34:22.444003 137274321021824 utils.py:1231] [69850] core_hours = 121.7022951654461
+I1202 23:34:22.444065 137274321021824 train.py:125] NOTE: Steps:69850/112603 [62.0%]
+Walltime:5d1h44m (0s eval)
+ETA:3d2h29m
+Total train time:8d4h11m
+I1202 23:39:34.261423 137274321021824 utils.py:1231] [69900] l2_params = 277.56632650914213
+I1202 23:39:34.261639 137274321021824 utils.py:1231] [69900] train/loss = 2.142068326473236
+I1202 23:39:34.261738 137274321021824 utils.py:1231] [69900] l2_grads = 1.9501787424087524
+I1202 23:39:34.261808 137274321021824 utils.py:1231] [69900] lr = 0.00036989188952946974
+I1202 23:39:34.261868 137274321021824 utils.py:1231] [69900] uptime = 438563.6242293
+I1202 23:39:34.261933 137274321021824 utils.py:1231] [69900] examples_seen = 71577600.0
+I1202 23:39:34.261987 137274321021824 utils.py:1231] [69900] progress = 0.6207649884994183
+I1202 23:39:34.262042 137274321021824 utils.py:1231] [69900] epoch = 55.869063127601635
+I1202 23:39:34.262098 137274321021824 utils.py:1231] [69900] img/sec/core = 164.19820681612325
+I1202 23:39:34.262166 137274321021824 utils.py:1231] [69900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.78891135022667
+I1202 23:39:34.262220 137274321021824 utils.py:1231] [69900] core_hours = 121.78891135022667
+I1202 23:39:34.262297 137274321021824 train.py:125] NOTE: Steps:69900/112603 [62.1%]
+Walltime:5d1h49m (0s eval)
+ETA:3d2h24m
+Total train time:8d4h11m
+I1202 23:44:46.062765 137274321021824 utils.py:1231] [69950] l2_params = 277.47290453446533
+I1202 23:44:46.063057 137274321021824 utils.py:1231] [69950] train/loss = 4.314089775085449
+I1202 23:44:46.063289 137274321021824 utils.py:1231] [69950] l2_grads = 1.758893609046936
+I1202 23:44:46.063400 137274321021824 utils.py:1231] [69950] lr = 0.0003691529396455893
+I1202 23:44:46.063493 137274321021824 utils.py:1231] [69950] uptime = 438875.4258419
+I1202 23:44:46.063577 137274321021824 utils.py:1231] [69950] examples_seen = 71628800.0
+I1202 23:44:46.063654 137274321021824 utils.py:1231] [69950] progress = 0.6212090264024938
+I1202 23:44:46.063728 137274321021824 utils.py:1231] [69950] epoch = 55.90902669207059
+I1202 23:44:46.063793 137274321021824 utils.py:1231] [69950] img/sec/core = 164.20697626630027
+I1202 23:44:46.063876 137274321021824 utils.py:1231] [69950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.87552290928221
+I1202 23:44:46.063978 137274321021824 utils.py:1231] [69950] core_hours = 121.87552290928221
+I1202 23:44:46.064064 137274321021824 train.py:125] NOTE: Steps:69950/112603 [62.1%]
+Walltime:5d1h54m (0s eval)
+ETA:3d2h19m
+Total train time:8d4h11m
+I1202 23:49:57.860193 137274321021824 utils.py:1231] [70000] l2_params = 277.3862042039067
+I1202 23:49:57.860419 137274321021824 utils.py:1231] [70000] train/loss = 2.820934444665909
+I1202 23:49:57.860516 137274321021824 utils.py:1231] [70000] l2_grads = 1.5919538736343384
+I1202 23:49:57.860577 137274321021824 utils.py:1231] [70000] lr = 0.0003684142964403433
+I1202 23:49:57.860634 137274321021824 utils.py:1231] [70000] uptime = 439187.222995916
+I1202 23:49:57.860687 137274321021824 utils.py:1231] [70000] examples_seen = 71680000.0
+I1202 23:49:57.860737 137274321021824 utils.py:1231] [70000] progress = 0.6216530643055691
+I1202 23:49:57.860787 137274321021824 utils.py:1231] [70000] epoch = 55.948990256539545
+I1202 23:49:57.860838 137274321021824 utils.py:1231] [70000] img/sec/core = 164.20932436534133
+I1202 23:49:57.860904 137274321021824 utils.py:1231] [70000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 121.96213322984221
+I1202 23:49:57.860955 137274321021824 utils.py:1231] [70000] core_hours = 121.96213322984221
+I1202 23:49:57.861015 137274321021824 train.py:125] NOTE: Steps:70000/112603 [62.2%]
+Walltime:5d1h59m (0s eval)
+ETA:3d2h13m
+Total train time:8d4h11m
+I1202 23:49:58.216814 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:70000/112603 [62.2%]
+Walltime:5d1h59m (0s eval)
+ETA:3d2h13m
+Total train time:8d4h11m
+I1202 23:51:35.203155 137274321021824 utils.py:1231] [70000] val/acc@1 = 0.6947544642857143
+I1202 23:51:35.203393 137274321021824 utils.py:1231] [70000] val/loss = 1.2311567530340077
+I1202 23:51:35.203549 137274321021824 utils.py:1231] [70000] z/secs/eval/val = 96.98648774600588
+I1202 23:51:35.203643 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 96.98648774600588
+I1202 23:56:45.647460 137274321021824 utils.py:1231] [70050] l2_params = 277.3006908295855
+I1202 23:56:45.647684 137274321021824 utils.py:1231] [70050] train/loss = 2.257009834051132
+I1202 23:56:45.647844 137274321021824 utils.py:1231] [70050] l2_grads = 1.923614501953125
+I1202 23:56:45.647947 137274321021824 utils.py:1231] [70050] lr = 0.00036767596164495966
+I1202 23:56:45.648010 137274321021824 utils.py:1231] [70050] uptime = 439595.010371572
+I1202 23:56:45.648071 137274321021824 utils.py:1231] [70050] examples_seen = 71731200.0
+I1202 23:56:45.648128 137274321021824 utils.py:1231] [70050] progress = 0.6220971022086446
+I1202 23:56:45.648189 137274321021824 utils.py:1231] [70050] epoch = 55.9889538210085
+I1202 23:56:45.648247 137274321021824 utils.py:1231] [70050] img/sec/core = 125.55562789954851
+I1202 23:56:45.648303 137274321021824 utils.py:1231] [70050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.07540750085778
+I1202 23:56:45.648357 137274321021824 utils.py:1231] [70050] core_hours = 122.07540750085778
+I1202 23:56:45.648428 137274321021824 train.py:125] NOTE: Steps:70050/112603 [62.2%]
+Walltime:5d2h6m (0s eval)
+ETA:3d2h9m
+Total train time:8d4h14m
+I1203 00:01:57.436229 137274321021824 utils.py:1231] [70100] l2_params = 277.2125327647592
+I1203 00:01:57.436431 137274321021824 utils.py:1231] [70100] train/loss = 3.5247892439365387
+I1203 00:01:57.436533 137274321021824 utils.py:1231] [70100] l2_grads = 1.7442774772644043
+I1203 00:01:57.436595 137274321021824 utils.py:1231] [70100] lr = 0.00036693793698994355
+I1203 00:01:57.436648 137274321021824 utils.py:1231] [70100] uptime = 439906.799009743
+I1203 00:01:57.436702 137274321021824 utils.py:1231] [70100] examples_seen = 71782400.0
+I1203 00:01:57.436752 137274321021824 utils.py:1231] [70100] progress = 0.6225411401117199
+I1203 00:01:57.436811 137274321021824 utils.py:1231] [70100] epoch = 56.028917385477456
+I1203 00:01:57.436863 137274321021824 utils.py:1231] [70100] img/sec/core = 164.21380939456236
+I1203 00:01:57.436929 137274321021824 utils.py:1231] [70100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.16201545590526
+I1203 00:01:57.436981 137274321021824 utils.py:1231] [70100] core_hours = 122.16201545590526
+I1203 00:01:57.437052 137274321021824 train.py:125] NOTE: Steps:70100/112603 [62.3%]
+Walltime:5d2h11m (0s eval)
+ETA:3d2h4m
+Total train time:8d4h14m
+I1203 00:07:09.209535 137274321021824 utils.py:1231] [70150] l2_params = 277.1310796880561
+I1203 00:07:09.209796 137274321021824 utils.py:1231] [70150] train/loss = 2.268866181373596
+I1203 00:07:09.209947 137274321021824 utils.py:1231] [70150] l2_grads = 1.8529456853866577
+I1203 00:07:09.210081 137274321021824 utils.py:1231] [70150] lr = 0.0003662002242050735
+I1203 00:07:09.210150 137274321021824 utils.py:1231] [70150] uptime = 440218.572510575
+I1203 00:07:09.210208 137274321021824 utils.py:1231] [70150] examples_seen = 71833600.0
+I1203 00:07:09.210268 137274321021824 utils.py:1231] [70150] progress = 0.6229851780147954
+I1203 00:07:09.210323 137274321021824 utils.py:1231] [70150] epoch = 56.06888094994642
+I1203 00:07:09.210382 137274321021824 utils.py:1231] [70150] img/sec/core = 164.22178236239597
+I1203 00:07:09.210440 137274321021824 utils.py:1231] [70150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.2486192061364
+I1203 00:07:09.210498 137274321021824 utils.py:1231] [70150] core_hours = 122.2486192061364
+I1203 00:07:09.210571 137274321021824 train.py:125] NOTE: Steps:70150/112603 [62.3%]
+Walltime:5d2h16m (0s eval)
+ETA:3d1h59m
+Total train time:8d4h14m
+I1203 00:12:20.970700 137274321021824 utils.py:1231] [70200] l2_params = 277.04002952595164
+I1203 00:12:20.970948 137274321021824 utils.py:1231] [70200] train/loss = 4.250284373760223
+I1203 00:12:20.971099 137274321021824 utils.py:1231] [70200] l2_grads = 1.748180627822876
+I1203 00:12:20.971204 137274321021824 utils.py:1231] [70200] lr = 0.00036546282501939605
+I1203 00:12:20.971281 137274321021824 utils.py:1231] [70200] uptime = 440530.333638558
+I1203 00:12:20.971362 137274321021824 utils.py:1231] [70200] examples_seen = 71884800.0
+I1203 00:12:20.971443 137274321021824 utils.py:1231] [70200] progress = 0.6234292159178707
+I1203 00:12:20.971527 137274321021824 utils.py:1231] [70200] epoch = 56.108844514415374
+I1203 00:12:20.971598 137274321021824 utils.py:1231] [70200] img/sec/core = 164.22829982445407
+I1203 00:12:20.971680 137274321021824 utils.py:1231] [70200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.335219519465
+I1203 00:12:20.971747 137274321021824 utils.py:1231] [70200] core_hours = 122.335219519465
+I1203 00:12:20.971817 137274321021824 train.py:125] NOTE: Steps:70200/112603 [62.3%]
+Walltime:5d2h22m (0s eval)
+ETA:3d1h53m
+Total train time:8d4h14m
+I1203 00:17:32.652011 137274321021824 utils.py:1231] [70250] l2_params = 276.95096142621867
+I1203 00:17:32.652298 137274321021824 utils.py:1231] [70250] train/loss = 4.13088059425354
+I1203 00:17:32.652481 137274321021824 utils.py:1231] [70250] l2_grads = 1.7940438985824585
+I1203 00:17:32.652592 137274321021824 utils.py:1231] [70250] lr = 0.00036472574116122423
+I1203 00:17:32.652667 137274321021824 utils.py:1231] [70250] uptime = 440842.01502828096
+I1203 00:17:32.652729 137274321021824 utils.py:1231] [70250] examples_seen = 71936000.0
+I1203 00:17:32.652785 137274321021824 utils.py:1231] [70250] progress = 0.6238732538209462
+I1203 00:17:32.652840 137274321021824 utils.py:1231] [70250] epoch = 56.14880807888433
+I1203 00:17:32.652901 137274321021824 utils.py:1231] [70250] img/sec/core = 164.2703147772387
+I1203 00:17:32.652965 137274321021824 utils.py:1231] [70250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.42179768327694
+I1203 00:17:32.653018 137274321021824 utils.py:1231] [70250] core_hours = 122.42179768327694
+I1203 00:17:32.653080 137274321021824 train.py:125] NOTE: Steps:70250/112603 [62.4%]
+Walltime:5d2h27m (0s eval)
+ETA:3d1h48m
+Total train time:8d4h14m
+I1203 00:22:44.360984 137274321021824 utils.py:1231] [70300] l2_params = 276.8561646304257
+I1203 00:22:44.361195 137274321021824 utils.py:1231] [70300] train/loss = 4.003956168889999
+I1203 00:22:44.361305 137274321021824 utils.py:1231] [70300] l2_grads = 1.788343906402588
+I1203 00:22:44.361377 137274321021824 utils.py:1231] [70300] lr = 0.0003639889743581306
+I1203 00:22:44.361440 137274321021824 utils.py:1231] [70300] uptime = 441153.723801193
+I1203 00:22:44.361510 137274321021824 utils.py:1231] [70300] examples_seen = 71987200.0
+I1203 00:22:44.361570 137274321021824 utils.py:1231] [70300] progress = 0.6243172917240216
+I1203 00:22:44.361640 137274321021824 utils.py:1231] [70300] epoch = 56.188771643353284
+I1203 00:22:44.361700 137274321021824 utils.py:1231] [70300] img/sec/core = 164.25588385492134
+I1203 00:22:44.361766 137274321021824 utils.py:1231] [70300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.50838345353027
+I1203 00:22:44.361836 137274321021824 utils.py:1231] [70300] core_hours = 122.50838345353027
+I1203 00:22:44.361915 137274321021824 train.py:125] NOTE: Steps:70300/112603 [62.4%]
+Walltime:5d2h32m (0s eval)
+ETA:3d1h43m
+Total train time:8d4h14m
+I1203 00:27:55.994381 137274321021824 utils.py:1231] [70350] l2_params = 276.7663505445407
+I1203 00:27:55.994584 137274321021824 utils.py:1231] [70350] train/loss = 3.5938092172145844
+I1203 00:27:55.994680 137274321021824 utils.py:1231] [70350] l2_grads = 1.657423973083496
+I1203 00:27:55.994740 137274321021824 utils.py:1231] [70350] lr = 0.0003632525263369456
+I1203 00:27:55.994792 137274321021824 utils.py:1231] [70350] uptime = 441465.357154489
+I1203 00:27:55.994845 137274321021824 utils.py:1231] [70350] examples_seen = 72038400.0
+I1203 00:27:55.994899 137274321021824 utils.py:1231] [70350] progress = 0.624761329627097
+I1203 00:27:55.994947 137274321021824 utils.py:1231] [70350] epoch = 56.22873520782224
+I1203 00:27:55.994996 137274321021824 utils.py:1231] [70350] img/sec/core = 164.2956360687542
+I1203 00:27:55.995054 137274321021824 utils.py:1231] [70350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.59494827389027
+I1203 00:27:55.995104 137274321021824 utils.py:1231] [70350] core_hours = 122.59494827389027
+I1203 00:27:55.995164 137274321021824 train.py:125] NOTE: Steps:70350/112603 [62.5%]
+Walltime:5d2h37m (0s eval)
+ETA:3d1h38m
+Total train time:8d4h13m
+I1203 00:33:07.780296 137274321021824 utils.py:1231] [70400] l2_params = 276.67642388331586
+I1203 00:33:07.780543 137274321021824 utils.py:1231] [70400] train/loss = 3.5161123871803284
+I1203 00:33:07.780675 137274321021824 utils.py:1231] [70400] l2_grads = 1.665767788887024
+I1203 00:33:07.780755 137274321021824 utils.py:1231] [70400] lr = 0.00036251639882375187
+I1203 00:33:07.780830 137274321021824 utils.py:1231] [70400] uptime = 441777.143188164
+I1203 00:33:07.780910 137274321021824 utils.py:1231] [70400] examples_seen = 72089600.0
+I1203 00:33:07.780979 137274321021824 utils.py:1231] [70400] progress = 0.6252053675301724
+I1203 00:33:07.781047 137274321021824 utils.py:1231] [70400] epoch = 56.2686987722912
+I1203 00:33:07.781112 137274321021824 utils.py:1231] [70400] img/sec/core = 164.21518115005512
+I1203 00:33:07.781186 137274321021824 utils.py:1231] [70400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.68155550546666
+I1203 00:33:07.781247 137274321021824 utils.py:1231] [70400] core_hours = 122.68155550546666
+I1203 00:33:07.781318 137274321021824 train.py:125] NOTE: Steps:70400/112603 [62.5%]
+Walltime:5d2h42m (0s eval)
+ETA:3d1h32m
+Total train time:8d4h13m
+I1203 00:38:19.576333 137274321021824 utils.py:1231] [70450] l2_params = 276.5953014235667
+I1203 00:38:19.576590 137274321021824 utils.py:1231] [70450] train/loss = 2.0133697539567947
+I1203 00:38:19.576690 137274321021824 utils.py:1231] [70450] l2_grads = 1.8641310930252075
+I1203 00:38:19.576752 137274321021824 utils.py:1231] [70450] lr = 0.0003617805935438813
+I1203 00:38:19.576803 137274321021824 utils.py:1231] [70450] uptime = 442088.93916498596
+I1203 00:38:19.576854 137274321021824 utils.py:1231] [70450] examples_seen = 72140800.0
+I1203 00:38:19.576908 137274321021824 utils.py:1231] [70450] progress = 0.6256494054332478
+I1203 00:38:19.576958 137274321021824 utils.py:1231] [70450] epoch = 56.30866233676016
+I1203 00:38:19.577011 137274321021824 utils.py:1231] [70450] img/sec/core = 164.20994434201353
+I1203 00:38:19.577067 137274321021824 utils.py:1231] [70450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.76816549902831
+I1203 00:38:19.577121 137274321021824 utils.py:1231] [70450] core_hours = 122.76816549902831
+I1203 00:38:19.577181 137274321021824 train.py:125] NOTE: Steps:70450/112603 [62.6%]
+Walltime:5d2h48m (0s eval)
+ETA:3d1h27m
+Total train time:8d4h13m
+I1203 00:43:31.369766 137274321021824 utils.py:1231] [70500] l2_params = 276.499616798563
+I1203 00:43:31.369973 137274321021824 utils.py:1231] [70500] train/loss = 3.3365767300128937
+I1203 00:43:31.370079 137274321021824 utils.py:1231] [70500] l2_grads = 1.8323440551757812
+I1203 00:43:31.370156 137274321021824 utils.py:1231] [70500] lr = 0.00036104511222191004
+I1203 00:43:31.370218 137274321021824 utils.py:1231] [70500] uptime = 442400.732579206
+I1203 00:43:31.370283 137274321021824 utils.py:1231] [70500] examples_seen = 72192000.0
+I1203 00:43:31.370341 137274321021824 utils.py:1231] [70500] progress = 0.6260934433363232
+I1203 00:43:31.370405 137274321021824 utils.py:1231] [70500] epoch = 56.34862590122911
+I1203 00:43:31.370461 137274321021824 utils.py:1231] [70500] img/sec/core = 164.2112939687272
+I1203 00:43:31.370519 137274321021824 utils.py:1231] [70500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.8547747807561
+I1203 00:43:31.370575 137274321021824 utils.py:1231] [70500] core_hours = 122.8547747807561
+I1203 00:43:31.370652 137274321021824 train.py:125] NOTE: Steps:70500/112603 [62.6%]
+Walltime:5d2h53m (0s eval)
+ETA:3d1h22m
+Total train time:8d4h13m
+I1203 00:48:43.166718 137274321021824 utils.py:1231] [70550] l2_params = 276.4101495748053
+I1203 00:48:43.166997 137274321021824 utils.py:1231] [70550] train/loss = 2.436916708946228
+I1203 00:48:43.167191 137274321021824 utils.py:1231] [70550] l2_grads = 1.8382303714752197
+I1203 00:48:43.167310 137274321021824 utils.py:1231] [70550] lr = 0.0003603099565816554
+I1203 00:48:43.167440 137274321021824 utils.py:1231] [70550] uptime = 442712.529790228
+I1203 00:48:43.167519 137274321021824 utils.py:1231] [70550] examples_seen = 72243200.0
+I1203 00:48:43.167596 137274321021824 utils.py:1231] [70550] progress = 0.6265374812393986
+I1203 00:48:43.167698 137274321021824 utils.py:1231] [70550] epoch = 56.38858946569807
+I1203 00:48:43.167777 137274321021824 utils.py:1231] [70550] img/sec/core = 164.20929434287797
+I1203 00:48:43.167874 137274321021824 utils.py:1231] [70550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 122.94138511715111
+I1203 00:48:43.167962 137274321021824 utils.py:1231] [70550] core_hours = 122.94138511715111
+I1203 00:48:43.168054 137274321021824 train.py:125] NOTE: Steps:70550/112603 [62.7%]
+Walltime:5d2h58m (0s eval)
+ETA:3d1h17m
+Total train time:8d4h13m
+I1203 00:53:54.951431 137274321021824 utils.py:1231] [70600] l2_params = 276.3320912349095
+I1203 00:53:54.951670 137274321021824 utils.py:1231] [70600] train/loss = 2.328557848930359
+I1203 00:53:54.951801 137274321021824 utils.py:1231] [70600] l2_grads = 1.740302562713623
+I1203 00:53:54.951907 137274321021824 utils.py:1231] [70600] lr = 0.0003595751283461715
+I1203 00:53:54.951968 137274321021824 utils.py:1231] [70600] uptime = 443024.31432994
+I1203 00:53:54.952030 137274321021824 utils.py:1231] [70600] examples_seen = 72294400.0
+I1203 00:53:54.952086 137274321021824 utils.py:1231] [70600] progress = 0.626981519142474
+I1203 00:53:54.952154 137274321021824 utils.py:1231] [70600] epoch = 56.42855303016703
+I1203 00:53:54.952235 137274321021824 utils.py:1231] [70600] img/sec/core = 164.2159680120522
+I1203 00:53:54.952311 137274321021824 utils.py:1231] [70600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.02799193373777
+I1203 00:53:54.952383 137274321021824 utils.py:1231] [70600] core_hours = 123.02799193373777
+I1203 00:53:54.952466 137274321021824 train.py:125] NOTE: Steps:70600/112603 [62.7%]
+Walltime:5d3h3m (0s eval)
+ETA:3d1h11m
+Total train time:8d4h13m
+I1203 00:59:06.731488 137274321021824 utils.py:1231] [70650] l2_params = 276.2353736969002
+I1203 00:59:06.731682 137274321021824 utils.py:1231] [70650] train/loss = 2.7024313509464264
+I1203 00:59:06.731777 137274321021824 utils.py:1231] [70650] l2_grads = 1.7132759094238281
+I1203 00:59:06.731838 137274321021824 utils.py:1231] [70650] lr = 0.00035884062923774456
+I1203 00:59:06.731898 137274321021824 utils.py:1231] [70650] uptime = 443336.094251189
+I1203 00:59:06.731953 137274321021824 utils.py:1231] [70650] examples_seen = 72345600.0
+I1203 00:59:06.732004 137274321021824 utils.py:1231] [70650] progress = 0.6274255570455494
+I1203 00:59:06.732051 137274321021824 utils.py:1231] [70650] epoch = 56.468516594635986
+I1203 00:59:06.732103 137274321021824 utils.py:1231] [70650] img/sec/core = 164.21840057851497
+I1203 00:59:06.732158 137274321021824 utils.py:1231] [70650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.11459746741804
+I1203 00:59:06.732212 137274321021824 utils.py:1231] [70650] core_hours = 123.11459746741804
+I1203 00:59:06.732274 137274321021824 train.py:125] NOTE: Steps:70650/112603 [62.7%]
+Walltime:5d3h8m (0s eval)
+ETA:3d1h6m
+Total train time:8d4h13m
+I1203 01:04:18.508050 137274321021824 utils.py:1231] [70700] l2_params = 276.14579043103237
+I1203 01:04:18.508264 137274321021824 utils.py:1231] [70700] train/loss = 2.706030458211899
+I1203 01:04:18.508358 137274321021824 utils.py:1231] [70700] l2_grads = 1.606532335281372
+I1203 01:04:18.508423 137274321021824 utils.py:1231] [70700] lr = 0.0003581064609778893
+I1203 01:04:18.508473 137274321021824 utils.py:1231] [70700] uptime = 443647.870835203
+I1203 01:04:18.508525 137274321021824 utils.py:1231] [70700] examples_seen = 72396800.0
+I1203 01:04:18.508574 137274321021824 utils.py:1231] [70700] progress = 0.6278695949486248
+I1203 01:04:18.508627 137274321021824 utils.py:1231] [70700] epoch = 56.50848015910494
+I1203 01:04:18.508701 137274321021824 utils.py:1231] [70700] img/sec/core = 164.22015836089864
+I1203 01:04:18.508758 137274321021824 utils.py:1231] [70700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.2012020740886
+I1203 01:04:18.508807 137274321021824 utils.py:1231] [70700] core_hours = 123.2012020740886
+I1203 01:04:18.508868 137274321021824 train.py:125] NOTE: Steps:70700/112603 [62.8%]
+Walltime:5d3h14m (0s eval)
+ETA:3d1h1m
+Total train time:8d4h13m
+I1203 01:09:30.289540 137274321021824 utils.py:1231] [70750] l2_params = 276.06305633194796
+I1203 01:09:30.289752 137274321021824 utils.py:1231] [70750] train/loss = 2.7169246077537537
+I1203 01:09:30.289843 137274321021824 utils.py:1231] [70750] l2_grads = 1.6786011457443237
+I1203 01:09:30.289908 137274321021824 utils.py:1231] [70750] lr = 0.0003573726252873454
+I1203 01:09:30.289960 137274321021824 utils.py:1231] [70750] uptime = 443959.652322333
+I1203 01:09:30.290012 137274321021824 utils.py:1231] [70750] examples_seen = 72448000.0
+I1203 01:09:30.290060 137274321021824 utils.py:1231] [70750] progress = 0.6283136328517002
+I1203 01:09:30.290110 137274321021824 utils.py:1231] [70750] epoch = 56.5484437235739
+I1203 01:09:30.290162 137274321021824 utils.py:1231] [70750] img/sec/core = 164.21757581342789
+I1203 01:09:30.290216 137274321021824 utils.py:1231] [70750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.28780804273583
+I1203 01:09:30.290269 137274321021824 utils.py:1231] [70750] core_hours = 123.28780804273583
+I1203 01:09:30.290329 137274321021824 train.py:125] NOTE: Steps:70750/112603 [62.8%]
+Walltime:5d3h19m (0s eval)
+ETA:3d0h56m
+Total train time:8d4h13m
+I1203 01:14:41.987155 137274321021824 utils.py:1231] [70800] l2_params = 275.973152867965
+I1203 01:14:41.987370 137274321021824 utils.py:1231] [70800] train/loss = 2.0702345073223114
+I1203 01:14:41.987493 137274321021824 utils.py:1231] [70800] l2_grads = 1.9081093072891235
+I1203 01:14:41.987572 137274321021824 utils.py:1231] [70800] lr = 0.0003566391238860735
+I1203 01:14:41.987642 137274321021824 utils.py:1231] [70800] uptime = 444271.349996277
+I1203 01:14:41.987706 137274321021824 utils.py:1231] [70800] examples_seen = 72499200.0
+I1203 01:14:41.987768 137274321021824 utils.py:1231] [70800] progress = 0.6287576707547756
+I1203 01:14:41.987828 137274321021824 utils.py:1231] [70800] epoch = 56.58840728804285
+I1203 01:14:41.987899 137274321021824 utils.py:1231] [70800] img/sec/core = 164.26173269807174
+I1203 01:14:41.987968 137274321021824 utils.py:1231] [70800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.37439072994249
+I1203 01:14:41.988029 137274321021824 utils.py:1231] [70800] core_hours = 123.37439072994249
+I1203 01:14:41.988099 137274321021824 train.py:125] NOTE: Steps:70800/112603 [62.9%]
+Walltime:5d3h24m (0s eval)
+ETA:3d0h50m
+Total train time:8d4h13m
+I1203 01:19:53.744773 137274321021824 utils.py:1231] [70850] l2_params = 275.88399740121355
+I1203 01:19:53.744981 137274321021824 utils.py:1231] [70850] train/loss = 4.451379060745239
+I1203 01:19:53.745092 137274321021824 utils.py:1231] [70850] l2_grads = 1.8055115938186646
+I1203 01:19:53.745161 137274321021824 utils.py:1231] [70850] lr = 0.0003559059584932502
+I1203 01:19:53.745222 137274321021824 utils.py:1231] [70850] uptime = 444583.107583399
+I1203 01:19:53.745282 137274321021824 utils.py:1231] [70850] examples_seen = 72550400.0
+I1203 01:19:53.745342 137274321021824 utils.py:1231] [70850] progress = 0.6292017086578511
+I1203 01:19:53.745399 137274321021824 utils.py:1231] [70850] epoch = 56.628370852511814
+I1203 01:19:53.745457 137274321021824 utils.py:1231] [70850] img/sec/core = 164.23016508646558
+I1203 01:19:53.745519 137274321021824 utils.py:1231] [70850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.4609900596986
+I1203 01:19:53.745575 137274321021824 utils.py:1231] [70850] core_hours = 123.4609900596986
+I1203 01:19:53.745640 137274321021824 train.py:125] NOTE: Steps:70850/112603 [62.9%]
+Walltime:5d3h29m (0s eval)
+ETA:3d0h45m
+Total train time:8d4h13m
+I1203 01:25:05.517801 137274321021824 utils.py:1231] [70900] l2_params = 275.7930088284429
+I1203 01:25:05.518072 137274321021824 utils.py:1231] [70900] train/loss = 2.09156996011734
+I1203 01:25:05.518204 137274321021824 utils.py:1231] [70900] l2_grads = 2.0459461212158203
+I1203 01:25:05.518298 137274321021824 utils.py:1231] [70900] lr = 0.0003551731308272641
+I1203 01:25:05.518371 137274321021824 utils.py:1231] [70900] uptime = 444894.880733144
+I1203 01:25:05.518450 137274321021824 utils.py:1231] [70900] examples_seen = 72601600.0
+I1203 01:25:05.518515 137274321021824 utils.py:1231] [70900] progress = 0.6296457465609264
+I1203 01:25:05.518575 137274321021824 utils.py:1231] [70900] epoch = 56.66833441698077
+I1203 01:25:05.518639 137274321021824 utils.py:1231] [70900] img/sec/core = 164.22196729216145
+I1203 01:25:05.518718 137274321021824 utils.py:1231] [70900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.54759371240556
+I1203 01:25:05.518778 137274321021824 utils.py:1231] [70900] core_hours = 123.54759371240556
+I1203 01:25:05.518856 137274321021824 train.py:125] NOTE: Steps:70900/112603 [63.0%]
+Walltime:5d3h34m (0s eval)
+ETA:3d0h40m
+Total train time:8d4h13m
+I1203 01:30:17.283730 137274321021824 utils.py:1231] [70950] l2_params = 275.71259557835015
+I1203 01:30:17.283969 137274321021824 utils.py:1231] [70950] train/loss = 3.0745368599891663
+I1203 01:30:17.284105 137274321021824 utils.py:1231] [70950] l2_grads = 1.658502459526062
+I1203 01:30:17.284174 137274321021824 utils.py:1231] [70950] lr = 0.00035444064260571303
+I1203 01:30:17.284236 137274321021824 utils.py:1231] [70950] uptime = 445206.646598312
+I1203 01:30:17.284292 137274321021824 utils.py:1231] [70950] examples_seen = 72652800.0
+I1203 01:30:17.284341 137274321021824 utils.py:1231] [70950] progress = 0.6300897844640019
+I1203 01:30:17.284387 137274321021824 utils.py:1231] [70950] epoch = 56.708297981449725
+I1203 01:30:17.284435 137274321021824 utils.py:1231] [70950] img/sec/core = 164.2258044266917
+I1203 01:30:17.284488 137274321021824 utils.py:1231] [70950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.63419534161889
+I1203 01:30:17.284537 137274321021824 utils.py:1231] [70950] core_hours = 123.63419534161889
+I1203 01:30:17.284593 137274321021824 train.py:125] NOTE: Steps:70950/112603 [63.0%]
+Walltime:5d3h40m (0s eval)
+ETA:3d0h35m
+Total train time:8d4h13m
+I1203 01:35:29.056883 137274321021824 utils.py:1231] [71000] l2_params = 275.6204116736727
+I1203 01:35:29.057091 137274321021824 utils.py:1231] [71000] train/loss = 1.9387506991624832
+I1203 01:35:29.057183 137274321021824 utils.py:1231] [71000] l2_grads = 1.8165186643600464
+I1203 01:35:29.057239 137274321021824 utils.py:1231] [71000] lr = 0.00035370849554539904
+I1203 01:35:29.057288 137274321021824 utils.py:1231] [71000] uptime = 445518.419650694
+I1203 01:35:29.057342 137274321021824 utils.py:1231] [71000] examples_seen = 72704000.0
+I1203 01:35:29.057392 137274321021824 utils.py:1231] [71000] progress = 0.6305338223670772
+I1203 01:35:29.057438 137274321021824 utils.py:1231] [71000] epoch = 56.74826154591868
+I1203 01:35:29.057486 137274321021824 utils.py:1231] [71000] img/sec/core = 164.2220185767323
+I1203 01:35:29.057539 137274321021824 utils.py:1231] [71000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.72079896728056
+I1203 01:35:29.057588 137274321021824 utils.py:1231] [71000] core_hours = 123.72079896728056
+I1203 01:35:29.057645 137274321021824 train.py:125] NOTE: Steps:71000/112603 [63.1%]
+Walltime:5d3h45m (0s eval)
+ETA:3d0h29m
+Total train time:8d4h13m
+I1203 01:40:41.185594 137274321021824 utils.py:1231] [71050] l2_params = 275.53145810783843
+I1203 01:40:41.185849 137274321021824 utils.py:1231] [71050] train/loss = 2.2956627011299133
+I1203 01:40:41.185996 137274321021824 utils.py:1231] [71050] l2_grads = 1.7879538536071777
+I1203 01:40:41.186095 137274321021824 utils.py:1231] [71050] lr = 0.00035297669136232423
+I1203 01:40:41.186180 137274321021824 utils.py:1231] [71050] uptime = 445830.548533733
+I1203 01:40:41.186259 137274321021824 utils.py:1231] [71050] examples_seen = 72755200.0
+I1203 01:40:41.186335 137274321021824 utils.py:1231] [71050] progress = 0.6309778602701527
+I1203 01:40:41.186402 137274321021824 utils.py:1231] [71050] epoch = 56.788225110387636
+I1203 01:40:41.186467 137274321021824 utils.py:1231] [71050] img/sec/core = 164.03480351289113
+I1203 01:40:41.186531 137274321021824 utils.py:1231] [71050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.80750143479139
+I1203 01:40:41.186592 137274321021824 utils.py:1231] [71050] core_hours = 123.80750143479139
+I1203 01:40:41.186666 137274321021824 train.py:125] NOTE: Steps:71050/112603 [63.1%]
+Walltime:5d3h50m (0s eval)
+ETA:3d0h24m
+Total train time:8d4h13m
+I1203 01:45:52.963978 137274321021824 utils.py:1231] [71100] l2_params = 275.44325593056783
+I1203 01:45:52.964191 137274321021824 utils.py:1231] [71100] train/loss = 2.3778382539749146
+I1203 01:45:52.964291 137274321021824 utils.py:1231] [71100] l2_grads = 1.6918702125549316
+I1203 01:45:52.964377 137274321021824 utils.py:1231] [71100] lr = 0.00035224523177168756
+I1203 01:45:52.964450 137274321021824 utils.py:1231] [71100] uptime = 446142.326810549
+I1203 01:45:52.964523 137274321021824 utils.py:1231] [71100] examples_seen = 72806400.0
+I1203 01:45:52.964617 137274321021824 utils.py:1231] [71100] progress = 0.631421898173228
+I1203 01:45:52.964697 137274321021824 utils.py:1231] [71100] epoch = 56.8281886748566
+I1203 01:45:52.964775 137274321021824 utils.py:1231] [71100] img/sec/core = 164.21926672657386
+I1203 01:45:52.964872 137274321021824 utils.py:1231] [71100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.89410651168471
+I1203 01:45:52.964950 137274321021824 utils.py:1231] [71100] core_hours = 123.89410651168471
+I1203 01:45:52.965026 137274321021824 train.py:125] NOTE: Steps:71100/112603 [63.1%]
+Walltime:5d3h55m (0s eval)
+ETA:3d0h19m
+Total train time:8d4h13m
+I1203 01:51:04.754008 137274321021824 utils.py:1231] [71150] l2_params = 275.35622909819085
+I1203 01:51:04.754212 137274321021824 utils.py:1231] [71150] train/loss = 3.0947189331054688
+I1203 01:51:04.754314 137274321021824 utils.py:1231] [71150] l2_grads = 1.6844751834869385
+I1203 01:51:04.754383 137274321021824 utils.py:1231] [71150] lr = 0.0003515141184878798
+I1203 01:51:04.754451 137274321021824 utils.py:1231] [71150] uptime = 446454.116813128
+I1203 01:51:04.754513 137274321021824 utils.py:1231] [71150] examples_seen = 72857600.0
+I1203 01:51:04.754569 137274321021824 utils.py:1231] [71150] progress = 0.6318659360763035
+I1203 01:51:04.754623 137274321021824 utils.py:1231] [71150] epoch = 56.86815223932555
+I1203 01:51:04.754680 137274321021824 utils.py:1231] [71150] img/sec/core = 164.2130907870578
+I1203 01:51:04.754740 137274321021824 utils.py:1231] [71150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 123.98071484573444
+I1203 01:51:04.754794 137274321021824 utils.py:1231] [71150] core_hours = 123.98071484573444
+I1203 01:51:04.754858 137274321021824 train.py:125] NOTE: Steps:71150/112603 [63.2%]
+Walltime:5d4h0m (0s eval)
+ETA:3d0h14m
+Total train time:8d4h13m
+I1203 01:56:16.539570 137274321021824 utils.py:1231] [71200] l2_params = 275.2659234921581
+I1203 01:56:16.539773 137274321021824 utils.py:1231] [71200] train/loss = 2.9763376712799072
+I1203 01:56:16.539874 137274321021824 utils.py:1231] [71200] l2_grads = 1.728937029838562
+I1203 01:56:16.539940 137274321021824 utils.py:1231] [71200] lr = 0.00035078335322448007
+I1203 01:56:16.539992 137274321021824 utils.py:1231] [71200] uptime = 446765.90235456696
+I1203 01:56:16.540046 137274321021824 utils.py:1231] [71200] examples_seen = 72908800.0
+I1203 01:56:16.540098 137274321021824 utils.py:1231] [71200] progress = 0.6323099739793788
+I1203 01:56:16.540148 137274321021824 utils.py:1231] [71200] epoch = 56.90811580379451
+I1203 01:56:16.540200 137274321021824 utils.py:1231] [71200] img/sec/core = 164.2154404072107
+I1203 01:56:16.540255 137274321021824 utils.py:1231] [71200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.0673219405786
+I1203 01:56:16.540307 137274321021824 utils.py:1231] [71200] core_hours = 124.0673219405786
+I1203 01:56:16.540368 137274321021824 train.py:125] NOTE: Steps:71200/112603 [63.2%]
+Walltime:5d4h6m (0s eval)
+ETA:3d0h8m
+Total train time:8d4h13m
+I1203 02:01:28.323029 137274321021824 utils.py:1231] [71250] l2_params = 275.1671402708912
+I1203 02:01:28.323324 137274321021824 utils.py:1231] [71250] train/loss = 2.1066305339336395
+I1203 02:01:28.323436 137274321021824 utils.py:1231] [71250] l2_grads = 1.899750828742981
+I1203 02:01:28.323522 137274321021824 utils.py:1231] [71250] lr = 0.0003500529376942527
+I1203 02:01:28.323592 137274321021824 utils.py:1231] [71250] uptime = 447077.685951954
+I1203 02:01:28.323665 137274321021824 utils.py:1231] [71250] examples_seen = 72960000.0
+I1203 02:01:28.323725 137274321021824 utils.py:1231] [71250] progress = 0.6327540118824543
+I1203 02:01:28.323780 137274321021824 utils.py:1231] [71250] epoch = 56.948079368263464
+I1203 02:01:28.323832 137274321021824 utils.py:1231] [71250] img/sec/core = 164.21646433325438
+I1203 02:01:28.323904 137274321021824 utils.py:1231] [71250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.15392849540832
+I1203 02:01:28.323957 137274321021824 utils.py:1231] [71250] core_hours = 124.15392849540832
+I1203 02:01:28.324023 137274321021824 train.py:125] NOTE: Steps:71250/112603 [63.3%]
+Walltime:5d4h11m (0s eval)
+ETA:3d0h3m
+Total train time:8d4h13m
+I1203 02:06:40.101840 137274321021824 utils.py:1231] [71300] l2_params = 275.07728905825695
+I1203 02:06:40.102093 137274321021824 utils.py:1231] [71300] train/loss = 2.2960857450962067
+I1203 02:06:40.102245 137274321021824 utils.py:1231] [71300] l2_grads = 1.78061044216156
+I1203 02:06:40.102338 137274321021824 utils.py:1231] [71300] lr = 0.00034932287360914135
+I1203 02:06:40.102405 137274321021824 utils.py:1231] [71300] uptime = 447389.46476719796
+I1203 02:06:40.102479 137274321021824 utils.py:1231] [71300] examples_seen = 73011200.0
+I1203 02:06:40.102543 137274321021824 utils.py:1231] [71300] progress = 0.6331980497855297
+I1203 02:06:40.102604 137274321021824 utils.py:1231] [71300] epoch = 56.98804293273242
+I1203 02:06:40.102679 137274321021824 utils.py:1231] [71300] img/sec/core = 164.21898312730013
+I1203 02:06:40.102744 137274321021824 utils.py:1231] [71300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.24053372186498
+I1203 02:06:40.102796 137274321021824 utils.py:1231] [71300] core_hours = 124.24053372186498
+I1203 02:06:40.102860 137274321021824 train.py:125] NOTE: Steps:71300/112603 [63.3%]
+Walltime:5d4h16m (0s eval)
+ETA:2d23h58m
+Total train time:8d4h13m
+I1203 02:11:51.904582 137274321021824 utils.py:1231] [71350] l2_params = 274.9839526361299
+I1203 02:11:51.904817 137274321021824 utils.py:1231] [71350] train/loss = 1.948280319571495
+I1203 02:11:51.904951 137274321021824 utils.py:1231] [71350] l2_grads = 1.8744322061538696
+I1203 02:11:51.905043 137274321021824 utils.py:1231] [71350] lr = 0.0003485931626802663
+I1203 02:11:51.905113 137274321021824 utils.py:1231] [71350] uptime = 447701.267475512
+I1203 02:11:51.905188 137274321021824 utils.py:1231] [71350] examples_seen = 73062400.0
+I1203 02:11:51.905260 137274321021824 utils.py:1231] [71350] progress = 0.6336420876886051
+I1203 02:11:51.905327 137274321021824 utils.py:1231] [71350] epoch = 57.02800649720138
+I1203 02:11:51.905400 137274321021824 utils.py:1231] [71350] img/sec/core = 164.20639922227275
+I1203 02:11:51.905471 137274321021824 utils.py:1231] [71350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.32714558528555
+I1203 02:11:51.905528 137274321021824 utils.py:1231] [71350] core_hours = 124.32714558528555
+I1203 02:11:51.905592 137274321021824 train.py:125] NOTE: Steps:71350/112603 [63.4%]
+Walltime:5d4h21m (0s eval)
+ETA:2d23h53m
+Total train time:8d4h12m
+I1203 02:17:03.678757 137274321021824 utils.py:1231] [71400] l2_params = 274.8912359253758
+I1203 02:17:03.679015 137274321021824 utils.py:1231] [71400] train/loss = 2.119312286376953
+I1203 02:17:03.679118 137274321021824 utils.py:1231] [71400] l2_grads = 1.9992361068725586
+I1203 02:17:03.679204 137274321021824 utils.py:1231] [71400] lr = 0.00034786380661792
+I1203 02:17:03.679309 137274321021824 utils.py:1231] [71400] uptime = 448013.041664553
+I1203 02:17:03.679394 137274321021824 utils.py:1231] [71400] examples_seen = 73113600.0
+I1203 02:17:03.679470 137274321021824 utils.py:1231] [71400] progress = 0.6340861255916805
+I1203 02:17:03.679550 137274321021824 utils.py:1231] [71400] epoch = 57.06797006167034
+I1203 02:17:03.679628 137274321021824 utils.py:1231] [71400] img/sec/core = 164.22141985996012
+I1203 02:17:03.679713 137274321021824 utils.py:1231] [71400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.41374952668583
+I1203 02:17:03.679798 137274321021824 utils.py:1231] [71400] core_hours = 124.41374952668583
+I1203 02:17:03.679893 137274321021824 train.py:125] NOTE: Steps:71400/112603 [63.4%]
+Walltime:5d4h26m (0s eval)
+ETA:2d23h47m
+Total train time:8d4h12m
+I1203 02:22:15.461280 137274321021824 utils.py:1231] [71450] l2_params = 274.80630489315325
+I1203 02:22:15.461534 137274321021824 utils.py:1231] [71450] train/loss = 1.842520073056221
+I1203 02:22:15.461663 137274321021824 utils.py:1231] [71450] l2_grads = 1.9257475137710571
+I1203 02:22:15.461749 137274321021824 utils.py:1231] [71450] lr = 0.00034713480713156333
+I1203 02:22:15.461807 137274321021824 utils.py:1231] [71450] uptime = 448324.824168891
+I1203 02:22:15.461875 137274321021824 utils.py:1231] [71450] examples_seen = 73164800.0
+I1203 02:22:15.461946 137274321021824 utils.py:1231] [71450] progress = 0.6345301634947559
+I1203 02:22:15.462014 137274321021824 utils.py:1231] [71450] epoch = 57.10793362613929
+I1203 02:22:15.462083 137274321021824 utils.py:1231] [71450] img/sec/core = 164.21704004433897
+I1203 02:22:15.462155 137274321021824 utils.py:1231] [71450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.50035577789085
+I1203 02:22:15.462217 137274321021824 utils.py:1231] [71450] core_hours = 124.50035577789085
+I1203 02:22:15.462299 137274321021824 train.py:125] NOTE: Steps:71450/112603 [63.5%]
+Walltime:5d4h32m (0s eval)
+ETA:2d23h42m
+Total train time:8d4h12m
+I1203 02:27:27.223304 137274321021824 utils.py:1231] [71500] l2_params = 274.71933662038185
+I1203 02:27:27.223517 137274321021824 utils.py:1231] [71500] train/loss = 2.1342491507530212
+I1203 02:27:27.223648 137274321021824 utils.py:1231] [71500] l2_grads = 1.8806462287902832
+I1203 02:27:27.223721 137274321021824 utils.py:1231] [71500] lr = 0.00034640616592982154
+I1203 02:27:27.223778 137274321021824 utils.py:1231] [71500] uptime = 448636.586139482
+I1203 02:27:27.223835 137274321021824 utils.py:1231] [71500] examples_seen = 73216000.0
+I1203 02:27:27.223896 137274321021824 utils.py:1231] [71500] progress = 0.6349742013978313
+I1203 02:27:27.223953 137274321021824 utils.py:1231] [71500] epoch = 57.14789719060825
+I1203 02:27:27.224008 137274321021824 utils.py:1231] [71500] img/sec/core = 164.22785595993474
+I1203 02:27:27.224068 137274321021824 utils.py:1231] [71500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.58695632527721
+I1203 02:27:27.224125 137274321021824 utils.py:1231] [71500] core_hours = 124.58695632527721
+I1203 02:27:27.224189 137274321021824 train.py:125] NOTE: Steps:71500/112603 [63.5%]
+Walltime:5d4h37m (0s eval)
+ETA:2d23h37m
+Total train time:8d4h12m
+I1203 02:32:38.994214 137274321021824 utils.py:1231] [71550] l2_params = 274.6360152602598
+I1203 02:32:38.994408 137274321021824 utils.py:1231] [71550] train/loss = 3.3533015847206116
+I1203 02:32:38.994510 137274321021824 utils.py:1231] [71550] l2_grads = 1.7298450469970703
+I1203 02:32:38.994576 137274321021824 utils.py:1231] [71550] lr = 0.00034567788472047957
+I1203 02:32:38.994633 137274321021824 utils.py:1231] [71550] uptime = 448948.35699545
+I1203 02:32:38.994710 137274321021824 utils.py:1231] [71550] examples_seen = 73267200.0
+I1203 02:32:38.994765 137274321021824 utils.py:1231] [71550] progress = 0.6354182393009067
+I1203 02:32:38.994819 137274321021824 utils.py:1231] [71550] epoch = 57.18786075507721
+I1203 02:32:38.994895 137274321021824 utils.py:1231] [71550] img/sec/core = 164.22317551469476
+I1203 02:32:38.994963 137274321021824 utils.py:1231] [71550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.67355934082389
+I1203 02:32:38.995018 137274321021824 utils.py:1231] [71550] core_hours = 124.67355934082389
+I1203 02:32:38.995100 137274321021824 train.py:125] NOTE: Steps:71550/112603 [63.5%]
+Walltime:5d4h42m (0s eval)
+ETA:2d23h32m
+Total train time:8d4h12m
+I1203 02:37:50.763479 137274321021824 utils.py:1231] [71600] l2_params = 274.54492165702817
+I1203 02:37:50.763700 137274321021824 utils.py:1231] [71600] train/loss = 1.9837283343076706
+I1203 02:37:50.763831 137274321021824 utils.py:1231] [71600] l2_grads = 1.8936753273010254
+I1203 02:37:50.763922 137274321021824 utils.py:1231] [71600] lr = 0.00034494996521047946
+I1203 02:37:50.764003 137274321021824 utils.py:1231] [71600] uptime = 449260.126360041
+I1203 02:37:50.764083 137274321021824 utils.py:1231] [71600] examples_seen = 73318400.0
+I1203 02:37:50.764154 137274321021824 utils.py:1231] [71600] progress = 0.6358622772039821
+I1203 02:37:50.764233 137274321021824 utils.py:1231] [71600] epoch = 57.227824319546166
+I1203 02:37:50.764307 137274321021824 utils.py:1231] [71600] img/sec/core = 164.22396109113973
+I1203 02:37:50.764375 137274321021824 utils.py:1231] [71600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.76016194209916
+I1203 02:37:50.764438 137274321021824 utils.py:1231] [71600] core_hours = 124.76016194209916
+I1203 02:37:50.764541 137274321021824 train.py:125] NOTE: Steps:71600/112603 [63.6%]
+Walltime:5d4h47m (0s eval)
+ETA:2d23h26m
+Total train time:8d4h12m
+I1203 02:43:02.530254 137274321021824 utils.py:1231] [71650] l2_params = 274.4563757070029
+I1203 02:43:02.530493 137274321021824 utils.py:1231] [71650] train/loss = 4.356330931186676
+I1203 02:43:02.530630 137274321021824 utils.py:1231] [71650] l2_grads = 1.7741761207580566
+I1203 02:43:02.530726 137274321021824 utils.py:1231] [71650] lr = 0.0003442224091059148
+I1203 02:43:02.530814 137274321021824 utils.py:1231] [71650] uptime = 449571.893170869
+I1203 02:43:02.530897 137274321021824 utils.py:1231] [71650] examples_seen = 73369600.0
+I1203 02:43:02.530978 137274321021824 utils.py:1231] [71650] progress = 0.6363063151070575
+I1203 02:43:02.531046 137274321021824 utils.py:1231] [71650] epoch = 57.26778788401512
+I1203 02:43:02.531114 137274321021824 utils.py:1231] [71650] img/sec/core = 164.22530629229144
+I1203 02:43:02.531189 137274321021824 utils.py:1231] [71650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.84676383399582
+I1203 02:43:02.531254 137274321021824 utils.py:1231] [71650] core_hours = 124.84676383399582
+I1203 02:43:02.531328 137274321021824 train.py:125] NOTE: Steps:71650/112603 [63.6%]
+Walltime:5d4h52m (0s eval)
+ETA:2d23h21m
+Total train time:8d4h12m
+I1203 02:48:14.292774 137274321021824 utils.py:1231] [71700] l2_params = 274.37060206816227
+I1203 02:48:14.293039 137274321021824 utils.py:1231] [71700] train/loss = 3.6216775476932526
+I1203 02:48:14.293181 137274321021824 utils.py:1231] [71700] l2_grads = 1.8463871479034424
+I1203 02:48:14.293272 137274321021824 utils.py:1231] [71700] lr = 0.00034349521811202755
+I1203 02:48:14.293341 137274321021824 utils.py:1231] [71700] uptime = 449883.655702171
+I1203 02:48:14.293410 137274321021824 utils.py:1231] [71700] examples_seen = 73420800.0
+I1203 02:48:14.293478 137274321021824 utils.py:1231] [71700] progress = 0.6367503530101329
+I1203 02:48:14.293533 137274321021824 utils.py:1231] [71700] epoch = 57.307751448484076
+I1203 02:48:14.293607 137274321021824 utils.py:1231] [71700] img/sec/core = 164.22756059291777
+I1203 02:48:14.293685 137274321021824 utils.py:1231] [71700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 124.93336453713526
+I1203 02:48:14.293751 137274321021824 utils.py:1231] [71700] core_hours = 124.93336453713526
+I1203 02:48:14.293836 137274321021824 train.py:125] NOTE: Steps:71700/112603 [63.7%]
+Walltime:5d4h58m (0s eval)
+ETA:2d23h16m
+Total train time:8d4h12m
+I1203 02:53:26.063357 137274321021824 utils.py:1231] [71750] l2_params = 274.28862748117945
+I1203 02:53:26.063559 137274321021824 utils.py:1231] [71750] train/loss = 2.0747386515140533
+I1203 02:53:26.063657 137274321021824 utils.py:1231] [71750] l2_grads = 1.9303414821624756
+I1203 02:53:26.063720 137274321021824 utils.py:1231] [71750] lr = 0.00034276839393320406
+I1203 02:53:26.063774 137274321021824 utils.py:1231] [71750] uptime = 450195.426136403
+I1203 02:53:26.063833 137274321021824 utils.py:1231] [71750] examples_seen = 73472000.0
+I1203 02:53:26.063889 137274321021824 utils.py:1231] [71750] progress = 0.6371943909132084
+I1203 02:53:26.063941 137274321021824 utils.py:1231] [71750] epoch = 57.34771501295303
+I1203 02:53:26.063994 137274321021824 utils.py:1231] [71750] img/sec/core = 164.22339766155213
+I1203 02:53:26.064053 137274321021824 utils.py:1231] [71750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.01996743553305
+I1203 02:53:26.064106 137274321021824 utils.py:1231] [71750] core_hours = 125.01996743553305
+I1203 02:53:26.064168 137274321021824 train.py:125] NOTE: Steps:71750/112603 [63.7%]
+Walltime:5d5h3m (0s eval)
+ETA:2d23h11m
+Total train time:8d4h12m
+I1203 02:58:37.821265 137274321021824 utils.py:1231] [71800] l2_params = 274.1969133293336
+I1203 02:58:37.821510 137274321021824 utils.py:1231] [71800] train/loss = 4.149212658405304
+I1203 02:58:37.821623 137274321021824 utils.py:1231] [71800] l2_grads = 1.9678057432174683
+I1203 02:58:37.821684 137274321021824 utils.py:1231] [71800] lr = 0.00034204193827297094
+I1203 02:58:37.821741 137274321021824 utils.py:1231] [71800] uptime = 450507.184102995
+I1203 02:58:37.821794 137274321021824 utils.py:1231] [71800] examples_seen = 73523200.0
+I1203 02:58:37.821843 137274321021824 utils.py:1231] [71800] progress = 0.6376384288162837
+I1203 02:58:37.821897 137274321021824 utils.py:1231] [71800] epoch = 57.387678577421994
+I1203 02:58:37.821947 137274321021824 utils.py:1231] [71800] img/sec/core = 164.229965186454
+I1203 02:58:37.822002 137274321021824 utils.py:1231] [71800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.10656687069748
+I1203 02:58:37.822052 137274321021824 utils.py:1231] [71800] core_hours = 125.10656687069748
+I1203 02:58:37.822113 137274321021824 train.py:125] NOTE: Steps:71800/112603 [63.8%]
+Walltime:5d5h8m (0s eval)
+ETA:2d23h5m
+Total train time:8d4h12m
+I1203 03:03:49.582472 137274321021824 utils.py:1231] [71850] l2_params = 274.11108022210624
+I1203 03:03:49.582705 137274321021824 utils.py:1231] [71850] train/loss = 2.0889170318841934
+I1203 03:03:49.582828 137274321021824 utils.py:1231] [71850] l2_grads = 1.9216371774673462
+I1203 03:03:49.582910 137274321021824 utils.py:1231] [71850] lr = 0.000341315852833991
+I1203 03:03:49.582966 137274321021824 utils.py:1231] [71850] uptime = 450818.945328684
+I1203 03:03:49.583017 137274321021824 utils.py:1231] [71850] examples_seen = 73574400.0
+I1203 03:03:49.583065 137274321021824 utils.py:1231] [71850] progress = 0.6380824667193592
+I1203 03:03:49.583111 137274321021824 utils.py:1231] [71850] epoch = 57.42764214189095
+I1203 03:03:49.583164 137274321021824 utils.py:1231] [71850] img/sec/core = 164.2282483552666
+I1203 03:03:49.583220 137274321021824 utils.py:1231] [71850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.19316721116667
+I1203 03:03:49.583271 137274321021824 utils.py:1231] [71850] core_hours = 125.19316721116667
+I1203 03:03:49.583334 137274321021824 train.py:125] NOTE: Steps:71850/112603 [63.8%]
+Walltime:5d5h13m (0s eval)
+ETA:2d23h0m
+Total train time:8d4h12m
+I1203 03:09:01.349916 137274321021824 utils.py:1231] [71900] l2_params = 274.0235195296281
+I1203 03:09:01.350123 137274321021824 utils.py:1231] [71900] train/loss = 2.0368787348270416
+I1203 03:09:01.350225 137274321021824 utils.py:1231] [71900] l2_grads = 1.8580023050308228
+I1203 03:09:01.350301 137274321021824 utils.py:1231] [71900] lr = 0.00034059013931805974
+I1203 03:09:01.350360 137274321021824 utils.py:1231] [71900] uptime = 451130.712722302
+I1203 03:09:01.350419 137274321021824 utils.py:1231] [71900] examples_seen = 73625600.0
+I1203 03:09:01.350476 137274321021824 utils.py:1231] [71900] progress = 0.6385265046224345
+I1203 03:09:01.350532 137274321021824 utils.py:1231] [71900] epoch = 57.467605706359905
+I1203 03:09:01.350588 137274321021824 utils.py:1231] [71900] img/sec/core = 164.22499930423376
+I1203 03:09:01.350650 137274321021824 utils.py:1231] [71900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.27976926494944
+I1203 03:09:01.350705 137274321021824 utils.py:1231] [71900] core_hours = 125.27976926494944
+I1203 03:09:01.350766 137274321021824 train.py:125] NOTE: Steps:71900/112603 [63.9%]
+Walltime:5d5h18m (0s eval)
+ETA:2d22h55m
+Total train time:8d4h12m
+I1203 03:14:13.121306 137274321021824 utils.py:1231] [71950] l2_params = 273.93894303871065
+I1203 03:14:13.121546 137274321021824 utils.py:1231] [71950] train/loss = 2.5975866317749023
+I1203 03:14:13.121648 137274321021824 utils.py:1231] [71950] l2_grads = 1.7338882684707642
+I1203 03:14:13.121718 137274321021824 utils.py:1231] [71950] lr = 0.00033986479942610013
+I1203 03:14:13.121783 137274321021824 utils.py:1231] [71950] uptime = 451442.484142389
+I1203 03:14:13.121851 137274321021824 utils.py:1231] [71950] examples_seen = 73676800.0
+I1203 03:14:13.121930 137274321021824 utils.py:1231] [71950] progress = 0.63897054252551
+I1203 03:14:13.122024 137274321021824 utils.py:1231] [71950] epoch = 57.50756927082886
+I1203 03:14:13.122083 137274321021824 utils.py:1231] [71950] img/sec/core = 164.22287836940941
+I1203 03:14:13.122143 137274321021824 utils.py:1231] [71950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.36637243719582
+I1203 03:14:13.122201 137274321021824 utils.py:1231] [71950] core_hours = 125.36637243719582
+I1203 03:14:13.122264 137274321021824 train.py:125] NOTE: Steps:71950/112603 [63.9%]
+Walltime:5d5h24m (0s eval)
+ETA:2d22h50m
+Total train time:8d4h12m
+I1203 03:19:24.893684 137274321021824 utils.py:1231] [72000] l2_params = 273.8463383531293
+I1203 03:19:24.893893 137274321021824 utils.py:1231] [72000] train/loss = 2.2327124178409576
+I1203 03:19:24.893994 137274321021824 utils.py:1231] [72000] l2_grads = 1.900468349456787
+I1203 03:19:24.894061 137274321021824 utils.py:1231] [72000] lr = 0.00033913983485816054
+I1203 03:19:24.894125 137274321021824 utils.py:1231] [72000] uptime = 451754.25648711197
+I1203 03:19:24.894182 137274321021824 utils.py:1231] [72000] examples_seen = 73728000.0
+I1203 03:19:24.894237 137274321021824 utils.py:1231] [72000] progress = 0.6394145804285853
+I1203 03:19:24.894290 137274321021824 utils.py:1231] [72000] epoch = 57.547532835297815
+I1203 03:19:24.894347 137274321021824 utils.py:1231] [72000] img/sec/core = 164.22239132688466
+I1203 03:19:24.894405 137274321021824 utils.py:1231] [72000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.45297586628554
+I1203 03:19:24.894458 137274321021824 utils.py:1231] [72000] core_hours = 125.45297586628554
+I1203 03:19:24.894522 137274321021824 train.py:125] NOTE: Steps:72000/112603 [63.9%]
+Walltime:5d5h29m (0s eval)
+ETA:2d22h44m
+Total train time:8d4h12m
+I1203 03:24:37.022963 137274321021824 utils.py:1231] [72050] l2_params = 273.75746372991364
+I1203 03:24:37.023175 137274321021824 utils.py:1231] [72050] train/loss = 1.995011880993843
+I1203 03:24:37.023292 137274321021824 utils.py:1231] [72050] l2_grads = 1.972498893737793
+I1203 03:24:37.023357 137274321021824 utils.py:1231] [72050] lr = 0.000338415247313408
+I1203 03:24:37.023414 137274321021824 utils.py:1231] [72050] uptime = 452066.385776066
+I1203 03:24:37.023482 137274321021824 utils.py:1231] [72050] examples_seen = 73779200.0
+I1203 03:24:37.023539 137274321021824 utils.py:1231] [72050] progress = 0.6398586183316608
+I1203 03:24:37.023585 137274321021824 utils.py:1231] [72050] epoch = 57.58749639976678
+I1203 03:24:37.023634 137274321021824 utils.py:1231] [72050] img/sec/core = 164.03459019042046
+I1203 03:24:37.023688 137274321021824 utils.py:1231] [72050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.53967844655055
+I1203 03:24:37.023737 137274321021824 utils.py:1231] [72050] core_hours = 125.53967844655055
+I1203 03:24:37.023794 137274321021824 train.py:125] NOTE: Steps:72050/112603 [64.0%]
+Walltime:5d5h34m (0s eval)
+ETA:2d22h39m
+Total train time:8d4h12m
+I1203 03:29:48.789824 137274321021824 utils.py:1231] [72100] l2_params = 273.66359562885606
+I1203 03:29:48.790099 137274321021824 utils.py:1231] [72100] train/loss = 2.5788766145706177
+I1203 03:29:48.790224 137274321021824 utils.py:1231] [72100] l2_grads = 1.6770296096801758
+I1203 03:29:48.790295 137274321021824 utils.py:1231] [72100] lr = 0.0003376910384901277
+I1203 03:29:48.790356 137274321021824 utils.py:1231] [72100] uptime = 452378.152713117
+I1203 03:29:48.790420 137274321021824 utils.py:1231] [72100] examples_seen = 73830400.0
+I1203 03:29:48.790472 137274321021824 utils.py:1231] [72100] progress = 0.6403026562347361
+I1203 03:29:48.790524 137274321021824 utils.py:1231] [72100] epoch = 57.62745996423573
+I1203 03:29:48.790575 137274321021824 utils.py:1231] [72100] img/sec/core = 164.22523980347583
+I1203 03:29:48.790630 137274321021824 utils.py:1231] [72100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.62628037350918
+I1203 03:29:48.790686 137274321021824 utils.py:1231] [72100] core_hours = 125.62628037350918
+I1203 03:29:48.790747 137274321021824 train.py:125] NOTE: Steps:72100/112603 [64.0%]
+Walltime:5d5h39m (0s eval)
+ETA:2d22h34m
+Total train time:8d4h12m
+I1203 03:35:00.732450 137274321021824 utils.py:1231] [72150] l2_params = 273.571019922966
+I1203 03:35:00.732728 137274321021824 utils.py:1231] [72150] train/loss = 3.182625412940979
+I1203 03:35:00.732858 137274321021824 utils.py:1231] [72150] l2_grads = 1.6575922966003418
+I1203 03:35:00.732937 137274321021824 utils.py:1231] [72150] lr = 0.00033696721008571627
+I1203 03:35:00.732988 137274321021824 utils.py:1231] [72150] uptime = 452690.095350202
+I1203 03:35:00.733040 137274321021824 utils.py:1231] [72150] examples_seen = 73881600.0
+I1203 03:35:00.733090 137274321021824 utils.py:1231] [72150] progress = 0.6407466941378116
+I1203 03:35:00.733138 137274321021824 utils.py:1231] [72150] epoch = 57.66742352870469
+I1203 03:35:00.733189 137274321021824 utils.py:1231] [72150] img/sec/core = 164.13274080917466
+I1203 03:35:00.733246 137274321021824 utils.py:1231] [72150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.71293110603277
+I1203 03:35:00.733299 137274321021824 utils.py:1231] [72150] core_hours = 125.71293110603277
+I1203 03:35:00.733360 137274321021824 train.py:125] NOTE: Steps:72150/112603 [64.1%]
+Walltime:5d5h44m (0s eval)
+ETA:2d22h29m
+Total train time:8d4h12m
+I1203 03:40:12.509551 137274321021824 utils.py:1231] [72200] l2_params = 273.49013471986717
+I1203 03:40:12.509832 137274321021824 utils.py:1231] [72200] train/loss = 3.652677893638611
+I1203 03:40:12.510027 137274321021824 utils.py:1231] [72200] l2_grads = 1.7638413906097412
+I1203 03:40:12.510124 137274321021824 utils.py:1231] [72200] lr = 0.00033624376379667864
+I1203 03:40:12.510206 137274321021824 utils.py:1231] [72200] uptime = 453001.87256437
+I1203 03:40:12.510284 137274321021824 utils.py:1231] [72200] examples_seen = 73932800.0
+I1203 03:40:12.510373 137274321021824 utils.py:1231] [72200] progress = 0.6411907320408871
+I1203 03:40:12.510446 137274321021824 utils.py:1231] [72200] epoch = 57.707387093173644
+I1203 03:40:12.510513 137274321021824 utils.py:1231] [72200] img/sec/core = 164.219826444446
+I1203 03:40:12.510597 137274321021824 utils.py:1231] [72200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.7995358877461
+I1203 03:40:12.510673 137274321021824 utils.py:1231] [72200] core_hours = 125.7995358877461
+I1203 03:40:12.510738 137274321021824 train.py:125] NOTE: Steps:72200/112603 [64.1%]
+Walltime:5d5h50m (0s eval)
+ETA:2d22h23m
+Total train time:8d4h12m
+I1203 03:45:24.279632 137274321021824 utils.py:1231] [72250] l2_params = 273.39836956229897
+I1203 03:45:24.279875 137274321021824 utils.py:1231] [72250] train/loss = 2.1997048407793045
+I1203 03:45:24.280017 137274321021824 utils.py:1231] [72250] l2_grads = 1.9394562244415283
+I1203 03:45:24.280104 137274321021824 utils.py:1231] [72250] lr = 0.0003355207013186243
+I1203 03:45:24.280164 137274321021824 utils.py:1231] [72250] uptime = 453313.64252577897
+I1203 03:45:24.280237 137274321021824 utils.py:1231] [72250] examples_seen = 73984000.0
+I1203 03:45:24.280297 137274321021824 utils.py:1231] [72250] progress = 0.6416347699439624
+I1203 03:45:24.280367 137274321021824 utils.py:1231] [72250] epoch = 57.7473506576426
+I1203 03:45:24.280431 137274321021824 utils.py:1231] [72250] img/sec/core = 164.22364671892126
+I1203 03:45:24.280499 137274321021824 utils.py:1231] [72250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.88613865480416
+I1203 03:45:24.280563 137274321021824 utils.py:1231] [72250] core_hours = 125.88613865480416
+I1203 03:45:24.280646 137274321021824 train.py:125] NOTE: Steps:72250/112603 [64.2%]
+Walltime:5d5h55m (0s eval)
+ETA:2d22h18m
+Total train time:8d4h12m
+I1203 03:50:36.056839 137274321021824 utils.py:1231] [72300] l2_params = 273.3158303015815
+I1203 03:50:36.057068 137274321021824 utils.py:1231] [72300] train/loss = 2.155095413327217
+I1203 03:50:36.057224 137274321021824 utils.py:1231] [72300] l2_grads = 2.073925256729126
+I1203 03:50:36.057294 137274321021824 utils.py:1231] [72300] lr = 0.0003347980243462632
+I1203 03:50:36.057345 137274321021824 utils.py:1231] [72300] uptime = 453625.41970733297
+I1203 03:50:36.057396 137274321021824 utils.py:1231] [72300] examples_seen = 74035200.0
+I1203 03:50:36.057445 137274321021824 utils.py:1231] [72300] progress = 0.6420788078470379
+I1203 03:50:36.057498 137274321021824 utils.py:1231] [72300] epoch = 57.78731422211156
+I1203 03:50:36.057555 137274321021824 utils.py:1231] [72300] img/sec/core = 164.21984362294177
+I1203 03:50:36.057611 137274321021824 utils.py:1231] [72300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 125.97274342745804
+I1203 03:50:36.057663 137274321021824 utils.py:1231] [72300] core_hours = 125.97274342745804
+I1203 03:50:36.057722 137274321021824 train.py:125] NOTE: Steps:72300/112603 [64.2%]
+Walltime:5d6h0m (0s eval)
+ETA:2d22h13m
+Total train time:8d4h12m
+I1203 03:55:47.831205 137274321021824 utils.py:1231] [72350] l2_params = 273.2159549454979
+I1203 03:55:47.831393 137274321021824 utils.py:1231] [72350] train/loss = 1.9790352433919907
+I1203 03:55:47.831481 137274321021824 utils.py:1231] [72350] l2_grads = 1.877946376800537
+I1203 03:55:47.831537 137274321021824 utils.py:1231] [72350] lr = 0.0003340757345734019
+I1203 03:55:47.831587 137274321021824 utils.py:1231] [72350] uptime = 453937.193950045
+I1203 03:55:47.831636 137274321021824 utils.py:1231] [72350] examples_seen = 74086400.0
+I1203 03:55:47.831684 137274321021824 utils.py:1231] [72350] progress = 0.6425228457501132
+I1203 03:55:47.831729 137274321021824 utils.py:1231] [72350] epoch = 57.82727778658052
+I1203 03:55:47.831778 137274321021824 utils.py:1231] [72350] img/sec/core = 164.22139158972826
+I1203 03:55:47.831830 137274321021824 utils.py:1231] [72350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.05934738376693
+I1203 03:55:47.831879 137274321021824 utils.py:1231] [72350] core_hours = 126.05934738376693
+I1203 03:55:47.831943 137274321021824 train.py:125] NOTE: Steps:72350/112603 [64.3%]
+Walltime:5d6h5m (0s eval)
+ETA:2d22h8m
+Total train time:8d4h11m
+I1203 04:00:59.597069 137274321021824 utils.py:1231] [72400] l2_params = 273.1318273224574
+I1203 04:00:59.597352 137274321021824 utils.py:1231] [72400] train/loss = 2.5173660218715668
+I1203 04:00:59.597583 137274321021824 utils.py:1231] [72400] l2_grads = 1.8875807523727417
+I1203 04:00:59.597664 137274321021824 utils.py:1231] [72400] lr = 0.0003333538336929389
+I1203 04:00:59.597716 137274321021824 utils.py:1231] [72400] uptime = 454248.960078432
+I1203 04:00:59.597778 137274321021824 utils.py:1231] [72400] examples_seen = 74137600.0
+I1203 04:00:59.597832 137274321021824 utils.py:1231] [72400] progress = 0.6429668836531887
+I1203 04:00:59.597885 137274321021824 utils.py:1231] [72400] epoch = 57.86724135104947
+I1203 04:00:59.597939 137274321021824 utils.py:1231] [72400] img/sec/core = 164.22566577355332
+I1203 04:00:59.598001 137274321021824 utils.py:1231] [72400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.14594908609666
+I1203 04:00:59.598053 137274321021824 utils.py:1231] [72400] core_hours = 126.14594908609666
+I1203 04:00:59.598120 137274321021824 train.py:125] NOTE: Steps:72400/112603 [64.3%]
+Walltime:5d6h10m (0s eval)
+ETA:2d22h2m
+Total train time:8d4h11m
+I1203 04:06:09.287843 137274321021824 utils.py:1231] [72450] l2_params = 273.0371845826624
+I1203 04:06:09.288073 137274321021824 utils.py:1231] [72450] train/loss = 2.836915224790573
+I1203 04:06:09.288207 137274321021824 utils.py:1231] [72450] l2_grads = 1.7434252500534058
+I1203 04:06:09.288302 137274321021824 utils.py:1231] [72450] lr = 0.000332632323396862
+I1203 04:06:09.288380 137274321021824 utils.py:1231] [72450] uptime = 454558.65073023597
+I1203 04:06:09.288457 137274321021824 utils.py:1231] [72450] examples_seen = 74188800.0
+I1203 04:06:09.288512 137274321021824 utils.py:1231] [72450] progress = 0.643410921556264
+I1203 04:06:09.288567 137274321021824 utils.py:1231] [72450] epoch = 57.90720491551843
+I1203 04:06:09.288639 137274321021824 utils.py:1231] [72450] img/sec/core = 165.32626897762458
+I1203 04:06:09.288700 137274321021824 utils.py:1231] [72450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.23197426715333
+I1203 04:06:09.288767 137274321021824 utils.py:1231] [72450] core_hours = 126.23197426715333
+I1203 04:06:09.288831 137274321021824 train.py:125] NOTE: Steps:72450/112603 [64.3%]
+Walltime:5d6h15m (0s eval)
+ETA:2d21h57m
+Total train time:8d4h11m
+I1203 04:11:18.995735 137274321021824 utils.py:1231] [72500] l2_params = 272.95006734174956
+I1203 04:11:18.995967 137274321021824 utils.py:1231] [72500] train/loss = 2.52109232544899
+I1203 04:11:18.996063 137274321021824 utils.py:1231] [72500] l2_grads = 1.8189953565597534
+I1203 04:11:18.996125 137274321021824 utils.py:1231] [72500] lr = 0.00033191120537624325
+I1203 04:11:18.996178 137274321021824 utils.py:1231] [72500] uptime = 454868.358540251
+I1203 04:11:18.996232 137274321021824 utils.py:1231] [72500] examples_seen = 74240000.0
+I1203 04:11:18.996282 137274321021824 utils.py:1231] [72500] progress = 0.6438549594593395
+I1203 04:11:18.996331 137274321021824 utils.py:1231] [72500] epoch = 57.94716847998739
+I1203 04:11:18.996382 137274321021824 utils.py:1231] [72500] img/sec/core = 165.31710968967352
+I1203 04:11:18.996438 137274321021824 utils.py:1231] [72500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.31800421437971
+I1203 04:11:18.996490 137274321021824 utils.py:1231] [72500] core_hours = 126.31800421437971
+I1203 04:11:18.996550 137274321021824 train.py:125] NOTE: Steps:72500/112603 [64.4%]
+Walltime:5d6h21m (0s eval)
+ETA:2d21h52m
+Total train time:8d4h11m
+I1203 04:11:18.996652 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:72500/112603 [64.4%]
+Walltime:5d6h21m (0s eval)
+ETA:2d21h52m
+Total train time:8d4h11m
+I1203 04:12:53.601464 137274321021824 utils.py:1231] [72500] val/acc@1 = 0.7017099808673469
+I1203 04:12:53.601711 137274321021824 utils.py:1231] [72500] val/loss = 1.1991954047460944
+I1203 04:12:53.601907 137274321021824 utils.py:1231] [72500] z/secs/eval/val = 94.6051839350257
+I1203 04:12:53.601978 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 94.6051839350257
+I1203 04:18:03.241187 137274321021824 utils.py:1231] [72550] l2_params = 272.8622969678614
+I1203 04:18:03.241451 137274321021824 utils.py:1231] [72550] train/loss = 1.9337876439094543
+I1203 04:18:03.241592 137274321021824 utils.py:1231] [72550] l2_grads = 1.9753814935684204
+I1203 04:18:03.241685 137274321021824 utils.py:1231] [72550] lr = 0.0003311904813212346
+I1203 04:18:03.241764 137274321021824 utils.py:1231] [72550] uptime = 455272.60412506596
+I1203 04:18:03.241834 137274321021824 utils.py:1231] [72550] examples_seen = 74291200.0
+I1203 04:18:03.241909 137274321021824 utils.py:1231] [72550] progress = 0.6442989973624148
+I1203 04:18:03.241967 137274321021824 utils.py:1231] [72550] epoch = 57.987132044456345
+I1203 04:18:03.242023 137274321021824 utils.py:1231] [72550] img/sec/core = 126.65568140573097
+I1203 04:18:03.242084 137274321021824 utils.py:1231] [72550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.4302946546061
+I1203 04:18:03.242137 137274321021824 utils.py:1231] [72550] core_hours = 126.4302946546061
+I1203 04:18:03.242201 137274321021824 train.py:125] NOTE: Steps:72550/112603 [64.4%]
+Walltime:5d6h27m (0s eval)
+ETA:2d21h48m
+Total train time:8d4h14m
+I1203 04:23:12.764228 137274321021824 utils.py:1231] [72600] l2_params = 272.7722549734059
+I1203 04:23:12.764479 137274321021824 utils.py:1231] [72600] train/loss = 3.8397835195064545
+I1203 04:23:12.764639 137274321021824 utils.py:1231] [72600] l2_grads = 1.814430594444275
+I1203 04:23:12.764742 137274321021824 utils.py:1231] [72600] lr = 0.0003304701529210657
+I1203 04:23:12.764817 137274321021824 utils.py:1231] [72600] uptime = 455582.12717714
+I1203 04:23:12.764910 137274321021824 utils.py:1231] [72600] examples_seen = 74342400.0
+I1203 04:23:12.764994 137274321021824 utils.py:1231] [72600] progress = 0.6447430352654903
+I1203 04:23:12.765051 137274321021824 utils.py:1231] [72600] epoch = 58.0270956089253
+I1203 04:23:12.765115 137274321021824 utils.py:1231] [72600] img/sec/core = 165.41578941188757
+I1203 04:23:12.765183 137274321021824 utils.py:1231] [72600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.51627328018222
+I1203 04:23:12.765239 137274321021824 utils.py:1231] [72600] core_hours = 126.51627328018222
+I1203 04:23:12.765308 137274321021824 train.py:125] NOTE: Steps:72600/112603 [64.5%]
+Walltime:5d6h33m (0s eval)
+ETA:2d21h42m
+Total train time:8d4h13m
+I1203 04:28:24.540723 137274321021824 utils.py:1231] [72650] l2_params = 272.68297918560484
+I1203 04:28:24.540989 137274321021824 utils.py:1231] [72650] train/loss = 2.0249488055706024
+I1203 04:28:24.541128 137274321021824 utils.py:1231] [72650] l2_grads = 1.973745584487915
+I1203 04:28:24.541231 137274321021824 utils.py:1231] [72650] lr = 0.0003297502218640381
+I1203 04:28:24.541309 137274321021824 utils.py:1231] [72650] uptime = 455893.903662538
+I1203 04:28:24.541378 137274321021824 utils.py:1231] [72650] examples_seen = 74393600.0
+I1203 04:28:24.541446 137274321021824 utils.py:1231] [72650] progress = 0.6451870731685657
+I1203 04:28:24.541510 137274321021824 utils.py:1231] [72650] epoch = 58.067059173394256
+I1203 04:28:24.541576 137274321021824 utils.py:1231] [72650] img/sec/core = 164.22021030431353
+I1203 04:28:24.541648 137274321021824 utils.py:1231] [72650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.60287785945943
+I1203 04:28:24.541705 137274321021824 utils.py:1231] [72650] core_hours = 126.60287785945943
+I1203 04:28:24.541772 137274321021824 train.py:125] NOTE: Steps:72650/112603 [64.5%]
+Walltime:5d6h38m (0s eval)
+ETA:2d21h37m
+Total train time:8d4h13m
+I1203 04:33:36.327033 137274321021824 utils.py:1231] [72700] l2_params = 272.60085133063563
+I1203 04:33:36.327258 137274321021824 utils.py:1231] [72700] train/loss = 1.9781047701835632
+I1203 04:33:36.327391 137274321021824 utils.py:1231] [72700] l2_grads = 1.9172431230545044
+I1203 04:33:36.327473 137274321021824 utils.py:1231] [72700] lr = 0.0003290306898375224
+I1203 04:33:36.327536 137274321021824 utils.py:1231] [72700] uptime = 456205.68989750696
+I1203 04:33:36.327592 137274321021824 utils.py:1231] [72700] examples_seen = 74444800.0
+I1203 04:33:36.327652 137274321021824 utils.py:1231] [72700] progress = 0.6456311110716411
+I1203 04:33:36.327710 137274321021824 utils.py:1231] [72700] epoch = 58.10702273786321
+I1203 04:33:36.327774 137274321021824 utils.py:1231] [72700] img/sec/core = 164.21507513022053
+I1203 04:33:36.327833 137274321021824 utils.py:1231] [72700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.68948514695082
+I1203 04:33:36.327896 137274321021824 utils.py:1231] [72700] core_hours = 126.68948514695082
+I1203 04:33:36.327972 137274321021824 train.py:125] NOTE: Steps:72700/112603 [64.6%]
+Walltime:5d6h43m (0s eval)
+ETA:2d21h32m
+Total train time:8d4h13m
+I1203 04:38:47.458785 137274321021824 utils.py:1231] [72750] l2_params = 272.5112592359361
+I1203 04:38:47.459035 137274321021824 utils.py:1231] [72750] train/loss = 1.9126514494419098
+I1203 04:38:47.459173 137274321021824 utils.py:1231] [72750] l2_grads = 1.94992995262146
+I1203 04:38:47.459260 137274321021824 utils.py:1231] [72750] lr = 0.0003283115585279539
+I1203 04:38:47.459313 137274321021824 utils.py:1231] [72750] uptime = 456516.82167489
+I1203 04:38:47.459367 137274321021824 utils.py:1231] [72750] examples_seen = 74496000.0
+I1203 04:38:47.459417 137274321021824 utils.py:1231] [72750] progress = 0.6460751489747165
+I1203 04:38:47.459465 137274321021824 utils.py:1231] [72750] epoch = 58.146986302332174
+I1203 04:38:47.459521 137274321021824 utils.py:1231] [72750] img/sec/core = 164.56049726147742
+I1203 04:38:47.459587 137274321021824 utils.py:1231] [72750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.77591064066833
+I1203 04:38:47.459642 137274321021824 utils.py:1231] [72750] core_hours = 126.77591064066833
+I1203 04:38:47.459709 137274321021824 train.py:125] NOTE: Steps:72750/112603 [64.6%]
+Walltime:5d6h48m (0s eval)
+ETA:2d21h27m
+Total train time:8d4h13m
+I1203 04:43:58.598093 137274321021824 utils.py:1231] [72800] l2_params = 272.4181569629267
+I1203 04:43:58.598299 137274321021824 utils.py:1231] [72800] train/loss = 3.4113932847976685
+I1203 04:43:58.598402 137274321021824 utils.py:1231] [72800] l2_grads = 1.8040587902069092
+I1203 04:43:58.598460 137274321021824 utils.py:1231] [72800] lr = 0.000327592829620829
+I1203 04:43:58.598512 137274321021824 utils.py:1231] [72800] uptime = 456827.960874531
+I1203 04:43:58.598562 137274321021824 utils.py:1231] [72800] examples_seen = 74547200.0
+I1203 04:43:58.598610 137274321021824 utils.py:1231] [72800] progress = 0.6465191868777919
+I1203 04:43:58.598657 137274321021824 utils.py:1231] [72800] epoch = 58.18694986680113
+I1203 04:43:58.598705 137274321021824 utils.py:1231] [72800] img/sec/core = 164.556571653698
+I1203 04:43:58.598763 137274321021824 utils.py:1231] [72800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.86233819612417
+I1203 04:43:58.598813 137274321021824 utils.py:1231] [72800] core_hours = 126.86233819612417
+I1203 04:43:58.598870 137274321021824 train.py:125] NOTE: Steps:72800/112603 [64.7%]
+Walltime:5d6h53m (0s eval)
+ETA:2d21h21m
+Total train time:8d4h13m
+I1203 04:49:09.731873 137274321021824 utils.py:1231] [72850] l2_params = 272.335836231199
+I1203 04:49:09.732104 137274321021824 utils.py:1231] [72850] train/loss = 1.996460661292076
+I1203 04:49:09.732214 137274321021824 utils.py:1231] [72850] l2_grads = 1.9046235084533691
+I1203 04:49:09.732285 137274321021824 utils.py:1231] [72850] lr = 0.00032687450480069994
+I1203 04:49:09.732346 137274321021824 utils.py:1231] [72850] uptime = 457139.09470780997
+I1203 04:49:09.732407 137274321021824 utils.py:1231] [72850] examples_seen = 74598400.0
+I1203 04:49:09.732464 137274321021824 utils.py:1231] [72850] progress = 0.6469632247808673
+I1203 04:49:09.732533 137274321021824 utils.py:1231] [72850] epoch = 58.226913431270084
+I1203 04:49:09.732590 137274321021824 utils.py:1231] [72850] img/sec/core = 164.55940988614535
+I1203 04:49:09.732659 137274321021824 utils.py:1231] [72850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 126.94876426092387
+I1203 04:49:09.732729 137274321021824 utils.py:1231] [72850] core_hours = 126.94876426092387
+I1203 04:49:09.732795 137274321021824 train.py:125] NOTE: Steps:72850/112603 [64.7%]
+Walltime:5d6h58m (0s eval)
+ETA:2d21h16m
+Total train time:8d4h13m
+I1203 04:54:20.900108 137274321021824 utils.py:1231] [72900] l2_params = 272.24738148475643
+I1203 04:54:20.900374 137274321021824 utils.py:1231] [72900] train/loss = 4.295711636543274
+I1203 04:54:20.900494 137274321021824 utils.py:1231] [72900] l2_grads = 1.8264271020889282
+I1203 04:54:20.900576 137274321021824 utils.py:1231] [72900] lr = 0.00032615658575117307
+I1203 04:54:20.900644 137274321021824 utils.py:1231] [72900] uptime = 457450.263005948
+I1203 04:54:20.900698 137274321021824 utils.py:1231] [72900] examples_seen = 74649600.0
+I1203 04:54:20.900747 137274321021824 utils.py:1231] [72900] progress = 0.6474072626839427
+I1203 04:54:20.900795 137274321021824 utils.py:1231] [72900] epoch = 58.26687699573904
+I1203 04:54:20.900845 137274321021824 utils.py:1231] [72900] img/sec/core = 164.54118336081567
+I1203 04:54:20.900914 137274321021824 utils.py:1231] [72900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.03519989929555
+I1203 04:54:20.900989 137274321021824 utils.py:1231] [72900] core_hours = 127.03519989929555
+I1203 04:54:20.901053 137274321021824 train.py:125] NOTE: Steps:72900/112603 [64.7%]
+Walltime:5d7h4m (0s eval)
+ETA:2d21h11m
+Total train time:8d4h13m
+I1203 04:59:31.837200 137274321021824 utils.py:1231] [72950] l2_params = 272.1517909243496
+I1203 04:59:31.837474 137274321021824 utils.py:1231] [72950] train/loss = 2.339048981666565
+I1203 04:59:31.837690 137274321021824 utils.py:1231] [72950] l2_grads = 1.8074365854263306
+I1203 04:59:31.837807 137274321021824 utils.py:1231] [72950] lr = 0.0003254390741549035
+I1203 04:59:31.837919 137274321021824 utils.py:1231] [72950] uptime = 457761.200275898
+I1203 04:59:31.837997 137274321021824 utils.py:1231] [72950] examples_seen = 74700800.0
+I1203 04:59:31.838065 137274321021824 utils.py:1231] [72950] progress = 0.6478513005870181
+I1203 04:59:31.838132 137274321021824 utils.py:1231] [72950] epoch = 58.306840560207995
+I1203 04:59:31.838194 137274321021824 utils.py:1231] [72950] img/sec/core = 164.66343841069317
+I1203 04:59:31.838258 137274321021824 utils.py:1231] [72950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.12157136317055
+I1203 04:59:31.838321 137274321021824 utils.py:1231] [72950] core_hours = 127.12157136317055
+I1203 04:59:31.838394 137274321021824 train.py:125] NOTE: Steps:72950/112603 [64.8%]
+Walltime:5d7h9m (0s eval)
+ETA:2d21h6m
+Total train time:8d4h13m
+I1203 05:04:42.745545 137274321021824 utils.py:1231] [73000] l2_params = 272.06624530893237
+I1203 05:04:42.745788 137274321021824 utils.py:1231] [73000] train/loss = 2.5440632104873657
+I1203 05:04:42.745927 137274321021824 utils.py:1231] [73000] l2_grads = 1.7833682298660278
+I1203 05:04:42.746009 137274321021824 utils.py:1231] [73000] lr = 0.00032472197169359
+I1203 05:04:42.746075 137274321021824 utils.py:1231] [73000] uptime = 458072.108436962
+I1203 05:04:42.746135 137274321021824 utils.py:1231] [73000] examples_seen = 74752000.0
+I1203 05:04:42.746209 137274321021824 utils.py:1231] [73000] progress = 0.6482953384900935
+I1203 05:04:42.746276 137274321021824 utils.py:1231] [73000] epoch = 58.34680412467696
+I1203 05:04:42.746341 137274321021824 utils.py:1231] [73000] img/sec/core = 164.67885508304806
+I1203 05:04:42.746408 137274321021824 utils.py:1231] [73000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.20793474124389
+I1203 05:04:42.746464 137274321021824 utils.py:1231] [73000] core_hours = 127.20793474124389
+I1203 05:04:42.746536 137274321021824 train.py:125] NOTE: Steps:73000/112603 [64.8%]
+Walltime:5d7h14m (0s eval)
+ETA:2d21h0m
+Total train time:8d4h13m
+I1203 05:09:53.916400 137274321021824 utils.py:1231] [73050] l2_params = 271.98678112691516
+I1203 05:09:53.916610 137274321021824 utils.py:1231] [73050] train/loss = 2.110363930463791
+I1203 05:09:53.916729 137274321021824 utils.py:1231] [73050] l2_grads = 1.9841982126235962
+I1203 05:09:53.916814 137274321021824 utils.py:1231] [73050] lr = 0.00032400528004797454
+I1203 05:09:53.916878 137274321021824 utils.py:1231] [73050] uptime = 458383.279239885
+I1203 05:09:53.916948 137274321021824 utils.py:1231] [73050] examples_seen = 74803200.0
+I1203 05:09:53.917009 137274321021824 utils.py:1231] [73050] progress = 0.6487393763931689
+I1203 05:09:53.917067 137274321021824 utils.py:1231] [73050] epoch = 58.38676768914591
+I1203 05:09:53.917129 137274321021824 utils.py:1231] [73050] img/sec/core = 164.53985887831917
+I1203 05:09:53.917188 137274321021824 utils.py:1231] [73050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.29437107538915
+I1203 05:09:53.917244 137274321021824 utils.py:1231] [73050] core_hours = 127.29437107538915
+I1203 05:09:53.917313 137274321021824 train.py:125] NOTE: Steps:73050/112603 [64.9%]
+Walltime:5d7h19m (0s eval)
+ETA:2d20h55m
+Total train time:8d4h13m
+I1203 05:15:04.990920 137274321021824 utils.py:1231] [73100] l2_params = 271.8964919772662
+I1203 05:15:04.991247 137274321021824 utils.py:1231] [73100] train/loss = 4.453470408916473
+I1203 05:15:04.991485 137274321021824 utils.py:1231] [73100] l2_grads = 2.085068702697754
+I1203 05:15:04.991620 137274321021824 utils.py:1231] [73100] lr = 0.0003232890008978343
+I1203 05:15:04.991730 137274321021824 utils.py:1231] [73100] uptime = 458694.354086698
+I1203 05:15:04.991850 137274321021824 utils.py:1231] [73100] examples_seen = 74854400.0
+I1203 05:15:04.991955 137274321021824 utils.py:1231] [73100] progress = 0.6491834142962444
+I1203 05:15:04.992025 137274321021824 utils.py:1231] [73100] epoch = 58.42673125361487
+I1203 05:15:04.992099 137274321021824 utils.py:1231] [73100] img/sec/core = 164.59061388136007
+I1203 05:15:04.992178 137274321021824 utils.py:1231] [73100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.38078075505943
+I1203 05:15:04.992264 137274321021824 utils.py:1231] [73100] core_hours = 127.38078075505943
+I1203 05:15:04.992357 137274321021824 train.py:125] NOTE: Steps:73100/112603 [64.9%]
+Walltime:5d7h24m (0s eval)
+ETA:2d20h50m
+Total train time:8d4h13m
+I1203 05:20:16.149190 137274321021824 utils.py:1231] [73150] l2_params = 271.80251108879133
+I1203 05:20:16.149385 137274321021824 utils.py:1231] [73150] train/loss = 2.7154204547405243
+I1203 05:20:16.149488 137274321021824 utils.py:1231] [73150] l2_grads = 1.8503552675247192
+I1203 05:20:16.149548 137274321021824 utils.py:1231] [73150] lr = 0.00032257313592198096
+I1203 05:20:16.149600 137274321021824 utils.py:1231] [73150] uptime = 459005.511961717
+I1203 05:20:16.149653 137274321021824 utils.py:1231] [73150] examples_seen = 74905600.0
+I1203 05:20:16.149703 137274321021824 utils.py:1231] [73150] progress = 0.6496274521993197
+I1203 05:20:16.149751 137274321021824 utils.py:1231] [73150] epoch = 58.46669481808382
+I1203 05:20:16.149801 137274321021824 utils.py:1231] [73150] img/sec/core = 164.54669513626192
+I1203 05:20:16.149857 137274321021824 utils.py:1231] [73150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.46721349812027
+I1203 05:20:16.149922 137274321021824 utils.py:1231] [73150] core_hours = 127.46721349812027
+I1203 05:20:16.150005 137274321021824 train.py:125] NOTE: Steps:73150/112603 [65.0%]
+Walltime:5d7h30m (0s eval)
+ETA:2d20h45m
+Total train time:8d4h13m
+I1203 05:25:27.931672 137274321021824 utils.py:1231] [73200] l2_params = 271.719917729848
+I1203 05:25:27.931971 137274321021824 utils.py:1231] [73200] train/loss = 2.1479362845420837
+I1203 05:25:27.932130 137274321021824 utils.py:1231] [73200] l2_grads = 2.0994181632995605
+I1203 05:25:27.932209 137274321021824 utils.py:1231] [73200] lr = 0.00032185768679825503
+I1203 05:25:27.932262 137274321021824 utils.py:1231] [73200] uptime = 459317.294624366
+I1203 05:25:27.932316 137274321021824 utils.py:1231] [73200] examples_seen = 74956800.0
+I1203 05:25:27.932367 137274321021824 utils.py:1231] [73200] progress = 0.6500714901023952
+I1203 05:25:27.932418 137274321021824 utils.py:1231] [73200] epoch = 58.50665838255278
+I1203 05:25:27.932473 137274321021824 utils.py:1231] [73200] img/sec/core = 164.21695666137913
+I1203 05:25:27.932531 137274321021824 utils.py:1231] [73200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.55381979330055
+I1203 05:25:27.932583 137274321021824 utils.py:1231] [73200] core_hours = 127.55381979330055
+I1203 05:25:27.932645 137274321021824 train.py:125] NOTE: Steps:73200/112603 [65.0%]
+Walltime:5d7h35m (0s eval)
+ETA:2d20h39m
+Total train time:8d4h13m
+I1203 05:30:39.709340 137274321021824 utils.py:1231] [73250] l2_params = 271.63576796520016
+I1203 05:30:39.709574 137274321021824 utils.py:1231] [73250] train/loss = 2.6868561506271362
+I1203 05:30:39.709832 137274321021824 utils.py:1231] [73250] l2_grads = 1.7615042924880981
+I1203 05:30:39.709942 137274321021824 utils.py:1231] [73250] lr = 0.0003211426552035218
+I1203 05:30:39.710018 137274321021824 utils.py:1231] [73250] uptime = 459629.072377872
+I1203 05:30:39.710088 137274321021824 utils.py:1231] [73250] examples_seen = 75008000.0
+I1203 05:30:39.710147 137274321021824 utils.py:1231] [73250] progress = 0.6505155280054705
+I1203 05:30:39.710231 137274321021824 utils.py:1231] [73250] epoch = 58.54662194702174
+I1203 05:30:39.710320 137274321021824 utils.py:1231] [73250] img/sec/core = 164.21954236388353
+I1203 05:30:39.710446 137274321021824 utils.py:1231] [73250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.64042472483
+I1203 05:30:39.710542 137274321021824 utils.py:1231] [73250] core_hours = 127.64042472483
+I1203 05:30:39.710669 137274321021824 train.py:125] NOTE: Steps:73250/112603 [65.1%]
+Walltime:5d7h40m (0s eval)
+ETA:2d20h34m
+Total train time:8d4h13m
+I1203 05:35:46.302456 137274321021824 utils.py:1231] [73300] l2_params = 271.5519565845643
+I1203 05:35:46.302665 137274321021824 utils.py:1231] [73300] train/loss = 2.316071540117264
+I1203 05:35:46.302760 137274321021824 utils.py:1231] [73300] l2_grads = 1.693426489830017
+I1203 05:35:46.302821 137274321021824 utils.py:1231] [73300] lr = 0.0003204280428136689
+I1203 05:35:46.302873 137274321021824 utils.py:1231] [73300] uptime = 459935.66523489
+I1203 05:35:46.302932 137274321021824 utils.py:1231] [73300] examples_seen = 75059200.0
+I1203 05:35:46.302981 137274321021824 utils.py:1231] [73300] progress = 0.650959565908546
+I1203 05:35:46.303032 137274321021824 utils.py:1231] [73300] epoch = 58.586585511490696
+I1203 05:35:46.303084 137274321021824 utils.py:1231] [73300] img/sec/core = 166.9967151158861
+I1203 05:35:46.303138 137274321021824 utils.py:1231] [73300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.72558940733501
+I1203 05:35:46.303188 137274321021824 utils.py:1231] [73300] core_hours = 127.72558940733501
+I1203 05:35:46.303249 137274321021824 train.py:125] NOTE: Steps:73300/112603 [65.1%]
+Walltime:5d7h45m (0s eval)
+ETA:2d20h29m
+Total train time:8d4h12m
+I1203 05:40:57.809647 137274321021824 utils.py:1231] [73350] l2_params = 271.4617825361933
+I1203 05:40:57.809873 137274321021824 utils.py:1231] [73350] train/loss = 2.044428914785385
+I1203 05:40:57.810008 137274321021824 utils.py:1231] [73350] l2_grads = 1.9147529602050781
+I1203 05:40:57.810081 137274321021824 utils.py:1231] [73350] lr = 0.0003197138513036011
+I1203 05:40:57.810137 137274321021824 utils.py:1231] [73350] uptime = 460247.172499375
+I1203 05:40:57.810194 137274321021824 utils.py:1231] [73350] examples_seen = 75110400.0
+I1203 05:40:57.810249 137274321021824 utils.py:1231] [73350] progress = 0.6514036038116213
+I1203 05:40:57.810305 137274321021824 utils.py:1231] [73350] epoch = 58.62654907595965
+I1203 05:40:57.810362 137274321021824 utils.py:1231] [73350] img/sec/core = 164.3621380215527
+I1203 05:40:57.810425 137274321021824 utils.py:1231] [73350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.81211920302528
+I1203 05:40:57.810488 137274321021824 utils.py:1231] [73350] core_hours = 127.81211920302528
+I1203 05:40:57.810563 137274321021824 train.py:125] NOTE: Steps:73350/112603 [65.1%]
+Walltime:5d7h50m (0s eval)
+ETA:2d20h24m
+Total train time:8d4h12m
+I1203 05:46:09.586864 137274321021824 utils.py:1231] [73400] l2_params = 271.3788975612997
+I1203 05:46:09.587077 137274321021824 utils.py:1231] [73400] train/loss = 1.950067862868309
+I1203 05:46:09.587181 137274321021824 utils.py:1231] [73400] l2_grads = 1.877314567565918
+I1203 05:46:09.587253 137274321021824 utils.py:1231] [73400] lr = 0.0003190000823472359
+I1203 05:46:09.587321 137274321021824 utils.py:1231] [73400] uptime = 460558.949681938
+I1203 05:46:09.587381 137274321021824 utils.py:1231] [73400] examples_seen = 75161600.0
+I1203 05:46:09.587437 137274321021824 utils.py:1231] [73400] progress = 0.6518476417146968
+I1203 05:46:09.587493 137274321021824 utils.py:1231] [73400] epoch = 58.66651264042861
+I1203 05:46:09.587553 137274321021824 utils.py:1231] [73400] img/sec/core = 164.21984309149443
+I1203 05:46:09.587616 137274321021824 utils.py:1231] [73400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.89872397595944
+I1203 05:46:09.587673 137274321021824 utils.py:1231] [73400] core_hours = 127.89872397595944
+I1203 05:46:09.587740 137274321021824 train.py:125] NOTE: Steps:73400/112603 [65.2%]
+Walltime:5d7h55m (0s eval)
+ETA:2d20h18m
+Total train time:8d4h12m
+I1203 05:51:21.279156 137274321021824 utils.py:1231] [73450] l2_params = 271.289304351589
+I1203 05:51:21.279396 137274321021824 utils.py:1231] [73450] train/loss = 2.9153760969638824
+I1203 05:51:21.279519 137274321021824 utils.py:1231] [73450] l2_grads = 1.8423810005187988
+I1203 05:51:21.279594 137274321021824 utils.py:1231] [73450] lr = 0.0003182867376175022
+I1203 05:51:21.279653 137274321021824 utils.py:1231] [73450] uptime = 460870.642015313
+I1203 05:51:21.279714 137274321021824 utils.py:1231] [73450] examples_seen = 75212800.0
+I1203 05:51:21.279764 137274321021824 utils.py:1231] [73450] progress = 0.6522916796177721
+I1203 05:51:21.279811 137274321021824 utils.py:1231] [73450] epoch = 58.70647620489757
+I1203 05:51:21.279860 137274321021824 utils.py:1231] [73450] img/sec/core = 164.2645471757654
+I1203 05:51:21.279930 137274321021824 utils.py:1231] [73450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 127.98530517967471
+I1203 05:51:21.279980 137274321021824 utils.py:1231] [73450] core_hours = 127.98530517967471
+I1203 05:51:21.280039 137274321021824 train.py:125] NOTE: Steps:73450/112603 [65.2%]
+Walltime:5d8h1m (0s eval)
+ETA:2d20h13m
+Total train time:8d4h12m
+I1203 05:56:32.926339 137274321021824 utils.py:1231] [73500] l2_params = 271.2024966713581
+I1203 05:56:32.926561 137274321021824 utils.py:1231] [73500] train/loss = 3.8157162964344025
+I1203 05:56:32.926662 137274321021824 utils.py:1231] [73500] l2_grads = 1.7843977212905884
+I1203 05:56:32.926731 137274321021824 utils.py:1231] [73500] lr = 0.000317573818786333
+I1203 05:56:32.926792 137274321021824 utils.py:1231] [73500] uptime = 461182.289152811
+I1203 05:56:32.926851 137274321021824 utils.py:1231] [73500] examples_seen = 75264000.0
+I1203 05:56:32.926917 137274321021824 utils.py:1231] [73500] progress = 0.6527357175208476
+I1203 05:56:32.926975 137274321021824 utils.py:1231] [73500] epoch = 58.746439769366525
+I1203 05:56:32.927033 137274321021824 utils.py:1231] [73500] img/sec/core = 164.28836924685604
+I1203 05:56:32.927095 137274321021824 utils.py:1231] [73500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.0718738289797
+I1203 05:56:32.927152 137274321021824 utils.py:1231] [73500] core_hours = 128.0718738289797
+I1203 05:56:32.927218 137274321021824 train.py:125] NOTE: Steps:73500/112603 [65.3%]
+Walltime:5d8h6m (0s eval)
+ETA:2d20h8m
+Total train time:8d4h12m
+I1203 06:01:44.613019 137274321021824 utils.py:1231] [73550] l2_params = 271.1151157146427
+I1203 06:01:44.613261 137274321021824 utils.py:1231] [73550] train/loss = 3.1390212178230286
+I1203 06:01:44.613411 137274321021824 utils.py:1231] [73550] l2_grads = 1.7700163125991821
+I1203 06:01:44.613508 137274321021824 utils.py:1231] [73550] lr = 0.00031686132752466325
+I1203 06:01:44.613574 137274321021824 utils.py:1231] [73550] uptime = 461493.975936145
+I1203 06:01:44.613648 137274321021824 utils.py:1231] [73550] examples_seen = 75315200.0
+I1203 06:01:44.613714 137274321021824 utils.py:1231] [73550] progress = 0.653179755423923
+I1203 06:01:44.613780 137274321021824 utils.py:1231] [73550] epoch = 58.78640333383548
+I1203 06:01:44.613846 137274321021824 utils.py:1231] [73550] img/sec/core = 164.26747214729807
+I1203 06:01:44.613922 137274321021824 utils.py:1231] [73550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.15845349101696
+I1203 06:01:44.613979 137274321021824 utils.py:1231] [73550] core_hours = 128.15845349101696
+I1203 06:01:44.614043 137274321021824 train.py:125] NOTE: Steps:73550/112603 [65.3%]
+Walltime:5d8h11m (0s eval)
+ETA:2d20h3m
+Total train time:8d4h12m
+I1203 06:06:56.268344 137274321021824 utils.py:1231] [73600] l2_params = 271.02943265440433
+I1203 06:06:56.268543 137274321021824 utils.py:1231] [73600] train/loss = 4.1935156881809235
+I1203 06:06:56.268642 137274321021824 utils.py:1231] [73600] l2_grads = 1.7688305377960205
+I1203 06:06:56.268701 137274321021824 utils.py:1231] [73600] lr = 0.00031614926550242663
+I1203 06:06:56.268754 137274321021824 utils.py:1231] [73600] uptime = 461805.631116158
+I1203 06:06:56.268807 137274321021824 utils.py:1231] [73600] examples_seen = 75366400.0
+I1203 06:06:56.268857 137274321021824 utils.py:1231] [73600] progress = 0.6536237933269984
+I1203 06:06:56.268911 137274321021824 utils.py:1231] [73600] epoch = 58.826366898304435
+I1203 06:06:56.268962 137274321021824 utils.py:1231] [73600] img/sec/core = 164.28412965209588
+I1203 06:06:56.269018 137274321021824 utils.py:1231] [73600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.24502437435387
+I1203 06:06:56.269068 137274321021824 utils.py:1231] [73600] core_hours = 128.24502437435387
+I1203 06:06:56.269129 137274321021824 train.py:125] NOTE: Steps:73600/112603 [65.4%]
+Walltime:5d8h16m (0s eval)
+ETA:2d19h57m
+Total train time:8d4h12m
+I1203 06:12:07.873477 137274321021824 utils.py:1231] [73650] l2_params = 270.93647173771103
+I1203 06:12:07.873750 137274321021824 utils.py:1231] [73650] train/loss = 1.8916912227869034
+I1203 06:12:07.873879 137274321021824 utils.py:1231] [73650] l2_grads = 1.938578486442566
+I1203 06:12:07.873966 137274321021824 utils.py:1231] [73650] lr = 0.00031543763438855
+I1203 06:12:07.874028 137274321021824 utils.py:1231] [73650] uptime = 462117.23638655996
+I1203 06:12:07.874090 137274321021824 utils.py:1231] [73650] examples_seen = 75417600.0
+I1203 06:12:07.874139 137274321021824 utils.py:1231] [73650] progress = 0.6540678312300738
+I1203 06:12:07.874196 137274321021824 utils.py:1231] [73650] epoch = 58.86633046277339
+I1203 06:12:07.874247 137274321021824 utils.py:1231] [73650] img/sec/core = 164.3104429329778
+I1203 06:12:07.874305 137274321021824 utils.py:1231] [73650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.33158139390997
+I1203 06:12:07.874362 137274321021824 utils.py:1231] [73650] core_hours = 128.33158139390997
+I1203 06:12:07.874422 137274321021824 train.py:125] NOTE: Steps:73650/112603 [65.4%]
+Walltime:5d8h21m (0s eval)
+ETA:2d19h52m
+Total train time:8d4h12m
+I1203 06:17:19.624995 137274321021824 utils.py:1231] [73700] l2_params = 270.8494462388825
+I1203 06:17:19.625235 137274321021824 utils.py:1231] [73700] train/loss = 4.1996448040008545
+I1203 06:17:19.625349 137274321021824 utils.py:1231] [73700] l2_grads = 1.8405227661132812
+I1203 06:17:19.625433 137274321021824 utils.py:1231] [73700] lr = 0.0003147264358509509
+I1203 06:17:19.625495 137274321021824 utils.py:1231] [73700] uptime = 462428.987857093
+I1203 06:17:19.625565 137274321021824 utils.py:1231] [73700] examples_seen = 75468800.0
+I1203 06:17:19.625617 137274321021824 utils.py:1231] [73700] progress = 0.6545118691331492
+I1203 06:17:19.625665 137274321021824 utils.py:1231] [73700] epoch = 58.90629402724235
+I1203 06:17:19.625717 137274321021824 utils.py:1231] [73700] img/sec/core = 164.23338729552583
+I1203 06:17:19.625778 137274321021824 utils.py:1231] [73700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.4181790246136
+I1203 06:17:19.625827 137274321021824 utils.py:1231] [73700] core_hours = 128.4181790246136
+I1203 06:17:19.625890 137274321021824 train.py:125] NOTE: Steps:73700/112603 [65.5%]
+Walltime:5d8h27m (0s eval)
+ETA:2d19h47m
+Total train time:8d4h12m
+I1203 06:22:31.249580 137274321021824 utils.py:1231] [73750] l2_params = 270.76540178145234
+I1203 06:22:31.249787 137274321021824 utils.py:1231] [73750] train/loss = 2.359671950340271
+I1203 06:22:31.249878 137274321021824 utils.py:1231] [73750] l2_grads = 1.8662614822387695
+I1203 06:22:31.249945 137274321021824 utils.py:1231] [73750] lr = 0.0003140156715565325
+I1203 06:22:31.249999 137274321021824 utils.py:1231] [73750] uptime = 462740.612360993
+I1203 06:22:31.250051 137274321021824 utils.py:1231] [73750] examples_seen = 75520000.0
+I1203 06:22:31.250101 137274321021824 utils.py:1231] [73750] progress = 0.6549559070362246
+I1203 06:22:31.250148 137274321021824 utils.py:1231] [73750] epoch = 58.94625759171131
+I1203 06:22:31.250209 137274321021824 utils.py:1231] [73750] img/sec/core = 164.30030167470096
+I1203 06:22:31.250263 137274321021824 utils.py:1231] [73750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.50474138680804
+I1203 06:22:31.250312 137274321021824 utils.py:1231] [73750] core_hours = 128.50474138680804
+I1203 06:22:31.250370 137274321021824 train.py:125] NOTE: Steps:73750/112603 [65.5%]
+Walltime:5d8h32m (0s eval)
+ETA:2d19h42m
+Total train time:8d4h12m
+I1203 06:27:42.875562 137274321021824 utils.py:1231] [73800] l2_params = 270.67491209042464
+I1203 06:27:42.875771 137274321021824 utils.py:1231] [73800] train/loss = 2.0586263239383698
+I1203 06:27:42.875870 137274321021824 utils.py:1231] [73800] l2_grads = 1.778334617614746
+I1203 06:27:42.875950 137274321021824 utils.py:1231] [73800] lr = 0.0003133053431711799
+I1203 06:27:42.876010 137274321021824 utils.py:1231] [73800] uptime = 463052.23837186
+I1203 06:27:42.876071 137274321021824 utils.py:1231] [73800] examples_seen = 75571200.0
+I1203 06:27:42.876127 137274321021824 utils.py:1231] [73800] progress = 0.6553999449393
+I1203 06:27:42.876182 137274321021824 utils.py:1231] [73800] epoch = 58.986221156180264
+I1203 06:27:42.876241 137274321021824 utils.py:1231] [73800] img/sec/core = 164.2995071481698
+I1203 06:27:42.876304 137274321021824 utils.py:1231] [73800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.59130416760445
+I1203 06:27:42.876356 137274321021824 utils.py:1231] [73800] core_hours = 128.59130416760445
+I1203 06:27:42.876420 137274321021824 train.py:125] NOTE: Steps:73800/112603 [65.5%]
+Walltime:5d8h37m (0s eval)
+ETA:2d19h36m
+Total train time:8d4h12m
+I1203 06:32:54.452322 137274321021824 utils.py:1231] [73850] l2_params = 270.58659536469787
+I1203 06:32:54.452558 137274321021824 utils.py:1231] [73850] train/loss = 2.3036336600780487
+I1203 06:32:54.452709 137274321021824 utils.py:1231] [73850] l2_grads = 1.8577419519424438
+I1203 06:32:54.452811 137274321021824 utils.py:1231] [73850] lr = 0.0003125954523597573
+I1203 06:32:54.452902 137274321021824 utils.py:1231] [73850] uptime = 463363.815258384
+I1203 06:32:54.452985 137274321021824 utils.py:1231] [73850] examples_seen = 75622400.0
+I1203 06:32:54.453062 137274321021824 utils.py:1231] [73850] progress = 0.6558439828423754
+I1203 06:32:54.453144 137274321021824 utils.py:1231] [73850] epoch = 59.02618472064922
+I1203 06:32:54.453231 137274321021824 utils.py:1231] [73850] img/sec/core = 164.32541120491416
+I1203 06:32:54.453314 137274321021824 utils.py:1231] [73850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.67785330274998
+I1203 06:32:54.453383 137274321021824 utils.py:1231] [73850] core_hours = 128.67785330274998
+I1203 06:32:54.453481 137274321021824 train.py:125] NOTE: Steps:73850/112603 [65.6%]
+Walltime:5d8h42m (0s eval)
+ETA:2d19h31m
+Total train time:8d4h12m
+I1203 06:38:04.876006 137274321021824 utils.py:1231] [73900] l2_params = 270.4952161244947
+I1203 06:38:04.876416 137274321021824 utils.py:1231] [73900] train/loss = 4.184878647327423
+I1203 06:38:04.876668 137274321021824 utils.py:1231] [73900] l2_grads = 1.7020913362503052
+I1203 06:38:04.876783 137274321021824 utils.py:1231] [73900] lr = 0.0003118860007861026
+I1203 06:38:04.876892 137274321021824 utils.py:1231] [73900] uptime = 463674.23923623096
+I1203 06:38:04.876980 137274321021824 utils.py:1231] [73900] examples_seen = 75673600.0
+I1203 06:38:04.877051 137274321021824 utils.py:1231] [73900] progress = 0.6562880207454508
+I1203 06:38:04.877113 137274321021824 utils.py:1231] [73900] epoch = 59.066148285118174
+I1203 06:38:04.877192 137274321021824 utils.py:1231] [73900] img/sec/core = 164.9357126183041
+I1203 06:38:04.877266 137274321021824 utils.py:1231] [73900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.76408218548528
+I1203 06:38:04.877333 137274321021824 utils.py:1231] [73900] core_hours = 128.76408218548528
+I1203 06:38:04.877415 137274321021824 train.py:125] NOTE: Steps:73900/112603 [65.6%]
+Walltime:5d8h47m (0s eval)
+ETA:2d19h26m
+Total train time:8d4h12m
+I1203 06:43:14.761621 137274321021824 utils.py:1231] [73950] l2_params = 270.41673554279754
+I1203 06:43:14.761837 137274321021824 utils.py:1231] [73950] train/loss = 1.8260684460401535
+I1203 06:43:14.761991 137274321021824 utils.py:1231] [73950] l2_grads = 1.9149792194366455
+I1203 06:43:14.762076 137274321021824 utils.py:1231] [73950] lr = 0.0003111769901130252
+I1203 06:43:14.762139 137274321021824 utils.py:1231] [73950] uptime = 463984.124500756
+I1203 06:43:14.762201 137274321021824 utils.py:1231] [73950] examples_seen = 75724800.0
+I1203 06:43:14.762262 137274321021824 utils.py:1231] [73950] progress = 0.6567320586485262
+I1203 06:43:14.762319 137274321021824 utils.py:1231] [73950] epoch = 59.10611184958714
+I1203 06:43:14.762377 137274321021824 utils.py:1231] [73950] img/sec/core = 165.2224415332498
+I1203 06:43:14.762438 137274321021824 utils.py:1231] [73950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.85016142563111
+I1203 06:43:14.762500 137274321021824 utils.py:1231] [73950] core_hours = 128.85016142563111
+I1203 06:43:14.762592 137274321021824 train.py:125] NOTE: Steps:73950/112603 [65.7%]
+Walltime:5d8h53m (0s eval)
+ETA:2d19h21m
+Total train time:8d4h12m
+I1203 06:48:26.532755 137274321021824 utils.py:1231] [74000] l2_params = 270.3268290488044
+I1203 06:48:26.532954 137274321021824 utils.py:1231] [74000] train/loss = 1.9567946791648865
+I1203 06:48:26.533048 137274321021824 utils.py:1231] [74000] l2_grads = 2.006833076477051
+I1203 06:48:26.533104 137274321021824 utils.py:1231] [74000] lr = 0.00031046842200229985
+I1203 06:48:26.533154 137274321021824 utils.py:1231] [74000] uptime = 464295.89551667497
+I1203 06:48:26.533204 137274321021824 utils.py:1231] [74000] examples_seen = 75776000.0
+I1203 06:48:26.533252 137274321021824 utils.py:1231] [74000] progress = 0.6571760965516017
+I1203 06:48:26.533300 137274321021824 utils.py:1231] [74000] epoch = 59.14607541405609
+I1203 06:48:26.533350 137274321021824 utils.py:1231] [74000] img/sec/core = 164.2230912616457
+I1203 06:48:26.533406 137274321021824 utils.py:1231] [74000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 128.9367644856086
+I1203 06:48:26.533455 137274321021824 utils.py:1231] [74000] core_hours = 128.9367644856086
+I1203 06:48:26.533514 137274321021824 train.py:125] NOTE: Steps:74000/112603 [65.7%]
+Walltime:5d8h58m (0s eval)
+ETA:2d19h15m
+Total train time:8d4h12m
+I1203 06:53:36.866554 137274321021824 utils.py:1231] [74050] l2_params = 270.2432644857802
+I1203 06:53:36.866779 137274321021824 utils.py:1231] [74050] train/loss = 3.173198699951172
+I1203 06:53:36.866892 137274321021824 utils.py:1231] [74050] l2_grads = 1.7898838520050049
+I1203 06:53:36.866985 137274321021824 utils.py:1231] [74050] lr = 0.00030976029811466524
+I1203 06:53:36.867044 137274321021824 utils.py:1231] [74050] uptime = 464606.22940551996
+I1203 06:53:36.867097 137274321021824 utils.py:1231] [74050] examples_seen = 75827200.0
+I1203 06:53:36.867147 137274321021824 utils.py:1231] [74050] progress = 0.657620134454677
+I1203 06:53:36.867196 137274321021824 utils.py:1231] [74050] epoch = 59.18603897852505
+I1203 06:53:36.867249 137274321021824 utils.py:1231] [74050] img/sec/core = 164.98359296355605
+I1203 06:53:36.867308 137274321021824 utils.py:1231] [74050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.0229683436211
+I1203 06:53:36.867361 137274321021824 utils.py:1231] [74050] core_hours = 129.0229683436211
+I1203 06:53:36.867423 137274321021824 train.py:125] NOTE: Steps:74050/112603 [65.8%]
+Walltime:5d9h3m (0s eval)
+ETA:2d19h10m
+Total train time:8d4h12m
+I1203 06:58:48.647171 137274321021824 utils.py:1231] [74100] l2_params = 270.15788424589425
+I1203 06:58:48.647382 137274321021824 utils.py:1231] [74100] train/loss = 2.043710246682167
+I1203 06:58:48.647485 137274321021824 utils.py:1231] [74100] l2_grads = 1.9963022470474243
+I1203 06:58:48.647545 137274321021824 utils.py:1231] [74100] lr = 0.00030905262010981764
+I1203 06:58:48.647596 137274321021824 utils.py:1231] [74100] uptime = 464918.00995763997
+I1203 06:58:48.647654 137274321021824 utils.py:1231] [74100] examples_seen = 75878400.0
+I1203 06:58:48.647702 137274321021824 utils.py:1231] [74100] progress = 0.6580641723577525
+I1203 06:58:48.647749 137274321021824 utils.py:1231] [74100] epoch = 59.226002542994
+I1203 06:58:48.647800 137274321021824 utils.py:1231] [74100] img/sec/core = 164.21806829148477
+I1203 06:58:48.647858 137274321021824 utils.py:1231] [74100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.1095740525433
+I1203 06:58:48.647917 137274321021824 utils.py:1231] [74100] core_hours = 129.1095740525433
+I1203 06:58:48.647979 137274321021824 train.py:125] NOTE: Steps:74100/112603 [65.8%]
+Walltime:5d9h8m (0s eval)
+ETA:2d19h5m
+Total train time:8d4h12m
+I1203 07:03:58.057565 137274321021824 utils.py:1231] [74150] l2_params = 270.06955337439797
+I1203 07:03:58.057827 137274321021824 utils.py:1231] [74150] train/loss = 3.0142272114753723
+I1203 07:03:58.057960 137274321021824 utils.py:1231] [74150] l2_grads = 1.7665907144546509
+I1203 07:03:58.058037 137274321021824 utils.py:1231] [74150] lr = 0.00030834538964640977
+I1203 07:03:58.058095 137274321021824 utils.py:1231] [74150] uptime = 465227.420457476
+I1203 07:03:58.058167 137274321021824 utils.py:1231] [74150] examples_seen = 75929600.0
+I1203 07:03:58.058224 137274321021824 utils.py:1231] [74150] progress = 0.6585082102608278
+I1203 07:03:58.058273 137274321021824 utils.py:1231] [74150] epoch = 59.26596610746296
+I1203 07:03:58.058349 137274321021824 utils.py:1231] [74150] img/sec/core = 165.47596163392504
+I1203 07:03:58.058409 137274321021824 utils.py:1231] [74150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.19552141360887
+I1203 07:03:58.058460 137274321021824 utils.py:1231] [74150] core_hours = 129.19552141360887
+I1203 07:03:58.058521 137274321021824 train.py:125] NOTE: Steps:74150/112603 [65.9%]
+Walltime:5d9h13m (0s eval)
+ETA:2d19h0m
+Total train time:8d4h11m
+I1203 07:09:09.834354 137274321021824 utils.py:1231] [74200] l2_params = 269.9802608831114
+I1203 07:09:09.834605 137274321021824 utils.py:1231] [74200] train/loss = 4.104240596294403
+I1203 07:09:09.834732 137274321021824 utils.py:1231] [74200] l2_grads = 1.8387714624404907
+I1203 07:09:09.834818 137274321021824 utils.py:1231] [74200] lr = 0.00030763860838204407
+I1203 07:09:09.834879 137274321021824 utils.py:1231] [74200] uptime = 465539.197240746
+I1203 07:09:09.834967 137274321021824 utils.py:1231] [74200] examples_seen = 75980800.0
+I1203 07:09:09.835035 137274321021824 utils.py:1231] [74200] progress = 0.6589522481639033
+I1203 07:09:09.835109 137274321021824 utils.py:1231] [74200] epoch = 59.30592967193192
+I1203 07:09:09.835175 137274321021824 utils.py:1231] [74200] img/sec/core = 164.22005340808144
+I1203 07:09:09.835233 137274321021824 utils.py:1231] [74200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.28212607562833
+I1203 07:09:09.835300 137274321021824 utils.py:1231] [74200] core_hours = 129.28212607562833
+I1203 07:09:09.835361 137274321021824 train.py:125] NOTE: Steps:74200/112603 [65.9%]
+Walltime:5d9h18m (0s eval)
+ETA:2d18h54m
+Total train time:8d4h11m
+I1203 07:14:21.615260 137274321021824 utils.py:1231] [74250] l2_params = 269.89661816351963
+I1203 07:14:21.615464 137274321021824 utils.py:1231] [74250] train/loss = 1.950803741812706
+I1203 07:14:21.615583 137274321021824 utils.py:1231] [74250] l2_grads = 1.8831815719604492
+I1203 07:14:21.615653 137274321021824 utils.py:1231] [74250] lr = 0.00030693227797327064
+I1203 07:14:21.615709 137274321021824 utils.py:1231] [74250] uptime = 465850.97807156696
+I1203 07:14:21.615766 137274321021824 utils.py:1231] [74250] examples_seen = 76032000.0
+I1203 07:14:21.615821 137274321021824 utils.py:1231] [74250] progress = 0.6593962860669786
+I1203 07:14:21.615899 137274321021824 utils.py:1231] [74250] epoch = 59.345893236400876
+I1203 07:14:21.615958 137274321021824 utils.py:1231] [74250] img/sec/core = 164.2179214969149
+I1203 07:14:21.616026 137274321021824 utils.py:1231] [74250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.3687318619675
+I1203 07:14:21.616077 137274321021824 utils.py:1231] [74250] core_hours = 129.3687318619675
+I1203 07:14:21.616151 137274321021824 train.py:125] NOTE: Steps:74250/112603 [65.9%]
+Walltime:5d9h24m (0s eval)
+ETA:2d18h49m
+Total train time:8d4h11m
+I1203 07:19:30.649993 137274321021824 utils.py:1231] [74300] l2_params = 269.81012695394253
+I1203 07:19:30.650205 137274321021824 utils.py:1231] [74300] train/loss = 1.9064565151929855
+I1203 07:19:30.650300 137274321021824 utils.py:1231] [74300] l2_grads = 1.8971145153045654
+I1203 07:19:30.650366 137274321021824 utils.py:1231] [74300] lr = 0.000306226400075583
+I1203 07:19:30.650417 137274321021824 utils.py:1231] [74300] uptime = 466160.012779364
+I1203 07:19:30.650469 137274321021824 utils.py:1231] [74300] examples_seen = 76083200.0
+I1203 07:19:30.650519 137274321021824 utils.py:1231] [74300] progress = 0.6598403239700541
+I1203 07:19:30.650568 137274321021824 utils.py:1231] [74300] epoch = 59.38585680086983
+I1203 07:19:30.650621 137274321021824 utils.py:1231] [74300] img/sec/core = 165.67718352730796
+I1203 07:19:30.650678 137274321021824 utils.py:1231] [74300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.45457483635556
+I1203 07:19:30.650728 137274321021824 utils.py:1231] [74300] core_hours = 129.45457483635556
+I1203 07:19:30.650788 137274321021824 train.py:125] NOTE: Steps:74300/112603 [66.0%]
+Walltime:5d9h29m (0s eval)
+ETA:2d18h44m
+Total train time:8d4h11m
+I1203 07:24:40.329489 137274321021824 utils.py:1231] [74350] l2_params = 269.7263074297186
+I1203 07:24:40.329695 137274321021824 utils.py:1231] [74350] train/loss = 1.9218690097332
+I1203 07:24:40.329812 137274321021824 utils.py:1231] [74350] l2_grads = 1.9628139734268188
+I1203 07:24:40.329884 137274321021824 utils.py:1231] [74350] lr = 0.0003055209763434141
+I1203 07:24:40.329946 137274321021824 utils.py:1231] [74350] uptime = 466469.692306778
+I1203 07:24:40.330015 137274321021824 utils.py:1231] [74350] examples_seen = 76134400.0
+I1203 07:24:40.330073 137274321021824 utils.py:1231] [74350] progress = 0.6602843618731294
+I1203 07:24:40.330129 137274321021824 utils.py:1231] [74350] epoch = 59.42582036533879
+I1203 07:24:40.330187 137274321021824 utils.py:1231] [74350] img/sec/core = 165.33220787164566
+I1203 07:24:40.330254 137274321021824 utils.py:1231] [74350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.54059692730388
+I1203 07:24:40.330310 137274321021824 utils.py:1231] [74350] core_hours = 129.54059692730388
+I1203 07:24:40.330377 137274321021824 train.py:125] NOTE: Steps:74350/112603 [66.0%]
+Walltime:5d9h34m (0s eval)
+ETA:2d18h39m
+Total train time:8d4h11m
+I1203 07:29:48.444474 137274321021824 utils.py:1231] [74400] l2_params = 269.63895469977575
+I1203 07:29:48.444677 137274321021824 utils.py:1231] [74400] train/loss = 2.412098318338394
+I1203 07:29:48.444777 137274321021824 utils.py:1231] [74400] l2_grads = 1.8466891050338745
+I1203 07:29:48.444842 137274321021824 utils.py:1231] [74400] lr = 0.0003048160084301321
+I1203 07:29:48.444899 137274321021824 utils.py:1231] [74400] uptime = 466777.807261037
+I1203 07:29:48.444952 137274321021824 utils.py:1231] [74400] examples_seen = 76185600.0
+I1203 07:29:48.445000 137274321021824 utils.py:1231] [74400] progress = 0.6607283997762049
+I1203 07:29:48.445050 137274321021824 utils.py:1231] [74400] epoch = 59.46578392980775
+I1203 07:29:48.445101 137274321021824 utils.py:1231] [74400] img/sec/core = 166.17174626636015
+I1203 07:29:48.445159 137274321021824 utils.py:1231] [74400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.62618441459804
+I1203 07:29:48.445210 137274321021824 utils.py:1231] [74400] core_hours = 129.62618441459804
+I1203 07:29:48.445275 137274321021824 train.py:125] NOTE: Steps:74400/112603 [66.1%]
+Walltime:5d9h39m (0s eval)
+ETA:2d18h33m
+Total train time:8d4h11m
+I1203 07:35:00.216488 137274321021824 utils.py:1231] [74450] l2_params = 269.550571477557
+I1203 07:35:00.216707 137274321021824 utils.py:1231] [74450] train/loss = 2.0340523570775986
+I1203 07:35:00.216810 137274321021824 utils.py:1231] [74450] l2_grads = 1.9399806261062622
+I1203 07:35:00.216872 137274321021824 utils.py:1231] [74450] lr = 0.0003041114979880368
+I1203 07:35:00.216930 137274321021824 utils.py:1231] [74450] uptime = 467089.579292332
+I1203 07:35:00.216983 137274321021824 utils.py:1231] [74450] examples_seen = 76236800.0
+I1203 07:35:00.217039 137274321021824 utils.py:1231] [74450] progress = 0.6611724376792804
+I1203 07:35:00.217101 137274321021824 utils.py:1231] [74450] epoch = 59.505747494276704
+I1203 07:35:00.217151 137274321021824 utils.py:1231] [74450] img/sec/core = 164.22255642153758
+I1203 07:35:00.217208 137274321021824 utils.py:1231] [74450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.71278775662444
+I1203 07:35:00.217257 137274321021824 utils.py:1231] [74450] core_hours = 129.71278775662444
+I1203 07:35:00.217317 137274321021824 train.py:125] NOTE: Steps:74450/112603 [66.1%]
+Walltime:5d9h44m (0s eval)
+ETA:2d18h28m
+Total train time:8d4h11m
+I1203 07:40:07.102952 137274321021824 utils.py:1231] [74500] l2_params = 269.46992744999005
+I1203 07:40:07.103193 137274321021824 utils.py:1231] [74500] train/loss = 1.988926261663437
+I1203 07:40:07.103328 137274321021824 utils.py:1231] [74500] l2_grads = 1.803831934928894
+I1203 07:40:07.103397 137274321021824 utils.py:1231] [74500] lr = 0.0003034074466683564
+I1203 07:40:07.103463 137274321021824 utils.py:1231] [74500] uptime = 467396.465825337
+I1203 07:40:07.103524 137274321021824 utils.py:1231] [74500] examples_seen = 76288000.0
+I1203 07:40:07.103578 137274321021824 utils.py:1231] [74500] progress = 0.6616164755823557
+I1203 07:40:07.103630 137274321021824 utils.py:1231] [74500] epoch = 59.54571105874566
+I1203 07:40:07.103684 137274321021824 utils.py:1231] [74500] img/sec/core = 166.83690710914232
+I1203 07:40:07.103744 137274321021824 utils.py:1231] [74500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.7980340157925
+I1203 07:40:07.103798 137274321021824 utils.py:1231] [74500] core_hours = 129.7980340157925
+I1203 07:40:07.103858 137274321021824 train.py:125] NOTE: Steps:74500/112603 [66.2%]
+Walltime:5d9h49m (0s eval)
+ETA:2d18h23m
+Total train time:8d4h11m
+I1203 07:45:18.865275 137274321021824 utils.py:1231] [74550] l2_params = 269.3875166134698
+I1203 07:45:18.865530 137274321021824 utils.py:1231] [74550] train/loss = 2.0435429215431213
+I1203 07:45:18.865657 137274321021824 utils.py:1231] [74550] l2_grads = 2.0315210819244385
+I1203 07:45:18.865734 137274321021824 utils.py:1231] [74550] lr = 0.0003027038561212423
+I1203 07:45:18.865793 137274321021824 utils.py:1231] [74550] uptime = 467708.228154924
+I1203 07:45:18.865863 137274321021824 utils.py:1231] [74550] examples_seen = 76339200.0
+I1203 07:45:18.865916 137274321021824 utils.py:1231] [74550] progress = 0.6620605134854312
+I1203 07:45:18.865963 137274321021824 utils.py:1231] [74550] epoch = 59.585674623214615
+I1203 07:45:18.866014 137274321021824 utils.py:1231] [74550] img/sec/core = 164.22766685065403
+I1203 07:45:18.866069 137274321021824 utils.py:1231] [74550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.8846346629
+I1203 07:45:18.866119 137274321021824 utils.py:1231] [74550] core_hours = 129.8846346629
+I1203 07:45:18.866178 137274321021824 train.py:125] NOTE: Steps:74550/112603 [66.2%]
+Walltime:5d9h55m (0s eval)
+ETA:2d18h17m
+Total train time:8d4h11m
+I1203 07:50:27.366382 137274321021824 utils.py:1231] [74600] l2_params = 269.3067223644454
+I1203 07:50:27.366598 137274321021824 utils.py:1231] [74600] train/loss = 2.10373917222023
+I1203 07:50:27.366691 137274321021824 utils.py:1231] [74600] l2_grads = 1.900815725326538
+I1203 07:50:27.366750 137274321021824 utils.py:1231] [74600] lr = 0.00030200072799576623
+I1203 07:50:27.366801 137274321021824 utils.py:1231] [74600] uptime = 468016.729163682
+I1203 07:50:27.366857 137274321021824 utils.py:1231] [74600] examples_seen = 76390400.0
+I1203 07:50:27.366914 137274321021824 utils.py:1231] [74600] progress = 0.6625045513885065
+I1203 07:50:27.366964 137274321021824 utils.py:1231] [74600] epoch = 59.62563818768357
+I1203 07:50:27.367015 137274321021824 utils.py:1231] [74600] img/sec/core = 165.96380091634597
+I1203 07:50:27.367073 137274321021824 utils.py:1231] [74600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 129.97032938755498
+I1203 07:50:27.367128 137274321021824 utils.py:1231] [74600] core_hours = 129.97032938755498
+I1203 07:50:27.367190 137274321021824 train.py:125] NOTE: Steps:74600/112603 [66.3%]
+Walltime:5d10h0m (0s eval)
+ETA:2d18h12m
+Total train time:8d4h11m
+I1203 07:55:36.397648 137274321021824 utils.py:1231] [74650] l2_params = 269.2167733740351
+I1203 07:55:36.397876 137274321021824 utils.py:1231] [74650] train/loss = 3.8584262430667877
+I1203 07:55:36.398002 137274321021824 utils.py:1231] [74650] l2_grads = 1.8016905784606934
+I1203 07:55:36.398065 137274321021824 utils.py:1231] [74650] lr = 0.0003012980639399164
+I1203 07:55:36.398117 137274321021824 utils.py:1231] [74650] uptime = 468325.760479104
+I1203 07:55:36.398175 137274321021824 utils.py:1231] [74650] examples_seen = 76441600.0
+I1203 07:55:36.398224 137274321021824 utils.py:1231] [74650] progress = 0.662948589291582
+I1203 07:55:36.398291 137274321021824 utils.py:1231] [74650] epoch = 59.66560175215253
+I1203 07:55:36.398341 137274321021824 utils.py:1231] [74650] img/sec/core = 165.6790022398788
+I1203 07:55:36.398400 137274321021824 utils.py:1231] [74650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.05617141961667
+I1203 07:55:36.398448 137274321021824 utils.py:1231] [74650] core_hours = 130.05617141961667
+I1203 07:55:36.398508 137274321021824 train.py:125] NOTE: Steps:74650/112603 [66.3%]
+Walltime:5d10h5m (0s eval)
+ETA:2d18h7m
+Total train time:8d4h11m
+I1203 08:00:44.334171 137274321021824 utils.py:1231] [74700] l2_params = 269.13152921424336
+I1203 08:00:44.334378 137274321021824 utils.py:1231] [74700] train/loss = 2.843293070793152
+I1203 08:00:44.334475 137274321021824 utils.py:1231] [74700] l2_grads = 1.8246266841888428
+I1203 08:00:44.334547 137274321021824 utils.py:1231] [74700] lr = 0.00030059586560059263
+I1203 08:00:44.334609 137274321021824 utils.py:1231] [74700] uptime = 468633.696970474
+I1203 08:00:44.334671 137274321021824 utils.py:1231] [74700] examples_seen = 76492800.0
+I1203 08:00:44.334728 137274321021824 utils.py:1231] [74700] progress = 0.6633926271946573
+I1203 08:00:44.334785 137274321021824 utils.py:1231] [74700] epoch = 59.70556531662149
+I1203 08:00:44.334845 137274321021824 utils.py:1231] [74700] img/sec/core = 166.26805018209453
+I1203 08:00:44.334921 137274321021824 utils.py:1231] [74700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.1417093338861
+I1203 08:00:44.334978 137274321021824 utils.py:1231] [74700] core_hours = 130.1417093338861
+I1203 08:00:44.335042 137274321021824 train.py:125] NOTE: Steps:74700/112603 [66.3%]
+Walltime:5d10h10m (0s eval)
+ETA:2d18h2m
+Total train time:8d4h10m
+I1203 08:05:56.113988 137274321021824 utils.py:1231] [74750] l2_params = 269.0416597298072
+I1203 08:05:56.114223 137274321021824 utils.py:1231] [74750] train/loss = 2.7570317685604095
+I1203 08:05:56.114347 137274321021824 utils.py:1231] [74750] l2_grads = 1.7091234922409058
+I1203 08:05:56.114425 137274321021824 utils.py:1231] [74750] lr = 0.0002998941346236038
+I1203 08:05:56.114493 137274321021824 utils.py:1231] [74750] uptime = 468945.476854038
+I1203 08:05:56.114569 137274321021824 utils.py:1231] [74750] examples_seen = 76544000.0
+I1203 08:05:56.114634 137274321021824 utils.py:1231] [74750] progress = 0.6638366650977328
+I1203 08:05:56.114696 137274321021824 utils.py:1231] [74750] epoch = 59.74552888109044
+I1203 08:05:56.114776 137274321021824 utils.py:1231] [74750] img/sec/core = 164.2184204276382
+I1203 08:05:56.114852 137274321021824 utils.py:1231] [74750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.22831485709833
+I1203 08:05:56.114922 137274321021824 utils.py:1231] [74750] core_hours = 130.22831485709833
+I1203 08:05:56.114995 137274321021824 train.py:125] NOTE: Steps:74750/112603 [66.4%]
+Walltime:5d10h15m (0s eval)
+ETA:2d17h56m
+Total train time:8d4h10m
+I1203 08:11:05.367376 137274321021824 utils.py:1231] [74800] l2_params = 268.9566380169362
+I1203 08:11:05.367589 137274321021824 utils.py:1231] [74800] train/loss = 2.21013006567955
+I1203 08:11:05.367699 137274321021824 utils.py:1231] [74800] l2_grads = 1.8550304174423218
+I1203 08:11:05.367775 137274321021824 utils.py:1231] [74800] lr = 0.00029919287265366324
+I1203 08:11:05.367835 137274321021824 utils.py:1231] [74800] uptime = 469254.73019669
+I1203 08:11:05.367904 137274321021824 utils.py:1231] [74800] examples_seen = 76595200.0
+I1203 08:11:05.367966 137274321021824 utils.py:1231] [74800] progress = 0.6642807030008081
+I1203 08:11:05.368019 137274321021824 utils.py:1231] [74800] epoch = 59.7854924455594
+I1203 08:11:05.368076 137274321021824 utils.py:1231] [74800] img/sec/core = 165.56005364707002
+I1203 08:11:05.368149 137274321021824 utils.py:1231] [74800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.31421856339054
+I1203 08:11:05.368206 137274321021824 utils.py:1231] [74800] core_hours = 130.31421856339054
+I1203 08:11:05.368271 137274321021824 train.py:125] NOTE: Steps:74800/112603 [66.4%]
+Walltime:5d10h20m (0s eval)
+ETA:2d17h51m
+Total train time:8d4h10m
+I1203 08:16:15.580620 137274321021824 utils.py:1231] [74850] l2_params = 268.87346654740213
+I1203 08:16:15.580839 137274321021824 utils.py:1231] [74850] train/loss = 1.8427743315696716
+I1203 08:16:15.580952 137274321021824 utils.py:1231] [74850] l2_grads = 1.8755838871002197
+I1203 08:16:15.581037 137274321021824 utils.py:1231] [74850] lr = 0.00029849208133438473
+I1203 08:16:15.581098 137274321021824 utils.py:1231] [74850] uptime = 469564.943459144
+I1203 08:16:15.581157 137274321021824 utils.py:1231] [74850] examples_seen = 76646400.0
+I1203 08:16:15.581214 137274321021824 utils.py:1231] [74850] progress = 0.6647247409038836
+I1203 08:16:15.581274 137274321021824 utils.py:1231] [74850] epoch = 59.825456010028354
+I1203 08:16:15.581331 137274321021824 utils.py:1231] [74850] img/sec/core = 165.04774681449268
+I1203 08:16:15.581396 137274321021824 utils.py:1231] [74850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.4003889140722
+I1203 08:16:15.581453 137274321021824 utils.py:1231] [74850] core_hours = 130.4003889140722
+I1203 08:16:15.581519 137274321021824 train.py:125] NOTE: Steps:74850/112603 [66.5%]
+Walltime:5d10h26m (0s eval)
+ETA:2d17h46m
+Total train time:8d4h10m
+I1203 08:21:26.798803 137274321021824 utils.py:1231] [74900] l2_params = 268.78674817901066
+I1203 08:21:26.799013 137274321021824 utils.py:1231] [74900] train/loss = 2.108561798930168
+I1203 08:21:26.799117 137274321021824 utils.py:1231] [74900] l2_grads = 2.0198633670806885
+I1203 08:21:26.799178 137274321021824 utils.py:1231] [74900] lr = 0.0002977917623082793
+I1203 08:21:26.799231 137274321021824 utils.py:1231] [74900] uptime = 469876.161592984
+I1203 08:21:26.799285 137274321021824 utils.py:1231] [74900] examples_seen = 76697600.0
+I1203 08:21:26.799335 137274321021824 utils.py:1231] [74900] progress = 0.665168778806959
+I1203 08:21:26.799386 137274321021824 utils.py:1231] [74900] epoch = 59.86541957449732
+I1203 08:21:26.799440 137274321021824 utils.py:1231] [74900] img/sec/core = 164.5148352001895
+I1203 08:21:26.799507 137274321021824 utils.py:1231] [74900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.48683839569443
+I1203 08:21:26.799568 137274321021824 utils.py:1231] [74900] core_hours = 130.48683839569443
+I1203 08:21:26.799630 137274321021824 train.py:125] NOTE: Steps:74900/112603 [66.5%]
+Walltime:5d10h31m (0s eval)
+ETA:2d17h41m
+Total train time:8d4h10m
+I1203 08:26:36.637976 137274321021824 utils.py:1231] [74950] l2_params = 268.69973052209286
+I1203 08:26:36.638216 137274321021824 utils.py:1231] [74950] train/loss = 2.0212612748146057
+I1203 08:26:36.638336 137274321021824 utils.py:1231] [74950] l2_grads = 1.8874077796936035
+I1203 08:26:36.638419 137274321021824 utils.py:1231] [74950] lr = 0.00029709191721675137
+I1203 08:26:36.638481 137274321021824 utils.py:1231] [74950] uptime = 470186.00084313
+I1203 08:26:36.638554 137274321021824 utils.py:1231] [74950] examples_seen = 76748800.0
+I1203 08:26:36.638601 137274321021824 utils.py:1231] [74950] progress = 0.6656128167100344
+I1203 08:26:36.638647 137274321021824 utils.py:1231] [74950] epoch = 59.90538313896627
+I1203 08:26:36.638696 137274321021824 utils.py:1231] [74950] img/sec/core = 165.24697879908143
+I1203 08:26:36.638751 137274321021824 utils.py:1231] [74950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.57290485406833
+I1203 08:26:36.638799 137274321021824 utils.py:1231] [74950] core_hours = 130.57290485406833
+I1203 08:26:36.638857 137274321021824 train.py:125] NOTE: Steps:74950/112603 [66.6%]
+Walltime:5d10h36m (0s eval)
+ETA:2d17h35m
+Total train time:8d4h10m
+I1203 08:31:46.637757 137274321021824 utils.py:1231] [75000] l2_params = 268.61501947651993
+I1203 08:31:46.637987 137274321021824 utils.py:1231] [75000] train/loss = 1.8730918318033218
+I1203 08:31:46.638110 137274321021824 utils.py:1231] [75000] l2_grads = 1.984401822090149
+I1203 08:31:46.638178 137274321021824 utils.py:1231] [75000] lr = 0.0002963925477000936
+I1203 08:31:46.638228 137274321021824 utils.py:1231] [75000] uptime = 470496.000589551
+I1203 08:31:46.638282 137274321021824 utils.py:1231] [75000] examples_seen = 76800000.0
+I1203 08:31:46.638336 137274321021824 utils.py:1231] [75000] progress = 0.6660568546131098
+I1203 08:31:46.638384 137274321021824 utils.py:1231] [75000] epoch = 59.94534670343523
+I1203 08:31:46.638432 137274321021824 utils.py:1231] [75000] img/sec/core = 165.16142542409239
+I1203 08:31:46.638487 137274321021824 utils.py:1231] [75000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.65901589474083
+I1203 08:31:46.638536 137274321021824 utils.py:1231] [75000] core_hours = 130.65901589474083
+I1203 08:31:46.638594 137274321021824 train.py:125] NOTE: Steps:75000/112603 [66.6%]
+Walltime:5d10h41m (0s eval)
+ETA:2d17h30m
+Total train time:8d4h10m
+I1203 08:31:46.992579 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:75000/112603 [66.6%]
+Walltime:5d10h41m (0s eval)
+ETA:2d17h30m
+Total train time:8d4h10m
+I1203 08:33:19.207695 137274321021824 utils.py:1231] [75000] val/acc@1 = 0.7108378507653061
+I1203 08:33:19.207941 137274321021824 utils.py:1231] [75000] val/loss = 1.1495065644991642
+I1203 08:33:19.208085 137274321021824 utils.py:1231] [75000] z/secs/eval/val = 92.21526118501788
+I1203 08:33:19.208165 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 92.21526118501788
+I1203 08:38:29.950884 137274321021824 utils.py:1231] [75050] l2_params = 268.5250445111357
+I1203 08:38:29.951156 137274321021824 utils.py:1231] [75050] train/loss = 2.0535203367471695
+I1203 08:38:29.951266 137274321021824 utils.py:1231] [75050] l2_grads = 2.056272268295288
+I1203 08:38:29.951333 137274321021824 utils.py:1231] [75050] lr = 0.0002956936553974843
+I1203 08:38:29.951404 137274321021824 utils.py:1231] [75050] uptime = 470899.313763473
+I1203 08:38:29.951459 137274321021824 utils.py:1231] [75050] examples_seen = 76851200.0
+I1203 08:38:29.951512 137274321021824 utils.py:1231] [75050] progress = 0.6665008925161852
+I1203 08:38:29.951565 137274321021824 utils.py:1231] [75050] epoch = 59.98531026790418
+I1203 08:38:29.951613 137274321021824 utils.py:1231] [75050] img/sec/core = 126.94849390141792
+I1203 08:38:29.951667 137274321021824 utils.py:1231] [75050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.7710473319414
+I1203 08:38:29.951714 137274321021824 utils.py:1231] [75050] core_hours = 130.7710473319414
+I1203 08:38:29.951772 137274321021824 train.py:125] NOTE: Steps:75050/112603 [66.7%]
+Walltime:5d10h48m (0s eval)
+ETA:2d17h26m
+Total train time:8d4h12m
+I1203 08:43:39.639437 137274321021824 utils.py:1231] [75100] l2_params = 268.43319010599777
+I1203 08:43:39.639640 137274321021824 utils.py:1231] [75100] train/loss = 2.407456785440445
+I1203 08:43:39.639752 137274321021824 utils.py:1231] [75100] l2_grads = 1.8616623878479004
+I1203 08:43:39.639827 137274321021824 utils.py:1231] [75100] lr = 0.0002949952419469839
+I1203 08:43:39.639942 137274321021824 utils.py:1231] [75100] uptime = 471209.00229813
+I1203 08:43:39.640009 137274321021824 utils.py:1231] [75100] examples_seen = 76902400.0
+I1203 08:43:39.640066 137274321021824 utils.py:1231] [75100] progress = 0.6669449304192606
+I1203 08:43:39.640121 137274321021824 utils.py:1231] [75100] epoch = 60.02527383237314
+I1203 08:43:39.640177 137274321021824 utils.py:1231] [75100] img/sec/core = 165.32739921000444
+I1203 08:43:39.640239 137274321021824 utils.py:1231] [75100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.85707192490167
+I1203 08:43:39.640291 137274321021824 utils.py:1231] [75100] core_hours = 130.85707192490167
+I1203 08:43:39.640355 137274321021824 train.py:125] NOTE: Steps:75100/112603 [66.7%]
+Walltime:5d10h53m (0s eval)
+ETA:2d17h20m
+Total train time:8d4h12m
+I1203 08:48:49.534059 137274321021824 utils.py:1231] [75150] l2_params = 268.35766037061444
+I1203 08:48:49.534306 137274321021824 utils.py:1231] [75150] train/loss = 2.020858258008957
+I1203 08:48:49.534420 137274321021824 utils.py:1231] [75150] l2_grads = 1.951766014099121
+I1203 08:48:49.534486 137274321021824 utils.py:1231] [75150] lr = 0.00029429730898552993
+I1203 08:48:49.534538 137274321021824 utils.py:1231] [75150] uptime = 471518.896900387
+I1203 08:48:49.534596 137274321021824 utils.py:1231] [75150] examples_seen = 76953600.0
+I1203 08:48:49.534643 137274321021824 utils.py:1231] [75150] progress = 0.667388968322336
+I1203 08:48:49.534690 137274321021824 utils.py:1231] [75150] epoch = 60.0652373968421
+I1203 08:48:49.534739 137274321021824 utils.py:1231] [75150] img/sec/core = 165.21746305712946
+I1203 08:48:49.534802 137274321021824 utils.py:1231] [75150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 130.94315375886194
+I1203 08:48:49.534851 137274321021824 utils.py:1231] [75150] core_hours = 130.94315375886194
+I1203 08:48:49.534919 137274321021824 train.py:125] NOTE: Steps:75150/112603 [66.7%]
+Walltime:5d10h58m (0s eval)
+ETA:2d17h15m
+Total train time:8d4h12m
+I1203 08:54:01.304636 137274321021824 utils.py:1231] [75200] l2_params = 268.2745386505035
+I1203 08:54:01.304827 137274321021824 utils.py:1231] [75200] train/loss = 4.33872663974762
+I1203 08:54:01.304921 137274321021824 utils.py:1231] [75200] l2_grads = 1.8926200866699219
+I1203 08:54:01.304988 137274321021824 utils.py:1231] [75200] lr = 0.00029359985814893393
+I1203 08:54:01.305036 137274321021824 utils.py:1231] [75200] uptime = 471830.667398602
+I1203 08:54:01.305089 137274321021824 utils.py:1231] [75200] examples_seen = 77004800.0
+I1203 08:54:01.305136 137274321021824 utils.py:1231] [75200] progress = 0.6678330062254114
+I1203 08:54:01.305183 137274321021824 utils.py:1231] [75200] epoch = 60.105200961311056
+I1203 08:54:01.305230 137274321021824 utils.py:1231] [75200] img/sec/core = 164.22336395887575
+I1203 08:54:01.305285 137274321021824 utils.py:1231] [75200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.02975667503276
+I1203 08:54:01.305336 137274321021824 utils.py:1231] [75200] core_hours = 131.02975667503276
+I1203 08:54:01.305394 137274321021824 train.py:125] NOTE: Steps:75200/112603 [66.8%]
+Walltime:5d11h3m (0s eval)
+ETA:2d17h10m
+Total train time:8d4h12m
+I1203 08:59:13.078898 137274321021824 utils.py:1231] [75250] l2_params = 268.1833175113237
+I1203 08:59:13.079189 137274321021824 utils.py:1231] [75250] train/loss = 1.9103724360466003
+I1203 08:59:13.079375 137274321021824 utils.py:1231] [75250] l2_grads = 1.995579481124878
+I1203 08:59:13.079468 137274321021824 utils.py:1231] [75250] lr = 0.0002929028910718774
+I1203 08:59:13.079557 137274321021824 utils.py:1231] [75250] uptime = 472142.441904779
+I1203 08:59:13.079635 137274321021824 utils.py:1231] [75250] examples_seen = 77056000.0
+I1203 08:59:13.079694 137274321021824 utils.py:1231] [75250] progress = 0.6682770441284868
+I1203 08:59:13.079751 137274321021824 utils.py:1231] [75250] epoch = 60.14516452578001
+I1203 08:59:13.079809 137274321021824 utils.py:1231] [75250] img/sec/core = 164.22125281447293
+I1203 08:59:13.079873 137274321021824 utils.py:1231] [75250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.1163607045264
+I1203 08:59:13.079936 137274321021824 utils.py:1231] [75250] core_hours = 131.1163607045264
+I1203 08:59:13.080003 137274321021824 train.py:125] NOTE: Steps:75250/112603 [66.8%]
+Walltime:5d11h9m (0s eval)
+ETA:2d17h5m
+Total train time:8d4h12m
+I1203 09:04:23.519509 137274321021824 utils.py:1231] [75300] l2_params = 268.1008479710232
+I1203 09:04:23.519710 137274321021824 utils.py:1231] [75300] train/loss = 2.976933032274246
+I1203 09:04:23.519815 137274321021824 utils.py:1231] [75300] l2_grads = 1.7642765045166016
+I1203 09:04:23.519892 137274321021824 utils.py:1231] [75300] lr = 0.0002922064093879079
+I1203 09:04:23.519955 137274321021824 utils.py:1231] [75300] uptime = 472452.882315914
+I1203 09:04:23.520015 137274321021824 utils.py:1231] [75300] examples_seen = 77107200.0
+I1203 09:04:23.520080 137274321021824 utils.py:1231] [75300] progress = 0.6687210820315622
+I1203 09:04:23.520135 137274321021824 utils.py:1231] [75300] epoch = 60.185128090248966
+I1203 09:04:23.520188 137274321021824 utils.py:1231] [75300] img/sec/core = 164.9269816800347
+I1203 09:04:23.520250 137274321021824 utils.py:1231] [75300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.20259415206388
+I1203 09:04:23.520315 137274321021824 utils.py:1231] [75300] core_hours = 131.20259415206388
+I1203 09:04:23.520412 137274321021824 train.py:125] NOTE: Steps:75300/112603 [66.9%]
+Walltime:5d11h14m (0s eval)
+ETA:2d16h59m
+Total train time:8d4h12m
+I1203 09:09:31.728079 137274321021824 utils.py:1231] [75350] l2_params = 268.0127311350826
+I1203 09:09:31.728284 137274321021824 utils.py:1231] [75350] train/loss = 3.2669794261455536
+I1203 09:09:31.728378 137274321021824 utils.py:1231] [75350] l2_grads = 1.8211839199066162
+I1203 09:09:31.728435 137274321021824 utils.py:1231] [75350] lr = 0.0002915104147294353
+I1203 09:09:31.728500 137274321021824 utils.py:1231] [75350] uptime = 472761.09086206596
+I1203 09:09:31.728550 137274321021824 utils.py:1231] [75350] examples_seen = 77158400.0
+I1203 09:09:31.728598 137274321021824 utils.py:1231] [75350] progress = 0.6691651199346377
+I1203 09:09:31.728644 137274321021824 utils.py:1231] [75350] epoch = 60.22509165471793
+I1203 09:09:31.728694 137274321021824 utils.py:1231] [75350] img/sec/core = 166.1212858606178
+I1203 09:09:31.728748 137274321021824 utils.py:1231] [75350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.2882076371061
+I1203 09:09:31.728796 137274321021824 utils.py:1231] [75350] core_hours = 131.2882076371061
+I1203 09:09:31.728854 137274321021824 train.py:125] NOTE: Steps:75350/112603 [66.9%]
+Walltime:5d11h19m (0s eval)
+ETA:2d16h54m
+Total train time:8d4h12m
+I1203 09:14:43.463040 137274321021824 utils.py:1231] [75400] l2_params = 267.92316542452767
+I1203 09:14:43.463252 137274321021824 utils.py:1231] [75400] train/loss = 3.5562689304351807
+I1203 09:14:43.463350 137274321021824 utils.py:1231] [75400] l2_grads = 1.8561397790908813
+I1203 09:14:43.463419 137274321021824 utils.py:1231] [75400] lr = 0.0002908149087277285
+I1203 09:14:43.463477 137274321021824 utils.py:1231] [75400] uptime = 473072.825839045
+I1203 09:14:43.463543 137274321021824 utils.py:1231] [75400] examples_seen = 77209600.0
+I1203 09:14:43.463598 137274321021824 utils.py:1231] [75400] progress = 0.669609157837713
+I1203 09:14:43.463654 137274321021824 utils.py:1231] [75400] epoch = 60.265055219186884
+I1203 09:14:43.463714 137274321021824 utils.py:1231] [75400] img/sec/core = 164.24207670300672
+I1203 09:14:43.463786 137274321021824 utils.py:1231] [75400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.37480068626692
+I1203 09:14:43.463839 137274321021824 utils.py:1231] [75400] core_hours = 131.37480068626692
+I1203 09:14:43.463909 137274321021824 train.py:125] NOTE: Steps:75400/112603 [67.0%]
+Walltime:5d11h24m (0s eval)
+ETA:2d16h49m
+Total train time:8d4h12m
+I1203 09:19:53.705119 137274321021824 utils.py:1231] [75450] l2_params = 267.8343815441906
+I1203 09:19:53.705336 137274321021824 utils.py:1231] [75450] train/loss = 4.202777981758118
+I1203 09:19:53.705480 137274321021824 utils.py:1231] [75450] l2_grads = 1.8384252786636353
+I1203 09:19:53.705565 137274321021824 utils.py:1231] [75450] lr = 0.000290119893012911
+I1203 09:19:53.705646 137274321021824 utils.py:1231] [75450] uptime = 473383.068006979
+I1203 09:19:53.705704 137274321021824 utils.py:1231] [75450] examples_seen = 77260800.0
+I1203 09:19:53.705759 137274321021824 utils.py:1231] [75450] progress = 0.6700531957407885
+I1203 09:19:53.705816 137274321021824 utils.py:1231] [75450] epoch = 60.30501878365584
+I1203 09:19:53.705890 137274321021824 utils.py:1231] [75450] img/sec/core = 165.03236920033726
+I1203 09:19:53.705952 137274321021824 utils.py:1231] [75450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.4609790662486
+I1203 09:19:53.706009 137274321021824 utils.py:1231] [75450] core_hours = 131.4609790662486
+I1203 09:19:53.706072 137274321021824 train.py:125] NOTE: Steps:75450/112603 [67.0%]
+Walltime:5d11h29m (0s eval)
+ETA:2d16h44m
+Total train time:8d4h12m
+I1203 09:25:02.053031 137274321021824 utils.py:1231] [75500] l2_params = 267.74634860491494
+I1203 09:25:02.053274 137274321021824 utils.py:1231] [75500] train/loss = 1.8537571430206299
+I1203 09:25:02.053406 137274321021824 utils.py:1231] [75500] l2_grads = 1.9456568956375122
+I1203 09:25:02.053498 137274321021824 utils.py:1231] [75500] lr = 0.00028942536921395634
+I1203 09:25:02.053578 137274321021824 utils.py:1231] [75500] uptime = 473691.415937545
+I1203 09:25:02.053645 137274321021824 utils.py:1231] [75500] examples_seen = 77312000.0
+I1203 09:25:02.053700 137274321021824 utils.py:1231] [75500] progress = 0.6704972336438638
+I1203 09:25:02.053755 137274321021824 utils.py:1231] [75500] epoch = 60.344982348124795
+I1203 09:25:02.053822 137274321021824 utils.py:1231] [75500] img/sec/core = 166.0461930327141
+I1203 09:25:02.053906 137274321021824 utils.py:1231] [75500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.54663126918362
+I1203 09:25:02.053959 137274321021824 utils.py:1231] [75500] core_hours = 131.54663126918362
+I1203 09:25:02.054023 137274321021824 train.py:125] NOTE: Steps:75500/112603 [67.0%]
+Walltime:5d11h34m (0s eval)
+ETA:2d16h38m
+Total train time:8d4h11m
+I1203 09:30:11.391328 137274321021824 utils.py:1231] [75550] l2_params = 267.6723026882133
+I1203 09:30:11.391528 137274321021824 utils.py:1231] [75550] train/loss = 2.9356725811958313
+I1203 09:30:11.391637 137274321021824 utils.py:1231] [75550] l2_grads = 1.8377643823623657
+I1203 09:30:11.391705 137274321021824 utils.py:1231] [75550] lr = 0.00028873133895868574
+I1203 09:30:11.391767 137274321021824 utils.py:1231] [75550] uptime = 474000.754129253
+I1203 09:30:11.391829 137274321021824 utils.py:1231] [75550] examples_seen = 77363200.0
+I1203 09:30:11.391908 137274321021824 utils.py:1231] [75550] progress = 0.6709412715469393
+I1203 09:30:11.391979 137274321021824 utils.py:1231] [75550] epoch = 60.38494591259375
+I1203 09:30:11.392037 137274321021824 utils.py:1231] [75550] img/sec/core = 165.51464181421093
+I1203 09:30:11.392103 137274321021824 utils.py:1231] [75550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.63255854465805
+I1203 09:30:11.392164 137274321021824 utils.py:1231] [75550] core_hours = 131.63255854465805
+I1203 09:30:11.392244 137274321021824 train.py:125] NOTE: Steps:75550/112603 [67.1%]
+Walltime:5d11h40m (0s eval)
+ETA:2d16h33m
+Total train time:8d4h11m
+I1203 09:35:20.204589 137274321021824 utils.py:1231] [75600] l2_params = 267.5902533729848
+I1203 09:35:20.204842 137274321021824 utils.py:1231] [75600] train/loss = 1.9322237223386765
+I1203 09:35:20.204972 137274321021824 utils.py:1231] [75600] l2_grads = 2.037177562713623
+I1203 09:35:20.205050 137274321021824 utils.py:1231] [75600] lr = 0.0002880378038737637
+I1203 09:35:20.205107 137274321021824 utils.py:1231] [75600] uptime = 474309.567469412
+I1203 09:35:20.205159 137274321021824 utils.py:1231] [75600] examples_seen = 77414400.0
+I1203 09:35:20.205208 137274321021824 utils.py:1231] [75600] progress = 0.6713853094500146
+I1203 09:35:20.205255 137274321021824 utils.py:1231] [75600] epoch = 60.42490947706271
+I1203 09:35:20.205307 137274321021824 utils.py:1231] [75600] img/sec/core = 165.7959464239326
+I1203 09:35:20.205364 137274321021824 utils.py:1231] [75600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.71834002803556
+I1203 09:35:20.205417 137274321021824 utils.py:1231] [75600] core_hours = 131.71834002803556
+I1203 09:35:20.205477 137274321021824 train.py:125] NOTE: Steps:75600/112603 [67.1%]
+Walltime:5d11h45m (0s eval)
+ETA:2d16h28m
+Total train time:8d4h11m
+I1203 09:40:25.792411 137274321021824 utils.py:1231] [75650] l2_params = 267.50454683361244
+I1203 09:40:25.792642 137274321021824 utils.py:1231] [75650] train/loss = 3.915977567434311
+I1203 09:40:25.792799 137274321021824 utils.py:1231] [75650] l2_grads = 1.843523383140564
+I1203 09:40:25.792876 137274321021824 utils.py:1231] [75650] lr = 0.00028734476558469376
+I1203 09:40:25.792945 137274321021824 utils.py:1231] [75650] uptime = 474615.155305922
+I1203 09:40:25.793008 137274321021824 utils.py:1231] [75650] examples_seen = 77465600.0
+I1203 09:40:25.793066 137274321021824 utils.py:1231] [75650] progress = 0.6718293473530901
+I1203 09:40:25.793123 137274321021824 utils.py:1231] [75650] epoch = 60.46487304153167
+I1203 09:40:25.793182 137274321021824 utils.py:1231] [75650] img/sec/core = 167.54593567839288
+I1203 09:40:25.793244 137274321021824 utils.py:1231] [75650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.8032255381772
+I1203 09:40:25.793304 137274321021824 utils.py:1231] [75650] core_hours = 131.8032255381772
+I1203 09:40:25.793370 137274321021824 train.py:125] NOTE: Steps:75650/112603 [67.2%]
+Walltime:5d11h50m (0s eval)
+ETA:2d16h23m
+Total train time:8d4h11m
+I1203 09:45:32.383428 137274321021824 utils.py:1231] [75700] l2_params = 267.4183510182891
+I1203 09:45:32.383651 137274321021824 utils.py:1231] [75700] train/loss = 3.2909159064292908
+I1203 09:45:32.383790 137274321021824 utils.py:1231] [75700] l2_grads = 1.8307504653930664
+I1203 09:45:32.383874 137274321021824 utils.py:1231] [75700] lr = 0.00028665222571581575
+I1203 09:45:32.383969 137274321021824 utils.py:1231] [75700] uptime = 474921.746325408
+I1203 09:45:32.384031 137274321021824 utils.py:1231] [75700] examples_seen = 77516800.0
+I1203 09:45:32.384090 137274321021824 utils.py:1231] [75700] progress = 0.6722733852561654
+I1203 09:45:32.384145 137274321021824 utils.py:1231] [75700] epoch = 60.50483660600062
+I1203 09:45:32.384204 137274321021824 utils.py:1231] [75700] img/sec/core = 166.99771599909764
+I1203 09:45:32.384268 137274321021824 utils.py:1231] [75700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.88838971025666
+I1203 09:45:32.384327 137274321021824 utils.py:1231] [75700] core_hours = 131.88838971025666
+I1203 09:45:32.384399 137274321021824 train.py:125] NOTE: Steps:75700/112603 [67.2%]
+Walltime:5d11h55m (0s eval)
+ETA:2d16h17m
+Total train time:8d4h11m
+I1203 09:50:35.048602 137274321021824 utils.py:1231] [75750] l2_params = 267.33401966045267
+I1203 09:50:35.048819 137274321021824 utils.py:1231] [75750] train/loss = 1.996830940246582
+I1203 09:50:35.048918 137274321021824 utils.py:1231] [75750] l2_grads = 2.109447717666626
+I1203 09:50:35.048991 137274321021824 utils.py:1231] [75750] lr = 0.0002859601858903004
+I1203 09:50:35.049063 137274321021824 utils.py:1231] [75750] uptime = 475224.411424996
+I1203 09:50:35.049147 137274321021824 utils.py:1231] [75750] examples_seen = 77568000.0
+I1203 09:50:35.049203 137274321021824 utils.py:1231] [75750] progress = 0.6727174231592409
+I1203 09:50:35.049256 137274321021824 utils.py:1231] [75750] epoch = 60.54480017046958
+I1203 09:50:35.049313 137274321021824 utils.py:1231] [75750] img/sec/core = 169.16387145295977
+I1203 09:50:35.049367 137274321021824 utils.py:1231] [75750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 131.9724633490311
+I1203 09:50:35.049417 137274321021824 utils.py:1231] [75750] core_hours = 131.9724633490311
+I1203 09:50:35.049479 137274321021824 train.py:125] NOTE: Steps:75750/112603 [67.3%]
+Walltime:5d12h0m (0s eval)
+ETA:2d16h12m
+Total train time:8d4h11m
+I1203 09:55:46.833423 137274321021824 utils.py:1231] [75800] l2_params = 267.24697387601685
+I1203 09:55:46.833627 137274321021824 utils.py:1231] [75800] train/loss = 2.2860490679740906
+I1203 09:55:46.833758 137274321021824 utils.py:1231] [75800] l2_grads = 1.8880568742752075
+I1203 09:55:46.833841 137274321021824 utils.py:1231] [75800] lr = 0.0002852686477301473
+I1203 09:55:46.833917 137274321021824 utils.py:1231] [75800] uptime = 475536.19627171196
+I1203 09:55:46.834003 137274321021824 utils.py:1231] [75800] examples_seen = 77619200.0
+I1203 09:55:46.834084 137274321021824 utils.py:1231] [75800] progress = 0.6731614610623162
+I1203 09:55:46.834166 137274321021824 utils.py:1231] [75800] epoch = 60.584763734938534
+I1203 09:55:46.834236 137274321021824 utils.py:1231] [75800] img/sec/core = 164.21580631415625
+I1203 09:55:46.834303 137274321021824 utils.py:1231] [75800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.05907025089667
+I1203 09:55:46.834363 137274321021824 utils.py:1231] [75800] core_hours = 132.05907025089667
+I1203 09:55:46.834431 137274321021824 train.py:125] NOTE: Steps:75800/112603 [67.3%]
+Walltime:5d12h5m (0s eval)
+ETA:2d16h7m
+Total train time:8d4h10m
+I1203 10:00:54.741981 137274321021824 utils.py:1231] [75850] l2_params = 267.1633207899468
+I1203 10:00:54.742176 137274321021824 utils.py:1231] [75850] train/loss = 1.900493398308754
+I1203 10:00:54.742287 137274321021824 utils.py:1231] [75850] l2_grads = 2.0411605834960938
+I1203 10:00:54.742354 137274321021824 utils.py:1231] [75850] lr = 0.0002845776128561796
+I1203 10:00:54.742429 137274321021824 utils.py:1231] [75850] uptime = 475844.104790925
+I1203 10:00:54.742497 137274321021824 utils.py:1231] [75850] examples_seen = 77670400.0
+I1203 10:00:54.742551 137274321021824 utils.py:1231] [75850] progress = 0.6736054989653917
+I1203 10:00:54.742619 137274321021824 utils.py:1231] [75850] epoch = 60.624727299407496
+I1203 10:00:54.742687 137274321021824 utils.py:1231] [75850] img/sec/core = 166.2831549151674
+I1203 10:00:54.742749 137274321021824 utils.py:1231] [75850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.1446003951225
+I1203 10:00:54.742813 137274321021824 utils.py:1231] [75850] core_hours = 132.1446003951225
+I1203 10:00:54.742893 137274321021824 train.py:125] NOTE: Steps:75850/112603 [67.4%]
+Walltime:5d12h10m (0s eval)
+ETA:2d16h1m
+Total train time:8d4h10m
+I1203 10:06:06.525357 137274321021824 utils.py:1231] [75900] l2_params = 267.08343210528466
+I1203 10:06:06.525602 137274321021824 utils.py:1231] [75900] train/loss = 1.9387026280164719
+I1203 10:06:06.525730 137274321021824 utils.py:1231] [75900] l2_grads = 1.9447354078292847
+I1203 10:06:06.525810 137274321021824 utils.py:1231] [75900] lr = 0.00028388708288804135
+I1203 10:06:06.525870 137274321021824 utils.py:1231] [75900] uptime = 476155.888231439
+I1203 10:06:06.525931 137274321021824 utils.py:1231] [75900] examples_seen = 77721600.0
+I1203 10:06:06.525981 137274321021824 utils.py:1231] [75900] progress = 0.6740495368684671
+I1203 10:06:06.526030 137274321021824 utils.py:1231] [75900] epoch = 60.66469086387645
+I1203 10:06:06.526083 137274321021824 utils.py:1231] [75900] img/sec/core = 164.21654695833868
+I1203 10:06:06.526139 137274321021824 utils.py:1231] [75900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.23120690637637
+I1203 10:06:06.526200 137274321021824 utils.py:1231] [75900] core_hours = 132.23120690637637
+I1203 10:06:06.526261 137274321021824 train.py:125] NOTE: Steps:75900/112603 [67.4%]
+Walltime:5d12h15m (0s eval)
+ETA:2d15h56m
+Total train time:8d4h10m
+I1203 10:11:18.308457 137274321021824 utils.py:1231] [75950] l2_params = 267.0072628634717
+I1203 10:11:18.308718 137274321021824 utils.py:1231] [75950] train/loss = 2.4423701763153076
+I1203 10:11:18.308842 137274321021824 utils.py:1231] [75950] l2_grads = 1.963035225868225
+I1203 10:11:18.308912 137274321021824 utils.py:1231] [75950] lr = 0.0002831970594441926
+I1203 10:11:18.308965 137274321021824 utils.py:1231] [75950] uptime = 476467.671327341
+I1203 10:11:18.309018 137274321021824 utils.py:1231] [75950] examples_seen = 77772800.0
+I1203 10:11:18.309066 137274321021824 utils.py:1231] [75950] progress = 0.6744935747715425
+I1203 10:11:18.309114 137274321021824 utils.py:1231] [75950] epoch = 60.70465442834541
+I1203 10:11:18.309167 137274321021824 utils.py:1231] [75950] img/sec/core = 164.2167284659081
+I1203 10:11:18.309224 137274321021824 utils.py:1231] [75950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.3178133219047
+I1203 10:11:18.309274 137274321021824 utils.py:1231] [75950] core_hours = 132.3178133219047
+I1203 10:11:18.309336 137274321021824 train.py:125] NOTE: Steps:75950/112603 [67.4%]
+Walltime:5d12h21m (0s eval)
+ETA:2d15h51m
+Total train time:8d4h10m
+I1203 10:16:26.509126 137274321021824 utils.py:1231] [76000] l2_params = 266.91964379304613
+I1203 10:16:26.509313 137274321021824 utils.py:1231] [76000] train/loss = 1.913922980427742
+I1203 10:16:26.509410 137274321021824 utils.py:1231] [76000] l2_grads = 1.9367018938064575
+I1203 10:16:26.509478 137274321021824 utils.py:1231] [76000] lr = 0.0002825075441419067
+I1203 10:16:26.509528 137274321021824 utils.py:1231] [76000] uptime = 476775.871890223
+I1203 10:16:26.509577 137274321021824 utils.py:1231] [76000] examples_seen = 77824000.0
+I1203 10:16:26.509624 137274321021824 utils.py:1231] [76000] progress = 0.6749376126746179
+I1203 10:16:26.509670 137274321021824 utils.py:1231] [76000] epoch = 60.74461799281436
+I1203 10:16:26.509719 137274321021824 utils.py:1231] [76000] img/sec/core = 166.1255888737689
+I1203 10:16:26.509772 137274321021824 utils.py:1231] [76000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.40342458937195
+I1203 10:16:26.509820 137274321021824 utils.py:1231] [76000] core_hours = 132.40342458937195
+I1203 10:16:26.509879 137274321021824 train.py:125] NOTE: Steps:76000/112603 [67.5%]
+Walltime:5d12h26m (0s eval)
+ETA:2d15h46m
+Total train time:8d4h10m
+I1203 10:21:35.816979 137274321021824 utils.py:1231] [76050] l2_params = 266.84329049052644
+I1203 10:21:35.817214 137274321021824 utils.py:1231] [76050] train/loss = 1.9949515014886856
+I1203 10:21:35.817319 137274321021824 utils.py:1231] [76050] l2_grads = 2.0465118885040283
+I1203 10:21:35.817408 137274321021824 utils.py:1231] [76050] lr = 0.0002818185385972662
+I1203 10:21:35.817499 137274321021824 utils.py:1231] [76050] uptime = 477085.179859493
+I1203 10:21:35.817563 137274321021824 utils.py:1231] [76050] examples_seen = 77875200.0
+I1203 10:21:35.817620 137274321021824 utils.py:1231] [76050] progress = 0.6753816505776933
+I1203 10:21:35.817677 137274321021824 utils.py:1231] [76050] epoch = 60.78458155728332
+I1203 10:21:35.817738 137274321021824 utils.py:1231] [76050] img/sec/core = 165.53081422648214
+I1203 10:21:35.817797 137274321021824 utils.py:1231] [76050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.48934346972473
+I1203 10:21:35.817853 137274321021824 utils.py:1231] [76050] core_hours = 132.48934346972473
+I1203 10:21:35.817931 137274321021824 train.py:125] NOTE: Steps:76050/112603 [67.5%]
+Walltime:5d12h31m (0s eval)
+ETA:2d15h40m
+Total train time:8d4h10m
+I1203 10:26:44.604802 137274321021824 utils.py:1231] [76100] l2_params = 266.7567032699772
+I1203 10:26:44.605048 137274321021824 utils.py:1231] [76100] train/loss = 2.598233699798584
+I1203 10:26:44.605183 137274321021824 utils.py:1231] [76100] l2_grads = 1.9678094387054443
+I1203 10:26:44.605263 137274321021824 utils.py:1231] [76100] lr = 0.0002811300444251584
+I1203 10:26:44.605330 137274321021824 utils.py:1231] [76100] uptime = 477393.967692163
+I1203 10:26:44.605401 137274321021824 utils.py:1231] [76100] examples_seen = 77926400.0
+I1203 10:26:44.605463 137274321021824 utils.py:1231] [76100] progress = 0.6758256884807687
+I1203 10:26:44.605524 137274321021824 utils.py:1231] [76100] epoch = 60.82454512175228
+I1203 10:26:44.605597 137274321021824 utils.py:1231] [76100] img/sec/core = 165.80964203570863
+I1203 10:26:44.605659 137274321021824 utils.py:1231] [76100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.57511786768862
+I1203 10:26:44.605714 137274321021824 utils.py:1231] [76100] core_hours = 132.57511786768862
+I1203 10:26:44.605780 137274321021824 train.py:125] NOTE: Steps:76100/112603 [67.6%]
+Walltime:5d12h36m (0s eval)
+ETA:2d15h35m
+Total train time:8d4h10m
+I1203 10:31:54.073097 137274321021824 utils.py:1231] [76150] l2_params = 266.67595552409347
+I1203 10:31:54.073333 137274321021824 utils.py:1231] [76150] train/loss = 2.6897529661655426
+I1203 10:31:54.073450 137274321021824 utils.py:1231] [76150] l2_grads = 1.8859851360321045
+I1203 10:31:54.073530 137274321021824 utils.py:1231] [76150] lr = 0.00028044206323927227
+I1203 10:31:54.073613 137274321021824 utils.py:1231] [76150] uptime = 477703.435966298
+I1203 10:31:54.073691 137274321021824 utils.py:1231] [76150] examples_seen = 77977600.0
+I1203 10:31:54.073751 137274321021824 utils.py:1231] [76150] progress = 0.6762697263838441
+I1203 10:31:54.073817 137274321021824 utils.py:1231] [76150] epoch = 60.864508686221235
+I1203 10:31:54.073875 137274321021824 utils.py:1231] [76150] img/sec/core = 165.44506910479043
+I1203 10:31:54.073947 137274321021824 utils.py:1231] [76150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.66108127717055
+I1203 10:31:54.074013 137274321021824 utils.py:1231] [76150] core_hours = 132.66108127717055
+I1203 10:31:54.074093 137274321021824 train.py:125] NOTE: Steps:76150/112603 [67.6%]
+Walltime:5d12h41m (0s eval)
+ETA:2d15h30m
+Total train time:8d4h10m
+I1203 10:37:01.708034 137274321021824 utils.py:1231] [76200] l2_params = 266.58585232557607
+I1203 10:37:01.708227 137274321021824 utils.py:1231] [76200] train/loss = 3.8871826827526093
+I1203 10:37:01.708328 137274321021824 utils.py:1231] [76200] l2_grads = 1.8456522226333618
+I1203 10:37:01.708398 137274321021824 utils.py:1231] [76200] lr = 0.0002797545966520944
+I1203 10:37:01.708465 137274321021824 utils.py:1231] [76200] uptime = 478011.070823442
+I1203 10:37:01.708534 137274321021824 utils.py:1231] [76200] examples_seen = 78028800.0
+I1203 10:37:01.708593 137274321021824 utils.py:1231] [76200] progress = 0.6767137642869195
+I1203 10:37:01.708657 137274321021824 utils.py:1231] [76200] epoch = 60.90447225069019
+I1203 10:37:01.708725 137274321021824 utils.py:1231] [76200] img/sec/core = 166.43107505866467
+I1203 10:37:01.708794 137274321021824 utils.py:1231] [76200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.746535404155
+I1203 10:37:01.708862 137274321021824 utils.py:1231] [76200] core_hours = 132.746535404155
+I1203 10:37:01.708941 137274321021824 train.py:125] NOTE: Steps:76200/112603 [67.7%]
+Walltime:5d12h46m (0s eval)
+ETA:2d15h25m
+Total train time:8d4h10m
+I1203 10:42:09.110175 137274321021824 utils.py:1231] [76250] l2_params = 266.50185491066657
+I1203 10:42:09.110383 137274321021824 utils.py:1231] [76250] train/loss = 2.1101842373609543
+I1203 10:42:09.110486 137274321021824 utils.py:1231] [76250] l2_grads = 2.023087978363037
+I1203 10:42:09.110554 137274321021824 utils.py:1231] [76250] lr = 0.000279067646274905
+I1203 10:42:09.110617 137274321021824 utils.py:1231] [76250] uptime = 478318.472978751
+I1203 10:42:09.110677 137274321021824 utils.py:1231] [76250] examples_seen = 78080000.0
+I1203 10:42:09.110732 137274321021824 utils.py:1231] [76250] progress = 0.6771578021899949
+I1203 10:42:09.110787 137274321021824 utils.py:1231] [76250] epoch = 60.944435815159146
+I1203 10:42:09.110858 137274321021824 utils.py:1231] [76250] img/sec/core = 166.5570625181092
+I1203 10:42:09.110922 137274321021824 utils.py:1231] [76250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.83192489174084
+I1203 10:42:09.110984 137274321021824 utils.py:1231] [76250] core_hours = 132.83192489174084
+I1203 10:42:09.111049 137274321021824 train.py:125] NOTE: Steps:76250/112603 [67.7%]
+Walltime:5d12h51m (0s eval)
+ETA:2d15h19m
+Total train time:8d4h9m
+I1203 10:47:16.575798 137274321021824 utils.py:1231] [76300] l2_params = 266.4245878880956
+I1203 10:47:16.575993 137274321021824 utils.py:1231] [76300] train/loss = 2.0179738253355026
+I1203 10:47:16.576097 137274321021824 utils.py:1231] [76300] l2_grads = 1.8808118104934692
+I1203 10:47:16.576170 137274321021824 utils.py:1231] [76300] lr = 0.0002783812137177752
+I1203 10:47:16.576224 137274321021824 utils.py:1231] [76300] uptime = 478625.938585406
+I1203 10:47:16.576277 137274321021824 utils.py:1231] [76300] examples_seen = 78131200.0
+I1203 10:47:16.576326 137274321021824 utils.py:1231] [76300] progress = 0.6776018400930703
+I1203 10:47:16.576373 137274321021824 utils.py:1231] [76300] epoch = 60.9843993796281
+I1203 10:47:16.576422 137274321021824 utils.py:1231] [76300] img/sec/core = 166.52269031655075
+I1203 10:47:16.576477 137274321021824 utils.py:1231] [76300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 132.91733200470054
+I1203 10:47:16.576526 137274321021824 utils.py:1231] [76300] core_hours = 132.91733200470054
+I1203 10:47:16.576585 137274321021824 train.py:125] NOTE: Steps:76300/112603 [67.8%]
+Walltime:5d12h57m (0s eval)
+ETA:2d15h14m
+Total train time:8d4h9m
+I1203 10:52:23.760295 137274321021824 utils.py:1231] [76350] l2_params = 266.34255464292977
+I1203 10:52:23.760492 137274321021824 utils.py:1231] [76350] train/loss = 4.007927417755127
+I1203 10:52:23.760595 137274321021824 utils.py:1231] [76350] l2_grads = 1.872867465019226
+I1203 10:52:23.760671 137274321021824 utils.py:1231] [76350] lr = 0.00027769530058956133
+I1203 10:52:23.760738 137274321021824 utils.py:1231] [76350] uptime = 478933.123096902
+I1203 10:52:23.760801 137274321021824 utils.py:1231] [76350] examples_seen = 78182400.0
+I1203 10:52:23.760860 137274321021824 utils.py:1231] [76350] progress = 0.6780458779961458
+I1203 10:52:23.760928 137274321021824 utils.py:1231] [76350] epoch = 61.024362944097064
+I1203 10:52:23.760999 137274321021824 utils.py:1231] [76350] img/sec/core = 166.675070141578
+I1203 10:52:23.761067 137274321021824 utils.py:1231] [76350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.00266103567165
+I1203 10:52:23.761124 137274321021824 utils.py:1231] [76350] core_hours = 133.00266103567165
+I1203 10:52:23.761191 137274321021824 train.py:125] NOTE: Steps:76350/112603 [67.8%]
+Walltime:5d13h2m (0s eval)
+ETA:2d15h9m
+Total train time:8d4h9m
+I1203 10:57:30.945310 137274321021824 utils.py:1231] [76400] l2_params = 266.26272966182825
+I1203 10:57:30.945551 137274321021824 utils.py:1231] [76400] train/loss = 1.9674699753522873
+I1203 10:57:30.945646 137274321021824 utils.py:1231] [76400] l2_grads = 2.0581085681915283
+I1203 10:57:30.945706 137274321021824 utils.py:1231] [76400] lr = 0.0002770099084979035
+I1203 10:57:30.945776 137274321021824 utils.py:1231] [76400] uptime = 479240.30813290697
+I1203 10:57:30.945833 137274321021824 utils.py:1231] [76400] examples_seen = 78233600.0
+I1203 10:57:30.945886 137274321021824 utils.py:1231] [76400] progress = 0.6784899158992211
+I1203 10:57:30.945939 137274321021824 utils.py:1231] [76400] epoch = 61.06432650856602
+I1203 10:57:30.945989 137274321021824 utils.py:1231] [76400] img/sec/core = 166.67478554902283
+I1203 10:57:30.946045 137274321021824 utils.py:1231] [76400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.0879902123397
+I1203 10:57:30.946095 137274321021824 utils.py:1231] [76400] core_hours = 133.0879902123397
+I1203 10:57:30.946154 137274321021824 train.py:125] NOTE: Steps:76400/112603 [67.8%]
+Walltime:5d13h7m (0s eval)
+ETA:2d15h4m
+Total train time:8d4h9m
+I1203 11:02:39.527704 137274321021824 utils.py:1231] [76450] l2_params = 266.17624086058254
+I1203 11:02:39.527933 137274321021824 utils.py:1231] [76450] train/loss = 2.8215459883213043
+I1203 11:02:39.528051 137274321021824 utils.py:1231] [76450] l2_grads = 1.6902507543563843
+I1203 11:02:39.528125 137274321021824 utils.py:1231] [76450] lr = 0.0002763250390492193
+I1203 11:02:39.528185 137274321021824 utils.py:1231] [76450] uptime = 479548.890546807
+I1203 11:02:39.528247 137274321021824 utils.py:1231] [76450] examples_seen = 78284800.0
+I1203 11:02:39.528306 137274321021824 utils.py:1231] [76450] progress = 0.6789339538022966
+I1203 11:02:39.528363 137274321021824 utils.py:1231] [76450] epoch = 61.104290073034974
+I1203 11:02:39.528422 137274321021824 utils.py:1231] [76450] img/sec/core = 165.92001907338587
+I1203 11:02:39.528484 137274321021824 utils.py:1231] [76450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.17370754953416
+I1203 11:02:39.528540 137274321021824 utils.py:1231] [76450] core_hours = 133.17370754953416
+I1203 11:02:39.528615 137274321021824 train.py:125] NOTE: Steps:76450/112603 [67.9%]
+Walltime:5d13h12m (0s eval)
+ETA:2d14h58m
+Total train time:8d4h9m
+I1203 11:07:44.675225 137274321021824 utils.py:1231] [76500] l2_params = 266.0866091574774
+I1203 11:07:44.675429 137274321021824 utils.py:1231] [76500] train/loss = 1.7625106275081635
+I1203 11:07:44.675524 137274321021824 utils.py:1231] [76500] l2_grads = 1.9241104125976562
+I1203 11:07:44.675596 137274321021824 utils.py:1231] [76500] lr = 0.0002756406938487025
+I1203 11:07:44.675658 137274321021824 utils.py:1231] [76500] uptime = 479854.03801917797
+I1203 11:07:44.675717 137274321021824 utils.py:1231] [76500] examples_seen = 78336000.0
+I1203 11:07:44.675774 137274321021824 utils.py:1231] [76500] progress = 0.6793779917053719
+I1203 11:07:44.675839 137274321021824 utils.py:1231] [76500] epoch = 61.14425363750393
+I1203 11:07:44.675916 137274321021824 utils.py:1231] [76500] img/sec/core = 167.78772441461055
+I1203 11:07:44.675983 137274321021824 utils.py:1231] [76500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.25847073630388
+I1203 11:07:44.676039 137274321021824 utils.py:1231] [76500] core_hours = 133.25847073630388
+I1203 11:07:44.676104 137274321021824 train.py:125] NOTE: Steps:76500/112603 [67.9%]
+Walltime:5d13h17m (0s eval)
+ETA:2d14h53m
+Total train time:8d4h9m
+I1203 11:12:56.485143 137274321021824 utils.py:1231] [76550] l2_params = 266.0111881638953
+I1203 11:12:56.485365 137274321021824 utils.py:1231] [76550] train/loss = 4.262731611728668
+I1203 11:12:56.485471 137274321021824 utils.py:1231] [76550] l2_grads = 1.9213600158691406
+I1203 11:12:56.485545 137274321021824 utils.py:1231] [76550] lr = 0.00027495687450031773
+I1203 11:12:56.485613 137274321021824 utils.py:1231] [76550] uptime = 480165.847973822
+I1203 11:12:56.485674 137274321021824 utils.py:1231] [76550] examples_seen = 78387200.0
+I1203 11:12:56.485734 137274321021824 utils.py:1231] [76550] progress = 0.6798220296084474
+I1203 11:12:56.485794 137274321021824 utils.py:1231] [76550] epoch = 61.18421720197289
+I1203 11:12:56.485853 137274321021824 utils.py:1231] [76550] img/sec/core = 164.2025831357921
+I1203 11:12:56.485936 137274321021824 utils.py:1231] [76550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.3450846125939
+I1203 11:12:56.485996 137274321021824 utils.py:1231] [76550] core_hours = 133.3450846125939
+I1203 11:12:56.486064 137274321021824 train.py:125] NOTE: Steps:76550/112603 [68.0%]
+Walltime:5d13h22m (0s eval)
+ETA:2d14h48m
+Total train time:8d4h9m
+I1203 11:18:06.825138 137274321021824 utils.py:1231] [76600] l2_params = 265.9308352854538
+I1203 11:18:06.825341 137274321021824 utils.py:1231] [76600] train/loss = 1.9560069739818573
+I1203 11:18:06.825435 137274321021824 utils.py:1231] [76600] l2_grads = 2.144498586654663
+I1203 11:18:06.825496 137274321021824 utils.py:1231] [76600] lr = 0.00027427358260679696
+I1203 11:18:06.825558 137274321021824 utils.py:1231] [76600] uptime = 480476.187919672
+I1203 11:18:06.825620 137274321021824 utils.py:1231] [76600] examples_seen = 78438400.0
+I1203 11:18:06.825671 137274321021824 utils.py:1231] [76600] progress = 0.6802660675115227
+I1203 11:18:06.825721 137274321021824 utils.py:1231] [76600] epoch = 61.22418076644185
+I1203 11:18:06.825773 137274321021824 utils.py:1231] [76600] img/sec/core = 164.98037292544876
+I1203 11:18:06.825837 137274321021824 utils.py:1231] [76600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.43129015310777
+I1203 11:18:06.825904 137274321021824 utils.py:1231] [76600] core_hours = 133.43129015310777
+I1203 11:18:06.825978 137274321021824 train.py:125] NOTE: Steps:76600/112603 [68.0%]
+Walltime:5d13h27m (0s eval)
+ETA:2d14h42m
+Total train time:8d4h9m
+I1203 11:23:16.803678 137274321021824 utils.py:1231] [76650] l2_params = 265.84317557040896
+I1203 11:23:16.803893 137274321021824 utils.py:1231] [76650] train/loss = 1.7643945664167404
+I1203 11:23:16.803990 137274321021824 utils.py:1231] [76650] l2_grads = 2.05670166015625
+I1203 11:23:16.804074 137274321021824 utils.py:1231] [76650] lr = 0.00027359081976963606
+I1203 11:23:16.804162 137274321021824 utils.py:1231] [76650] uptime = 480786.166518162
+I1203 11:23:16.804225 137274321021824 utils.py:1231] [76650] examples_seen = 78489600.0
+I1203 11:23:16.804275 137274321021824 utils.py:1231] [76650] progress = 0.6807101054145982
+I1203 11:23:16.804330 137274321021824 utils.py:1231] [76650] epoch = 61.2641443309108
+I1203 11:23:16.804385 137274321021824 utils.py:1231] [76650] img/sec/core = 165.17269337112117
+I1203 11:23:16.804442 137274321021824 utils.py:1231] [76650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.517395319355
+I1203 11:23:16.804495 137274321021824 utils.py:1231] [76650] core_hours = 133.517395319355
+I1203 11:23:16.804557 137274321021824 train.py:125] NOTE: Steps:76650/112603 [68.1%]
+Walltime:5d13h33m (0s eval)
+ETA:2d14h37m
+Total train time:8d4h8m
+I1203 11:28:28.528822 137274321021824 utils.py:1231] [76700] l2_params = 265.7552150356374
+I1203 11:28:28.529011 137274321021824 utils.py:1231] [76700] train/loss = 1.9819037020206451
+I1203 11:28:28.529107 137274321021824 utils.py:1231] [76700] l2_grads = 1.9568054676055908
+I1203 11:28:28.529166 137274321021824 utils.py:1231] [76700] lr = 0.00027290858758909065
+I1203 11:28:28.529232 137274321021824 utils.py:1231] [76700] uptime = 481097.89159141097
+I1203 11:28:28.529281 137274321021824 utils.py:1231] [76700] examples_seen = 78540800.0
+I1203 11:28:28.529335 137274321021824 utils.py:1231] [76700] progress = 0.6811541433176735
+I1203 11:28:28.529381 137274321021824 utils.py:1231] [76700] epoch = 61.30410789537976
+I1203 11:28:28.529429 137274321021824 utils.py:1231] [76700] img/sec/core = 164.2472947920635
+I1203 11:28:28.529480 137274321021824 utils.py:1231] [76700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.6039856174797
+I1203 11:28:28.529528 137274321021824 utils.py:1231] [76700] core_hours = 133.6039856174797
+I1203 11:28:28.529585 137274321021824 train.py:125] NOTE: Steps:76700/112603 [68.1%]
+Walltime:5d13h38m (0s eval)
+ETA:2d14h32m
+Total train time:8d4h8m
+I1203 11:33:40.251749 137274321021824 utils.py:1231] [76750] l2_params = 265.67298175628576
+I1203 11:33:40.251993 137274321021824 utils.py:1231] [76750] train/loss = 2.618408203125
+I1203 11:33:40.252135 137274321021824 utils.py:1231] [76750] l2_grads = 1.8693948984146118
+I1203 11:33:40.252217 137274321021824 utils.py:1231] [76750] lr = 0.00027222688766417294
+I1203 11:33:40.252280 137274321021824 utils.py:1231] [76750] uptime = 481409.614641917
+I1203 11:33:40.252341 137274321021824 utils.py:1231] [76750] examples_seen = 78592000.0
+I1203 11:33:40.252397 137274321021824 utils.py:1231] [76750] progress = 0.681598181220749
+I1203 11:33:40.252460 137274321021824 utils.py:1231] [76750] epoch = 61.34407145984871
+I1203 11:33:40.252517 137274321021824 utils.py:1231] [76750] img/sec/core = 164.24836057803745
+I1203 11:33:40.252580 137274321021824 utils.py:1231] [76750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.6905753537314
+I1203 11:33:40.252639 137274321021824 utils.py:1231] [76750] core_hours = 133.6905753537314
+I1203 11:33:40.252701 137274321021824 train.py:125] NOTE: Steps:76750/112603 [68.2%]
+Walltime:5d13h43m (0s eval)
+ETA:2d14h27m
+Total train time:8d4h8m
+I1203 11:38:47.752129 137274321021824 utils.py:1231] [76800] l2_params = 265.58954025183664
+I1203 11:38:47.752330 137274321021824 utils.py:1231] [76800] train/loss = 3.2319932878017426
+I1203 11:38:47.752427 137274321021824 utils.py:1231] [76800] l2_grads = 1.904765009880066
+I1203 11:38:47.752486 137274321021824 utils.py:1231] [76800] lr = 0.00027154572159264815
+I1203 11:38:47.752538 137274321021824 utils.py:1231] [76800] uptime = 481717.114900106
+I1203 11:38:47.752591 137274321021824 utils.py:1231] [76800] examples_seen = 78643200.0
+I1203 11:38:47.752640 137274321021824 utils.py:1231] [76800] progress = 0.6820422191238245
+I1203 11:38:47.752689 137274321021824 utils.py:1231] [76800] epoch = 61.384035024317676
+I1203 11:38:47.752740 137274321021824 utils.py:1231] [76800] img/sec/core = 166.50392523746478
+I1203 11:38:47.752797 137274321021824 utils.py:1231] [76800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.7759920921172
+I1203 11:38:47.752848 137274321021824 utils.py:1231] [76800] core_hours = 133.7759920921172
+I1203 11:38:47.752923 137274321021824 train.py:125] NOTE: Steps:76800/112603 [68.2%]
+Walltime:5d13h48m (0s eval)
+ETA:2d14h21m
+Total train time:8d4h8m
+I1203 11:43:57.005248 137274321021824 utils.py:1231] [76850] l2_params = 265.5123789841478
+I1203 11:43:57.005467 137274321021824 utils.py:1231] [76850] train/loss = 2.8103072941303253
+I1203 11:43:57.005588 137274321021824 utils.py:1231] [76850] l2_grads = 1.7938402891159058
+I1203 11:43:57.005658 137274321021824 utils.py:1231] [76850] lr = 0.0002708650909710289
+I1203 11:43:57.005717 137274321021824 utils.py:1231] [76850] uptime = 482026.36807927
+I1203 11:43:57.005771 137274321021824 utils.py:1231] [76850] examples_seen = 78694400.0
+I1203 11:43:57.005818 137274321021824 utils.py:1231] [76850] progress = 0.6824862570268998
+I1203 11:43:57.005864 137274321021824 utils.py:1231] [76850] epoch = 61.42399858878663
+I1203 11:43:57.005918 137274321021824 utils.py:1231] [76850] img/sec/core = 165.56014117107787
+I1203 11:43:57.005973 137274321021824 utils.py:1231] [76850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.8618957529961
+I1203 11:43:57.006021 137274321021824 utils.py:1231] [76850] core_hours = 133.8618957529961
+I1203 11:43:57.006079 137274321021824 train.py:125] NOTE: Steps:76850/112603 [68.2%]
+Walltime:5d13h53m (0s eval)
+ETA:2d14h16m
+Total train time:8d4h8m
+I1203 11:49:06.449090 137274321021824 utils.py:1231] [76900] l2_params = 265.4315682750591
+I1203 11:49:06.449311 137274321021824 utils.py:1231] [76900] train/loss = 3.0805689990520477
+I1203 11:49:06.449415 137274321021824 utils.py:1231] [76900] l2_grads = 1.7848881483078003
+I1203 11:49:06.449491 137274321021824 utils.py:1231] [76900] lr = 0.0002701849973945738
+I1203 11:49:06.449553 137274321021824 utils.py:1231] [76900] uptime = 482335.811913007
+I1203 11:49:06.449637 137274321021824 utils.py:1231] [76900] examples_seen = 78745600.0
+I1203 11:49:06.449710 137274321021824 utils.py:1231] [76900] progress = 0.6829302949299753
+I1203 11:49:06.449781 137274321021824 utils.py:1231] [76900] epoch = 61.463962153255586
+I1203 11:49:06.449849 137274321021824 utils.py:1231] [76900] img/sec/core = 165.45813623649877
+I1203 11:49:06.449941 137274321021824 utils.py:1231] [76900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 133.9478523734786
+I1203 11:49:06.450047 137274321021824 utils.py:1231] [76900] core_hours = 133.9478523734786
+I1203 11:49:06.450164 137274321021824 train.py:125] NOTE: Steps:76900/112603 [68.3%]
+Walltime:5d13h58m (0s eval)
+ETA:2d14h11m
+Total train time:8d4h8m
+I1203 11:54:17.327556 137274321021824 utils.py:1231] [76950] l2_params = 265.3482105924514
+I1203 11:54:17.327754 137274321021824 utils.py:1231] [76950] train/loss = 1.8413697332143784
+I1203 11:54:17.327848 137274321021824 utils.py:1231] [76950] l2_grads = 2.1104393005371094
+I1203 11:54:17.327916 137274321021824 utils.py:1231] [76950] lr = 0.00026950544245728263
+I1203 11:54:17.327971 137274321021824 utils.py:1231] [76950] uptime = 482646.69033312896
+I1203 11:54:17.328030 137274321021824 utils.py:1231] [76950] examples_seen = 78796800.0
+I1203 11:54:17.328081 137274321021824 utils.py:1231] [76950] progress = 0.6833743328330506
+I1203 11:54:17.328129 137274321021824 utils.py:1231] [76950] epoch = 61.50392571772454
+I1203 11:54:17.328181 137274321021824 utils.py:1231] [76950] img/sec/core = 164.69460948724765
+I1203 11:54:17.328238 137274321021824 utils.py:1231] [76950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.03420749017917
+I1203 11:54:17.328295 137274321021824 utils.py:1231] [76950] core_hours = 134.03420749017917
+I1203 11:54:17.328355 137274321021824 train.py:125] NOTE: Steps:76950/112603 [68.3%]
+Walltime:5d14h4m (0s eval)
+ETA:2d14h6m
+Total train time:8d4h8m
+I1203 11:59:29.106314 137274321021824 utils.py:1231] [77000] l2_params = 265.2582113067332
+I1203 11:59:29.106505 137274321021824 utils.py:1231] [77000] train/loss = 2.0013388991355896
+I1203 11:59:29.106606 137274321021824 utils.py:1231] [77000] l2_grads = 1.9983642101287842
+I1203 11:59:29.106662 137274321021824 utils.py:1231] [77000] lr = 0.0002688264277518926
+I1203 11:59:29.106709 137274321021824 utils.py:1231] [77000] uptime = 482958.469071844
+I1203 11:59:29.106758 137274321021824 utils.py:1231] [77000] examples_seen = 78848000.0
+I1203 11:59:29.106804 137274321021824 utils.py:1231] [77000] progress = 0.6838183707361261
+I1203 11:59:29.106853 137274321021824 utils.py:1231] [77000] epoch = 61.5438892821935
+I1203 11:59:29.106906 137274321021824 utils.py:1231] [77000] img/sec/core = 164.2190234363445
+I1203 11:59:29.106960 137274321021824 utils.py:1231] [77000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.1208126953778
+I1203 11:59:29.107010 137274321021824 utils.py:1231] [77000] core_hours = 134.1208126953778
+I1203 11:59:29.107069 137274321021824 train.py:125] NOTE: Steps:77000/112603 [68.4%]
+Walltime:5d14h9m (0s eval)
+ETA:2d14h0m
+Total train time:8d4h8m
+I1203 12:04:41.245373 137274321021824 utils.py:1231] [77050] l2_params = 265.18194972259755
+I1203 12:04:41.245697 137274321021824 utils.py:1231] [77050] train/loss = 1.8404122442007065
+I1203 12:04:41.245888 137274321021824 utils.py:1231] [77050] l2_grads = 1.979459524154663
+I1203 12:04:41.245980 137274321021824 utils.py:1231] [77050] lr = 0.00026814795486987435
+I1203 12:04:41.246061 137274321021824 utils.py:1231] [77050] uptime = 483270.60841924
+I1203 12:04:41.246136 137274321021824 utils.py:1231] [77050] examples_seen = 78899200.0
+I1203 12:04:41.246221 137274321021824 utils.py:1231] [77050] progress = 0.6842624086392014
+I1203 12:04:41.246336 137274321021824 utils.py:1231] [77050] epoch = 61.58385284666246
+I1203 12:04:41.246448 137274321021824 utils.py:1231] [77050] img/sec/core = 164.02930430634981
+I1203 12:04:41.246555 137274321021824 utils.py:1231] [77050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.20751806965444
+I1203 12:04:41.246662 137274321021824 utils.py:1231] [77050] core_hours = 134.20751806965444
+I1203 12:04:41.246755 137274321021824 train.py:125] NOTE: Steps:77050/112603 [68.4%]
+Walltime:5d14h14m (0s eval)
+ETA:2d13h55m
+Total train time:8d4h8m
+I1203 12:09:53.030913 137274321021824 utils.py:1231] [77100] l2_params = 265.10333560645967
+I1203 12:09:53.031198 137274321021824 utils.py:1231] [77100] train/loss = 1.8756928145885468
+I1203 12:09:53.031453 137274321021824 utils.py:1231] [77100] l2_grads = 2.048893690109253
+I1203 12:09:53.031572 137274321021824 utils.py:1231] [77100] lr = 0.0002674700254014294
+I1203 12:09:53.031684 137274321021824 utils.py:1231] [77100] uptime = 483582.39403211797
+I1203 12:09:53.031776 137274321021824 utils.py:1231] [77100] examples_seen = 78950400.0
+I1203 12:09:53.031902 137274321021824 utils.py:1231] [77100] progress = 0.6847064465422769
+I1203 12:09:53.031981 137274321021824 utils.py:1231] [77100] epoch = 61.623816411131415
+I1203 12:09:53.032069 137274321021824 utils.py:1231] [77100] img/sec/core = 164.2154027807473
+I1203 12:09:53.032155 137274321021824 utils.py:1231] [77100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.29412518434276
+I1203 12:09:53.032221 137274321021824 utils.py:1231] [77100] core_hours = 134.29412518434276
+I1203 12:09:53.032325 137274321021824 train.py:125] NOTE: Steps:77100/112603 [68.5%]
+Walltime:5d14h19m (0s eval)
+ETA:2d13h50m
+Total train time:8d4h8m
+I1203 12:15:03.578957 137274321021824 utils.py:1231] [77150] l2_params = 265.0224253907259
+I1203 12:15:03.579160 137274321021824 utils.py:1231] [77150] train/loss = 2.4341930747032166
+I1203 12:15:03.579263 137274321021824 utils.py:1231] [77150] l2_grads = 1.832735538482666
+I1203 12:15:03.579335 137274321021824 utils.py:1231] [77150] lr = 0.0002667926409354846
+I1203 12:15:03.579398 137274321021824 utils.py:1231] [77150] uptime = 483892.941759212
+I1203 12:15:03.579457 137274321021824 utils.py:1231] [77150] examples_seen = 79001600.0
+I1203 12:15:03.579514 137274321021824 utils.py:1231] [77150] progress = 0.6851504844453522
+I1203 12:15:03.579569 137274321021824 utils.py:1231] [77150] epoch = 61.66377997560037
+I1203 12:15:03.579627 137274321021824 utils.py:1231] [77150] img/sec/core = 164.869987873062
+I1203 12:15:03.579693 137274321021824 utils.py:1231] [77150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.3803884418689
+I1203 12:15:03.579751 137274321021824 utils.py:1231] [77150] core_hours = 134.3803884418689
+I1203 12:15:03.579818 137274321021824 train.py:125] NOTE: Steps:77150/112603 [68.5%]
+Walltime:5d14h24m (0s eval)
+ETA:2d13h45m
+Total train time:8d4h8m
+I1203 12:20:15.195697 137274321021824 utils.py:1231] [77200] l2_params = 264.9453962180418
+I1203 12:20:15.195984 137274321021824 utils.py:1231] [77200] train/loss = 4.367486298084259
+I1203 12:20:15.196153 137274321021824 utils.py:1231] [77200] l2_grads = 1.8799431324005127
+I1203 12:20:15.196237 137274321021824 utils.py:1231] [77200] lr = 0.0002661158030596906
+I1203 12:20:15.196305 137274321021824 utils.py:1231] [77200] uptime = 484204.558660349
+I1203 12:20:15.196382 137274321021824 utils.py:1231] [77200] examples_seen = 79052800.0
+I1203 12:20:15.196445 137274321021824 utils.py:1231] [77200] progress = 0.6855945223484277
+I1203 12:20:15.196509 137274321021824 utils.py:1231] [77200] epoch = 61.703743540069325
+I1203 12:20:15.196574 137274321021824 utils.py:1231] [77200] img/sec/core = 164.30431023858833
+I1203 12:20:15.196637 137274321021824 utils.py:1231] [77200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.46694869218473
+I1203 12:20:15.196704 137274321021824 utils.py:1231] [77200] core_hours = 134.46694869218473
+I1203 12:20:15.196806 137274321021824 train.py:125] NOTE: Steps:77200/112603 [68.6%]
+Walltime:5d14h30m (0s eval)
+ETA:2d13h39m
+Total train time:8d4h8m
+I1203 12:25:27.000872 137274321021824 utils.py:1231] [77250] l2_params = 264.85597598548264
+I1203 12:25:27.001093 137274321021824 utils.py:1231] [77250] train/loss = 4.204144895076752
+I1203 12:25:27.001223 137274321021824 utils.py:1231] [77250] l2_grads = 1.9328840970993042
+I1203 12:25:27.001295 137274321021824 utils.py:1231] [77250] lr = 0.0002654395133604163
+I1203 12:25:27.001362 137274321021824 utils.py:1231] [77250] uptime = 484516.363719864
+I1203 12:25:27.001415 137274321021824 utils.py:1231] [77250] examples_seen = 79104000.0
+I1203 12:25:27.001467 137274321021824 utils.py:1231] [77250] progress = 0.6860385602515031
+I1203 12:25:27.001515 137274321021824 utils.py:1231] [77250] epoch = 61.74370710453828
+I1203 12:25:27.001570 137274321021824 utils.py:1231] [77250] img/sec/core = 164.20516100553286
+I1203 12:25:27.001642 137274321021824 utils.py:1231] [77250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.55356120871664
+I1203 12:25:27.001703 137274321021824 utils.py:1231] [77250] core_hours = 134.55356120871664
+I1203 12:25:27.001764 137274321021824 train.py:125] NOTE: Steps:77250/112603 [68.6%]
+Walltime:5d14h35m (0s eval)
+ETA:2d13h34m
+Total train time:8d4h8m
+I1203 12:30:38.807094 137274321021824 utils.py:1231] [77300] l2_params = 264.7710363688841
+I1203 12:30:38.807305 137274321021824 utils.py:1231] [77300] train/loss = 4.2683965265750885
+I1203 12:30:38.807404 137274321021824 utils.py:1231] [77300] l2_grads = 1.8659310340881348
+I1203 12:30:38.807477 137274321021824 utils.py:1231] [77300] lr = 0.00026476377342274566
+I1203 12:30:38.807538 137274321021824 utils.py:1231] [77300] uptime = 484828.169898865
+I1203 12:30:38.807599 137274321021824 utils.py:1231] [77300] examples_seen = 79155200.0
+I1203 12:30:38.807656 137274321021824 utils.py:1231] [77300] progress = 0.6864825981545785
+I1203 12:30:38.807712 137274321021824 utils.py:1231] [77300] epoch = 61.78367066900724
+I1203 12:30:38.807765 137274321021824 utils.py:1231] [77300] img/sec/core = 164.2045714553862
+I1203 12:30:38.807825 137274321021824 utils.py:1231] [77300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.64017403621693
+I1203 12:30:38.807889 137274321021824 utils.py:1231] [77300] core_hours = 134.64017403621693
+I1203 12:30:38.807960 137274321021824 train.py:125] NOTE: Steps:77300/112603 [68.6%]
+Walltime:5d14h40m (0s eval)
+ETA:2d13h29m
+Total train time:8d4h8m
+I1203 12:35:50.692820 137274321021824 utils.py:1231] [77350] l2_params = 264.69044323267764
+I1203 12:35:50.693059 137274321021824 utils.py:1231] [77350] train/loss = 1.8829382359981537
+I1203 12:35:50.693201 137274321021824 utils.py:1231] [77350] l2_grads = 1.9703688621520996
+I1203 12:35:50.693303 137274321021824 utils.py:1231] [77350] lr = 0.0002640885848304743
+I1203 12:35:50.693380 137274321021824 utils.py:1231] [77350] uptime = 485140.05573792
+I1203 12:35:50.693452 137274321021824 utils.py:1231] [77350] examples_seen = 79206400.0
+I1203 12:35:50.693520 137274321021824 utils.py:1231] [77350] progress = 0.6869266360576539
+I1203 12:35:50.693591 137274321021824 utils.py:1231] [77350] epoch = 61.8236342334762
+I1203 12:35:50.693660 137274321021824 utils.py:1231] [77350] img/sec/core = 164.16263128564438
+I1203 12:35:50.693729 137274321021824 utils.py:1231] [77350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.72680899150998
+I1203 12:35:50.693789 137274321021824 utils.py:1231] [77350] core_hours = 134.72680899150998
+I1203 12:35:50.693863 137274321021824 train.py:125] NOTE: Steps:77350/112603 [68.7%]
+Walltime:5d14h45m (0s eval)
+ETA:2d13h24m
+Total train time:8d4h8m
+I1203 12:41:02.480954 137274321021824 utils.py:1231] [77400] l2_params = 264.60872870581164
+I1203 12:41:02.481181 137274321021824 utils.py:1231] [77400] train/loss = 2.245223879814148
+I1203 12:41:02.481283 137274321021824 utils.py:1231] [77400] l2_grads = 1.9879839420318604
+I1203 12:41:02.481343 137274321021824 utils.py:1231] [77400] lr = 0.0002634139491661055
+I1203 12:41:02.481395 137274321021824 utils.py:1231] [77400] uptime = 485451.843757199
+I1203 12:41:02.481448 137274321021824 utils.py:1231] [77400] examples_seen = 79257600.0
+I1203 12:41:02.481501 137274321021824 utils.py:1231] [77400] progress = 0.6873706739607293
+I1203 12:41:02.481549 137274321021824 utils.py:1231] [77400] epoch = 61.863597797945154
+I1203 12:41:02.481600 137274321021824 utils.py:1231] [77400] img/sec/core = 164.2141353551539
+I1203 12:41:02.481657 137274321021824 utils.py:1231] [77400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.81341677464306
+I1203 12:41:02.481708 137274321021824 utils.py:1231] [77400] core_hours = 134.81341677464306
+I1203 12:41:02.481794 137274321021824 train.py:125] NOTE: Steps:77400/112603 [68.7%]
+Walltime:5d14h50m (0s eval)
+ETA:2d13h19m
+Total train time:8d4h8m
+I1203 12:46:14.283073 137274321021824 utils.py:1231] [77450] l2_params = 264.5324807580616
+I1203 12:46:14.283343 137274321021824 utils.py:1231] [77450] train/loss = 2.459239959716797
+I1203 12:46:14.283468 137274321021824 utils.py:1231] [77450] l2_grads = 1.9994643926620483
+I1203 12:46:14.283566 137274321021824 utils.py:1231] [77450] lr = 0.0002627398680108469
+I1203 12:46:14.283646 137274321021824 utils.py:1231] [77450] uptime = 485763.646007752
+I1203 12:46:14.283721 137274321021824 utils.py:1231] [77450] examples_seen = 79308800.0
+I1203 12:46:14.283781 137274321021824 utils.py:1231] [77450] progress = 0.6878147118638047
+I1203 12:46:14.283843 137274321021824 utils.py:1231] [77450] epoch = 61.90356136241411
+I1203 12:46:14.283911 137274321021824 utils.py:1231] [77450] img/sec/core = 164.20664029588272
+I1203 12:46:14.283974 137274321021824 utils.py:1231] [77450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.9000285109078
+I1203 12:46:14.284029 137274321021824 utils.py:1231] [77450] core_hours = 134.9000285109078
+I1203 12:46:14.284093 137274321021824 train.py:125] NOTE: Steps:77450/112603 [68.8%]
+Walltime:5d14h56m (0s eval)
+ETA:2d13h13m
+Total train time:8d4h8m
+I1203 12:51:25.891958 137274321021824 utils.py:1231] [77500] l2_params = 264.44147155757304
+I1203 12:51:25.892175 137274321021824 utils.py:1231] [77500] train/loss = 1.8707342892885208
+I1203 12:51:25.892280 137274321021824 utils.py:1231] [77500] l2_grads = 2.0856759548187256
+I1203 12:51:25.892366 137274321021824 utils.py:1231] [77500] lr = 0.00026206634294460656
+I1203 12:51:25.892430 137274321021824 utils.py:1231] [77500] uptime = 486075.254790893
+I1203 12:51:25.892497 137274321021824 utils.py:1231] [77500] examples_seen = 79360000.0
+I1203 12:51:25.892552 137274321021824 utils.py:1231] [77500] progress = 0.6882587497668801
+I1203 12:51:25.892606 137274321021824 utils.py:1231] [77500] epoch = 61.94352492688307
+I1203 12:51:25.892674 137274321021824 utils.py:1231] [77500] img/sec/core = 164.30859067547726
+I1203 12:51:25.892738 137274321021824 utils.py:1231] [77500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 134.9865865062247
+I1203 12:51:25.892817 137274321021824 utils.py:1231] [77500] core_hours = 134.9865865062247
+I1203 12:51:25.892930 137274321021824 train.py:125] NOTE: Steps:77500/112603 [68.8%]
+Walltime:5d15h1m (0s eval)
+ETA:2d13h8m
+Total train time:8d4h7m
+I1203 12:51:25.893080 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:77500/112603 [68.8%]
+Walltime:5d15h1m (0s eval)
+ETA:2d13h8m
+Total train time:8d4h7m
+I1203 12:53:03.571765 137274321021824 utils.py:1231] [77500] val/acc@1 = 0.7165776466836735
+I1203 12:53:03.571989 137274321021824 utils.py:1231] [77500] val/loss = 1.1347645385837068
+I1203 12:53:03.572135 137274321021824 utils.py:1231] [77500] z/secs/eval/val = 97.67894641601015
+I1203 12:53:03.572225 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.67894641601015
+I1203 12:58:14.262311 137274321021824 utils.py:1231] [77550] l2_params = 264.3635078696493
+I1203 12:58:14.262509 137274321021824 utils.py:1231] [77550] train/loss = 1.7794191390275955
+I1203 12:58:14.262598 137274321021824 utils.py:1231] [77550] l2_grads = 1.927625060081482
+I1203 12:58:14.262663 137274321021824 utils.py:1231] [77550] lr = 0.0002613933755459883
+I1203 12:58:14.262722 137274321021824 utils.py:1231] [77550] uptime = 486483.625084055
+I1203 12:58:14.262779 137274321021824 utils.py:1231] [77550] examples_seen = 79411200.0
+I1203 12:58:14.262833 137274321021824 utils.py:1231] [77550] progress = 0.6887027876699555
+I1203 12:58:14.262891 137274321021824 utils.py:1231] [77550] epoch = 61.98348849135203
+I1203 12:58:14.262947 137274321021824 utils.py:1231] [77550] img/sec/core = 125.37640679873896
+I1203 12:58:14.263002 137274321021824 utils.py:1231] [77550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.10002269876972
+I1203 12:58:14.263053 137274321021824 utils.py:1231] [77550] core_hours = 135.10002269876972
+I1203 12:58:14.263116 137274321021824 train.py:125] NOTE: Steps:77550/112603 [68.9%]
+Walltime:5d15h8m (0s eval)
+ETA:2d13h4m
+Total train time:8d4h10m
+I1203 13:03:26.035349 137274321021824 utils.py:1231] [77600] l2_params = 264.28708325852375
+I1203 13:03:26.035588 137274321021824 utils.py:1231] [77600] train/loss = 2.189223974943161
+I1203 13:03:26.035705 137274321021824 utils.py:1231] [77600] l2_grads = 1.9008715152740479
+I1203 13:03:26.035790 137274321021824 utils.py:1231] [77600] lr = 0.0002607209673922898
+I1203 13:03:26.035869 137274321021824 utils.py:1231] [77600] uptime = 486795.398227083
+I1203 13:03:26.035997 137274321021824 utils.py:1231] [77600] examples_seen = 79462400.0
+I1203 13:03:26.036086 137274321021824 utils.py:1231] [77600] progress = 0.6891468255730309
+I1203 13:03:26.036147 137274321021824 utils.py:1231] [77600] epoch = 62.02345205582098
+I1203 13:03:26.036207 137274321021824 utils.py:1231] [77600] img/sec/core = 164.2219708302686
+I1203 13:03:26.036273 137274321021824 utils.py:1231] [77600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.1866263496108
+I1203 13:03:26.036327 137274321021824 utils.py:1231] [77600] core_hours = 135.1866263496108
+I1203 13:03:26.036407 137274321021824 train.py:125] NOTE: Steps:77600/112603 [68.9%]
+Walltime:5d15h13m (0s eval)
+ETA:2d12h58m
+Total train time:8d4h10m
+I1203 13:08:37.788075 137274321021824 utils.py:1231] [77650] l2_params = 264.20320644362744
+I1203 13:08:37.788324 137274321021824 utils.py:1231] [77650] train/loss = 1.9743543416261673
+I1203 13:08:37.788453 137274321021824 utils.py:1231] [77650] l2_grads = 2.124868392944336
+I1203 13:08:37.788536 137274321021824 utils.py:1231] [77650] lr = 0.00026004912005949754
+I1203 13:08:37.788590 137274321021824 utils.py:1231] [77650] uptime = 487107.150951609
+I1203 13:08:37.788654 137274321021824 utils.py:1231] [77650] examples_seen = 79513600.0
+I1203 13:08:37.788706 137274321021824 utils.py:1231] [77650] progress = 0.6895908634761063
+I1203 13:08:37.788760 137274321021824 utils.py:1231] [77650] epoch = 62.06341562028994
+I1203 13:08:37.788820 137274321021824 utils.py:1231] [77650] img/sec/core = 164.23272668375014
+I1203 13:08:37.788896 137274321021824 utils.py:1231] [77650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.27322432864582
+I1203 13:08:37.788949 137274321021824 utils.py:1231] [77650] core_hours = 135.27322432864582
+I1203 13:08:37.789009 137274321021824 train.py:125] NOTE: Steps:77650/112603 [69.0%]
+Walltime:5d15h18m (0s eval)
+ETA:2d12h53m
+Total train time:8d4h10m
+I1203 13:13:49.568691 137274321021824 utils.py:1231] [77700] l2_params = 264.1323568657297
+I1203 13:13:49.568945 137274321021824 utils.py:1231] [77700] train/loss = 1.9151190221309662
+I1203 13:13:49.569067 137274321021824 utils.py:1231] [77700] l2_grads = 2.016904354095459
+I1203 13:13:49.569141 137274321021824 utils.py:1231] [77700] lr = 0.0002593778351222838
+I1203 13:13:49.569195 137274321021824 utils.py:1231] [77700] uptime = 487418.931556618
+I1203 13:13:49.569249 137274321021824 utils.py:1231] [77700] examples_seen = 79564800.0
+I1203 13:13:49.569299 137274321021824 utils.py:1231] [77700] progress = 0.6900349013791818
+I1203 13:13:49.569349 137274321021824 utils.py:1231] [77700] epoch = 62.10337918475889
+I1203 13:13:49.569406 137274321021824 utils.py:1231] [77700] img/sec/core = 164.21804043430473
+I1203 13:13:49.569471 137274321021824 utils.py:1231] [77700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.35983005225944
+I1203 13:13:49.569523 137274321021824 utils.py:1231] [77700] core_hours = 135.35983005225944
+I1203 13:13:49.569586 137274321021824 train.py:125] NOTE: Steps:77700/112603 [69.0%]
+Walltime:5d15h23m (0s eval)
+ETA:2d12h48m
+Total train time:8d4h10m
+I1203 13:19:01.352962 137274321021824 utils.py:1231] [77750] l2_params = 264.0509132835146
+I1203 13:19:01.353227 137274321021824 utils.py:1231] [77750] train/loss = 2.498083859682083
+I1203 13:19:01.353349 137274321021824 utils.py:1231] [77750] l2_grads = 1.768010139465332
+I1203 13:19:01.353441 137274321021824 utils.py:1231] [77750] lr = 0.0002587071141540027
+I1203 13:19:01.353502 137274321021824 utils.py:1231] [77750] uptime = 487730.71585909894
+I1203 13:19:01.353554 137274321021824 utils.py:1231] [77750] examples_seen = 79616000.0
+I1203 13:19:01.353607 137274321021824 utils.py:1231] [77750] progress = 0.6904789392822571
+I1203 13:19:01.353660 137274321021824 utils.py:1231] [77750] epoch = 62.143342749227855
+I1203 13:19:01.353726 137274321021824 utils.py:1231] [77750] img/sec/core = 164.21609296103742
+I1203 13:19:01.353787 137274321021824 utils.py:1231] [77750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.4464368029486
+I1203 13:19:01.353843 137274321021824 utils.py:1231] [77750] core_hours = 135.4464368029486
+I1203 13:19:01.353925 137274321021824 train.py:125] NOTE: Steps:77750/112603 [69.0%]
+Walltime:5d15h28m (0s eval)
+ETA:2d12h43m
+Total train time:8d4h10m
+I1203 13:24:13.156426 137274321021824 utils.py:1231] [77800] l2_params = 263.97031730145017
+I1203 13:24:13.156634 137274321021824 utils.py:1231] [77800] train/loss = 2.5626446306705475
+I1203 13:24:13.156734 137274321021824 utils.py:1231] [77800] l2_grads = 1.964997410774231
+I1203 13:24:13.156800 137274321021824 utils.py:1231] [77800] lr = 0.0002580369587266861
+I1203 13:24:13.156857 137274321021824 utils.py:1231] [77800] uptime = 488042.519219007
+I1203 13:24:13.156919 137274321021824 utils.py:1231] [77800] examples_seen = 79667200.0
+I1203 13:24:13.156974 137274321021824 utils.py:1231] [77800] progress = 0.6909229771853326
+I1203 13:24:13.157026 137274321021824 utils.py:1231] [77800] epoch = 62.18330631369681
+I1203 13:24:13.157081 137274321021824 utils.py:1231] [77800] img/sec/core = 164.20605607039786
+I1203 13:24:13.157147 137274321021824 utils.py:1231] [77800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.5330488473675
+I1203 13:24:13.157197 137274321021824 utils.py:1231] [77800] core_hours = 135.5330488473675
+I1203 13:24:13.157256 137274321021824 train.py:125] NOTE: Steps:77800/112603 [69.1%]
+Walltime:5d15h34m (0s eval)
+ETA:2d12h37m
+Total train time:8d4h10m
+I1203 13:29:24.948626 137274321021824 utils.py:1231] [77850] l2_params = 263.8856022870804
+I1203 13:29:24.948849 137274321021824 utils.py:1231] [77850] train/loss = 1.8342762291431427
+I1203 13:29:24.948961 137274321021824 utils.py:1231] [77850] l2_grads = 2.075180768966675
+I1203 13:29:24.949032 137274321021824 utils.py:1231] [77850] lr = 0.0002573673704110412
+I1203 13:29:24.949114 137274321021824 utils.py:1231] [77850] uptime = 488354.311476071
+I1203 13:29:24.949185 137274321021824 utils.py:1231] [77850] examples_seen = 79718400.0
+I1203 13:29:24.949246 137274321021824 utils.py:1231] [77850] progress = 0.6913670150884079
+I1203 13:29:24.949301 137274321021824 utils.py:1231] [77850] epoch = 62.223269878165766
+I1203 13:29:24.949361 137274321021824 utils.py:1231] [77850] img/sec/core = 164.21190340684697
+I1203 13:29:24.949432 137274321021824 utils.py:1231] [77850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.61965780766306
+I1203 13:29:24.949485 137274321021824 utils.py:1231] [77850] core_hours = 135.61965780766306
+I1203 13:29:24.949563 137274321021824 train.py:125] NOTE: Steps:77850/112603 [69.1%]
+Walltime:5d15h39m (0s eval)
+ETA:2d12h32m
+Total train time:8d4h9m
+I1203 13:34:36.748453 137274321021824 utils.py:1231] [77900] l2_params = 263.80225555551124
+I1203 13:34:36.748654 137274321021824 utils.py:1231] [77900] train/loss = 1.8067406713962555
+I1203 13:34:36.748748 137274321021824 utils.py:1231] [77900] l2_grads = 2.0165648460388184
+I1203 13:34:36.748807 137274321021824 utils.py:1231] [77900] lr = 0.0002566983507764448
+I1203 13:34:36.748857 137274321021824 utils.py:1231] [77900] uptime = 488666.11121977295
+I1203 13:34:36.748920 137274321021824 utils.py:1231] [77900] examples_seen = 79769600.0
+I1203 13:34:36.748969 137274321021824 utils.py:1231] [77900] progress = 0.6918110529914834
+I1203 13:34:36.749017 137274321021824 utils.py:1231] [77900] epoch = 62.26323344263472
+I1203 13:34:36.749068 137274321021824 utils.py:1231] [77900] img/sec/core = 164.20796050731963
+I1203 13:34:36.749123 137274321021824 utils.py:1231] [77900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.70626884758025
+I1203 13:34:36.749172 137274321021824 utils.py:1231] [77900] core_hours = 135.70626884758025
+I1203 13:34:36.749231 137274321021824 train.py:125] NOTE: Steps:77900/112603 [69.2%]
+Walltime:5d15h44m (0s eval)
+ETA:2d12h27m
+Total train time:8d4h9m
+I1203 13:39:48.534985 137274321021824 utils.py:1231] [77950] l2_params = 263.71182891521005
+I1203 13:39:48.535274 137274321021824 utils.py:1231] [77950] train/loss = 1.9389088451862335
+I1203 13:39:48.535386 137274321021824 utils.py:1231] [77950] l2_grads = 2.1554667949676514
+I1203 13:39:48.535464 137274321021824 utils.py:1231] [77950] lr = 0.000256029901390942
+I1203 13:39:48.535521 137274321021824 utils.py:1231] [77950] uptime = 488977.897883569
+I1203 13:39:48.535585 137274321021824 utils.py:1231] [77950] examples_seen = 79820800.0
+I1203 13:39:48.535635 137274321021824 utils.py:1231] [77950] progress = 0.6922550908945587
+I1203 13:39:48.535684 137274321021824 utils.py:1231] [77950] epoch = 62.30319700710368
+I1203 13:39:48.535735 137274321021824 utils.py:1231] [77950] img/sec/core = 164.2148492710738
+I1203 13:39:48.535794 137274321021824 utils.py:1231] [77950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.79287625419028
+I1203 13:39:48.535845 137274321021824 utils.py:1231] [77950] core_hours = 135.79287625419028
+I1203 13:39:48.535917 137274321021824 train.py:125] NOTE: Steps:77950/112603 [69.2%]
+Walltime:5d15h49m (0s eval)
+ETA:2d12h22m
+Total train time:8d4h9m
+I1203 13:45:00.317501 137274321021824 utils.py:1231] [78000] l2_params = 263.6310924698313
+I1203 13:45:00.317785 137274321021824 utils.py:1231] [78000] train/loss = 1.80961674451828
+I1203 13:45:00.318023 137274321021824 utils.py:1231] [78000] l2_grads = 1.9900894165039062
+I1203 13:45:00.318169 137274321021824 utils.py:1231] [78000] lr = 0.00025536202382124047
+I1203 13:45:00.318267 137274321021824 utils.py:1231] [78000] uptime = 489289.68061986694
+I1203 13:45:00.318348 137274321021824 utils.py:1231] [78000] examples_seen = 79872000.0
+I1203 13:45:00.318437 137274321021824 utils.py:1231] [78000] progress = 0.6926991287976342
+I1203 13:45:00.318517 137274321021824 utils.py:1231] [78000] epoch = 62.34316057157264
+I1203 13:45:00.318591 137274321021824 utils.py:1231] [78000] img/sec/core = 164.21691787025344
+I1203 13:45:00.318667 137274321021824 utils.py:1231] [78000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.87948256982858
+I1203 13:45:00.318728 137274321021824 utils.py:1231] [78000] core_hours = 135.87948256982858
+I1203 13:45:00.318803 137274321021824 train.py:125] NOTE: Steps:78000/112603 [69.3%]
+Walltime:5d15h54m (0s eval)
+ETA:2d12h16m
+Total train time:8d4h9m
+I1203 13:50:12.349152 137274321021824 utils.py:1231] [78050] l2_params = 263.5539750085735
+I1203 13:50:12.349367 137274321021824 utils.py:1231] [78050] train/loss = 2.0967212468385696
+I1203 13:50:12.349473 137274321021824 utils.py:1231] [78050] l2_grads = 2.1323471069335938
+I1203 13:50:12.349545 137274321021824 utils.py:1231] [78050] lr = 0.00025469471963270843
+I1203 13:50:12.349606 137274321021824 utils.py:1231] [78050] uptime = 489601.711967936
+I1203 13:50:12.349668 137274321021824 utils.py:1231] [78050] examples_seen = 79923200.0
+I1203 13:50:12.349725 137274321021824 utils.py:1231] [78050] progress = 0.6931431667007095
+I1203 13:50:12.349781 137274321021824 utils.py:1231] [78050] epoch = 62.383124136041594
+I1203 13:50:12.349844 137274321021824 utils.py:1231] [78050] img/sec/core = 164.08607762276694
+I1203 13:50:12.349926 137274321021824 utils.py:1231] [78050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 135.9661579442922
+I1203 13:50:12.350003 137274321021824 utils.py:1231] [78050] core_hours = 135.9661579442922
+I1203 13:50:12.350073 137274321021824 train.py:125] NOTE: Steps:78050/112603 [69.3%]
+Walltime:5d16h0m (0s eval)
+ETA:2d12h11m
+Total train time:8d4h9m
+I1203 13:55:24.130021 137274321021824 utils.py:1231] [78100] l2_params = 263.4736165550913
+I1203 13:55:24.130274 137274321021824 utils.py:1231] [78100] train/loss = 3.1264860033988953
+I1203 13:55:24.130419 137274321021824 utils.py:1231] [78100] l2_grads = 1.8774834871292114
+I1203 13:55:24.130511 137274321021824 utils.py:1231] [78100] lr = 0.00025402799038936984
+I1203 13:55:24.130594 137274321021824 utils.py:1231] [78100] uptime = 489913.49295466393
+I1203 13:55:24.130665 137274321021824 utils.py:1231] [78100] examples_seen = 79974400.0
+I1203 13:55:24.130731 137274321021824 utils.py:1231] [78100] progress = 0.693587204603785
+I1203 13:55:24.130802 137274321021824 utils.py:1231] [78100] epoch = 62.42308770051055
+I1203 13:55:24.130872 137274321021824 utils.py:1231] [78100] img/sec/core = 164.21783937928646
+I1203 13:55:24.130983 137274321021824 utils.py:1231] [78100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.05276377393886
+I1203 13:55:24.131079 137274321021824 utils.py:1231] [78100] core_hours = 136.05276377393886
+I1203 13:55:24.131177 137274321021824 train.py:125] NOTE: Steps:78100/112603 [69.4%]
+Walltime:5d16h5m (0s eval)
+ETA:2d12h6m
+Total train time:8d4h9m
+I1203 14:00:35.907216 137274321021824 utils.py:1231] [78150] l2_params = 263.39110926911553
+I1203 14:00:35.907489 137274321021824 utils.py:1231] [78150] train/loss = 1.794064313173294
+I1203 14:00:35.907629 137274321021824 utils.py:1231] [78150] l2_grads = 2.0495173931121826
+I1203 14:00:35.907719 137274321021824 utils.py:1231] [78150] lr = 0.00025336183765390063
+I1203 14:00:35.907785 137274321021824 utils.py:1231] [78150] uptime = 490225.27014308295
+I1203 14:00:35.907855 137274321021824 utils.py:1231] [78150] examples_seen = 80025600.0
+I1203 14:00:35.907917 137274321021824 utils.py:1231] [78150] progress = 0.6940312425068604
+I1203 14:00:35.907972 137274321021824 utils.py:1231] [78150] epoch = 62.463051264979505
+I1203 14:00:35.908027 137274321021824 utils.py:1231] [78150] img/sec/core = 164.2198400069905
+I1203 14:00:35.908094 137274321021824 utils.py:1231] [78150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.1393685484997
+I1203 14:00:35.908152 137274321021824 utils.py:1231] [78150] core_hours = 136.1393685484997
+I1203 14:00:35.908222 137274321021824 train.py:125] NOTE: Steps:78150/112603 [69.4%]
+Walltime:5d16h10m (0s eval)
+ETA:2d12h1m
+Total train time:8d4h9m
+I1203 14:05:47.706682 137274321021824 utils.py:1231] [78200] l2_params = 263.30931926982004
+I1203 14:05:47.706918 137274321021824 utils.py:1231] [78200] train/loss = 1.8679904639720917
+I1203 14:05:47.707018 137274321021824 utils.py:1231] [78200] l2_grads = 2.170994281768799
+I1203 14:05:47.707078 137274321021824 utils.py:1231] [78200] lr = 0.00025269626298762685
+I1203 14:05:47.707129 137274321021824 utils.py:1231] [78200] uptime = 490537.069491421
+I1203 14:05:47.707179 137274321021824 utils.py:1231] [78200] examples_seen = 80076800.0
+I1203 14:05:47.707227 137274321021824 utils.py:1231] [78200] progress = 0.6944752804099358
+I1203 14:05:47.707275 137274321021824 utils.py:1231] [78200] epoch = 62.50301482944846
+I1203 14:05:47.707324 137274321021824 utils.py:1231] [78200] img/sec/core = 164.20816872423617
+I1203 14:05:47.707378 137274321021824 utils.py:1231] [78200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.22597947859362
+I1203 14:05:47.707427 137274321021824 utils.py:1231] [78200] core_hours = 136.22597947859362
+I1203 14:05:47.707484 137274321021824 train.py:125] NOTE: Steps:78200/112603 [69.4%]
+Walltime:5d16h15m (0s eval)
+ETA:2d11h55m
+Total train time:8d4h9m
+I1203 14:10:59.509172 137274321021824 utils.py:1231] [78250] l2_params = 263.2312169352306
+I1203 14:10:59.509442 137274321021824 utils.py:1231] [78250] train/loss = 2.1533926129341125
+I1203 14:10:59.509559 137274321021824 utils.py:1231] [78250] l2_grads = 1.9732714891433716
+I1203 14:10:59.509642 137274321021824 utils.py:1231] [78250] lr = 0.00025203126795051806
+I1203 14:10:59.509715 137274321021824 utils.py:1231] [78250] uptime = 490848.87207547494
+I1203 14:10:59.509783 137274321021824 utils.py:1231] [78250] examples_seen = 80128000.0
+I1203 14:10:59.509832 137274321021824 utils.py:1231] [78250] progress = 0.6949193183130112
+I1203 14:10:59.509886 137274321021824 utils.py:1231] [78250] epoch = 62.54297839391742
+I1203 14:10:59.509943 137274321021824 utils.py:1231] [78250] img/sec/core = 164.20646466208024
+I1203 14:10:59.510001 137274321021824 utils.py:1231] [78250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.31259130749748
+I1203 14:10:59.510054 137274321021824 utils.py:1231] [78250] core_hours = 136.31259130749748
+I1203 14:10:59.510114 137274321021824 train.py:125] NOTE: Steps:78250/112603 [69.5%]
+Walltime:5d16h20m (0s eval)
+ETA:2d11h50m
+Total train time:8d4h9m
+I1203 14:16:11.304320 137274321021824 utils.py:1231] [78300] l2_params = 263.1466434869372
+I1203 14:16:11.304553 137274321021824 utils.py:1231] [78300] train/loss = 2.144268274307251
+I1203 14:16:11.304699 137274321021824 utils.py:1231] [78300] l2_grads = 1.9351813793182373
+I1203 14:16:11.304797 137274321021824 utils.py:1231] [78300] lr = 0.0002513668541011862
+I1203 14:16:11.304888 137274321021824 utils.py:1231] [78300] uptime = 491160.66724003304
+I1203 14:16:11.304973 137274321021824 utils.py:1231] [78300] examples_seen = 80179200.0
+I1203 14:16:11.305050 137274321021824 utils.py:1231] [78300] progress = 0.6953633562160866
+I1203 14:16:11.305122 137274321021824 utils.py:1231] [78300] epoch = 62.58294195838638
+I1203 14:16:11.305198 137274321021824 utils.py:1231] [78300] img/sec/core = 164.21037212864962
+I1203 14:16:11.305264 137274321021824 utils.py:1231] [78300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.39920107543028
+I1203 14:16:11.305338 137274321021824 utils.py:1231] [78300] core_hours = 136.39920107543028
+I1203 14:16:11.305421 137274321021824 train.py:125] NOTE: Steps:78300/112603 [69.5%]
+Walltime:5d16h26m (0s eval)
+ETA:2d11h45m
+Total train time:8d4h9m
+I1203 14:21:23.095600 137274321021824 utils.py:1231] [78350] l2_params = 263.06746164209744
+I1203 14:21:23.095954 137274321021824 utils.py:1231] [78350] train/loss = 1.9202460497617722
+I1203 14:21:23.096153 137274321021824 utils.py:1231] [78350] l2_grads = 2.0102977752685547
+I1203 14:21:23.096235 137274321021824 utils.py:1231] [78350] lr = 0.00025070302299688094
+I1203 14:21:23.096312 137274321021824 utils.py:1231] [78350] uptime = 491472.45867204096
+I1203 14:21:23.096384 137274321021824 utils.py:1231] [78350] examples_seen = 80230400.0
+I1203 14:21:23.096471 137274321021824 utils.py:1231] [78350] progress = 0.695807394119162
+I1203 14:21:23.096546 137274321021824 utils.py:1231] [78350] epoch = 62.62290552285533
+I1203 14:21:23.096617 137274321021824 utils.py:1231] [78350] img/sec/core = 164.21233794102577
+I1203 14:21:23.096724 137274321021824 utils.py:1231] [78350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.4858098065436
+I1203 14:21:23.096858 137274321021824 utils.py:1231] [78350] core_hours = 136.4858098065436
+I1203 14:21:23.096947 137274321021824 train.py:125] NOTE: Steps:78350/112603 [69.6%]
+Walltime:5d16h31m (0s eval)
+ETA:2d11h40m
+Total train time:8d4h9m
+I1203 14:26:34.879164 137274321021824 utils.py:1231] [78400] l2_params = 262.99147807935185
+I1203 14:26:34.879379 137274321021824 utils.py:1231] [78400] train/loss = 2.3276174068450928
+I1203 14:26:34.879492 137274321021824 utils.py:1231] [78400] l2_grads = 1.900734543800354
+I1203 14:26:34.879562 137274321021824 utils.py:1231] [78400] lr = 0.0002500397761934863
+I1203 14:26:34.879625 137274321021824 utils.py:1231] [78400] uptime = 491784.241986318
+I1203 14:26:34.879687 137274321021824 utils.py:1231] [78400] examples_seen = 80281600.0
+I1203 14:26:34.879745 137274321021824 utils.py:1231] [78400] progress = 0.6962514320222374
+I1203 14:26:34.879802 137274321021824 utils.py:1231] [78400] epoch = 62.66286908732429
+I1203 14:26:34.879863 137274321021824 utils.py:1231] [78400] img/sec/core = 164.2166134474429
+I1203 14:26:34.879934 137274321021824 utils.py:1231] [78400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.5724162827317
+I1203 14:26:34.879989 137274321021824 utils.py:1231] [78400] core_hours = 136.5724162827317
+I1203 14:26:34.880050 137274321021824 train.py:125] NOTE: Steps:78400/112603 [69.6%]
+Walltime:5d16h36m (0s eval)
+ETA:2d11h34m
+Total train time:8d4h9m
+I1203 14:31:46.668566 137274321021824 utils.py:1231] [78450] l2_params = 262.909124020196
+I1203 14:31:46.668780 137274321021824 utils.py:1231] [78450] train/loss = 4.2264761328697205
+I1203 14:31:46.668894 137274321021824 utils.py:1231] [78450] l2_grads = 2.037996530532837
+I1203 14:31:46.668968 137274321021824 utils.py:1231] [78450] lr = 0.000249377115245516
+I1203 14:31:46.669030 137274321021824 utils.py:1231] [78450] uptime = 492096.031391272
+I1203 14:31:46.669090 137274321021824 utils.py:1231] [78450] examples_seen = 80332800.0
+I1203 14:31:46.669152 137274321021824 utils.py:1231] [78450] progress = 0.6966954699253128
+I1203 14:31:46.669207 137274321021824 utils.py:1231] [78450] epoch = 62.70283265179325
+I1203 14:31:46.669265 137274321021824 utils.py:1231] [78450] img/sec/core = 164.21340554392035
+I1203 14:31:46.669323 137274321021824 utils.py:1231] [78450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.6590244507744
+I1203 14:31:46.669377 137274321021824 utils.py:1231] [78450] core_hours = 136.6590244507744
+I1203 14:31:46.669449 137274321021824 train.py:125] NOTE: Steps:78450/112603 [69.7%]
+Walltime:5d16h41m (0s eval)
+ETA:2d11h29m
+Total train time:8d4h9m
+I1203 14:36:58.635124 137274321021824 utils.py:1231] [78500] l2_params = 262.83138897155357
+I1203 14:36:58.635390 137274321021824 utils.py:1231] [78500] train/loss = 3.6101349890232086
+I1203 14:36:58.635557 137274321021824 utils.py:1231] [78500] l2_grads = 1.9153739213943481
+I1203 14:36:58.635645 137274321021824 utils.py:1231] [78500] lr = 0.00024871504170611154
+I1203 14:36:58.635705 137274321021824 utils.py:1231] [78500] uptime = 492407.998067098
+I1203 14:36:58.635764 137274321021824 utils.py:1231] [78500] examples_seen = 80384000.0
+I1203 14:36:58.635817 137274321021824 utils.py:1231] [78500] progress = 0.6971395078283882
+I1203 14:36:58.635871 137274321021824 utils.py:1231] [78500] epoch = 62.74279621626221
+I1203 14:36:58.635937 137274321021824 utils.py:1231] [78500] img/sec/core = 164.12009348253926
+I1203 14:36:58.635994 137274321021824 utils.py:1231] [78500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.7456818607261
+I1203 14:36:58.636047 137274321021824 utils.py:1231] [78500] core_hours = 136.7456818607261
+I1203 14:36:58.636108 137274321021824 train.py:125] NOTE: Steps:78500/112603 [69.7%]
+Walltime:5d16h46m (0s eval)
+ETA:2d11h24m
+Total train time:8d4h9m
+I1203 14:42:10.420445 137274321021824 utils.py:1231] [78550] l2_params = 262.7571859772214
+I1203 14:42:10.420751 137274321021824 utils.py:1231] [78550] train/loss = 1.814851924777031
+I1203 14:42:10.420910 137274321021824 utils.py:1231] [78550] l2_grads = 2.115628957748413
+I1203 14:42:10.420992 137274321021824 utils.py:1231] [78550] lr = 0.00024805355712703724
+I1203 14:42:10.421056 137274321021824 utils.py:1231] [78550] uptime = 492719.78341606003
+I1203 14:42:10.421118 137274321021824 utils.py:1231] [78550] examples_seen = 80435200.0
+I1203 14:42:10.421178 137274321021824 utils.py:1231] [78550] progress = 0.6975835457314636
+I1203 14:42:10.421236 137274321021824 utils.py:1231] [78550] epoch = 62.78275978073116
+I1203 14:42:10.421297 137274321021824 utils.py:1231] [78550] img/sec/core = 164.21554178362035
+I1203 14:42:10.421366 137274321021824 utils.py:1231] [78550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.83228890210447
+I1203 14:42:10.421425 137274321021824 utils.py:1231] [78550] core_hours = 136.83228890210447
+I1203 14:42:10.421494 137274321021824 train.py:125] NOTE: Steps:78550/112603 [69.8%]
+Walltime:5d16h51m (0s eval)
+ETA:2d11h19m
+Total train time:8d4h9m
+I1203 14:47:22.209440 137274321021824 utils.py:1231] [78600] l2_params = 262.6686502913038
+I1203 14:47:22.209694 137274321021824 utils.py:1231] [78600] train/loss = 3.5180200338363647
+I1203 14:47:22.209789 137274321021824 utils.py:1231] [78600] l2_grads = 1.9548231363296509
+I1203 14:47:22.209850 137274321021824 utils.py:1231] [78600] lr = 0.0002473926630586772
+I1203 14:47:22.209907 137274321021824 utils.py:1231] [78600] uptime = 493031.57226859103
+I1203 14:47:22.209960 137274321021824 utils.py:1231] [78600] examples_seen = 80486400.0
+I1203 14:47:22.210011 137274321021824 utils.py:1231] [78600] progress = 0.6980275836345391
+I1203 14:47:22.210059 137274321021824 utils.py:1231] [78600] epoch = 62.82272334520012
+I1203 14:47:22.210112 137274321021824 utils.py:1231] [78600] img/sec/core = 164.21369649484126
+I1203 14:47:22.210169 137274321021824 utils.py:1231] [78600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 136.91889691669638
+I1203 14:47:22.210220 137274321021824 utils.py:1231] [78600] core_hours = 136.91889691669638
+I1203 14:47:22.210282 137274321021824 train.py:125] NOTE: Steps:78600/112603 [69.8%]
+Walltime:5d16h57m (0s eval)
+ETA:2d11h14m
+Total train time:8d4h9m
+I1203 14:52:33.869090 137274321021824 utils.py:1231] [78650] l2_params = 262.58770265712593
+I1203 14:52:33.869303 137274321021824 utils.py:1231] [78650] train/loss = 2.0235267728567123
+I1203 14:52:33.869413 137274321021824 utils.py:1231] [78650] l2_grads = 2.1022536754608154
+I1203 14:52:33.869484 137274321021824 utils.py:1231] [78650] lr = 0.00024673236105003123
+I1203 14:52:33.869546 137274321021824 utils.py:1231] [78650] uptime = 493343.23190725304
+I1203 14:52:33.869607 137274321021824 utils.py:1231] [78650] examples_seen = 80537600.0
+I1203 14:52:33.869674 137274321021824 utils.py:1231] [78650] progress = 0.6984716215376144
+I1203 14:52:33.869731 137274321021824 utils.py:1231] [78650] epoch = 62.86268690966907
+I1203 14:52:33.869791 137274321021824 utils.py:1231] [78650] img/sec/core = 164.28177937896535
+I1203 14:52:33.869856 137274321021824 utils.py:1231] [78650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.00546903854695
+I1203 14:52:33.869939 137274321021824 utils.py:1231] [78650] core_hours = 137.00546903854695
+I1203 14:52:33.870030 137274321021824 train.py:125] NOTE: Steps:78650/112603 [69.8%]
+Walltime:5d17h2m (0s eval)
+ETA:2d11h8m
+Total train time:8d4h9m
+I1203 14:57:45.652792 137274321021824 utils.py:1231] [78700] l2_params = 262.51186301866284
+I1203 14:57:45.653036 137274321021824 utils.py:1231] [78700] train/loss = 3.1696263253688812
+I1203 14:57:45.653140 137274321021824 utils.py:1231] [78700] l2_grads = 1.9052311182022095
+I1203 14:57:45.653211 137274321021824 utils.py:1231] [78700] lr = 0.00024607265264871167
+I1203 14:57:45.653271 137274321021824 utils.py:1231] [78700] uptime = 493655.015632294
+I1203 14:57:45.653332 137274321021824 utils.py:1231] [78700] examples_seen = 80588800.0
+I1203 14:57:45.653388 137274321021824 utils.py:1231] [78700] progress = 0.6989156594406899
+I1203 14:57:45.653446 137274321021824 utils.py:1231] [78700] epoch = 62.902650474138035
+I1203 14:57:45.653505 137274321021824 utils.py:1231] [78700] img/sec/core = 164.21639709793354
+I1203 14:57:45.653567 137274321021824 utils.py:1231] [78700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.0920756288361
+I1203 14:57:45.653623 137274321021824 utils.py:1231] [78700] core_hours = 137.0920756288361
+I1203 14:57:45.653705 137274321021824 train.py:125] NOTE: Steps:78700/112603 [69.9%]
+Walltime:5d17h7m (0s eval)
+ETA:2d11h3m
+Total train time:8d4h9m
+I1203 15:02:57.439195 137274321021824 utils.py:1231] [78750] l2_params = 262.4328617364929
+I1203 15:02:57.439419 137274321021824 utils.py:1231] [78750] train/loss = 2.239465609192848
+I1203 15:02:57.439527 137274321021824 utils.py:1231] [78750] l2_grads = 2.0064964294433594
+I1203 15:02:57.439601 137274321021824 utils.py:1231] [78750] lr = 0.00024541353940093954
+I1203 15:02:57.439689 137274321021824 utils.py:1231] [78750] uptime = 493966.802047704
+I1203 15:02:57.439781 137274321021824 utils.py:1231] [78750] examples_seen = 80640000.0
+I1203 15:02:57.439852 137274321021824 utils.py:1231] [78750] progress = 0.6993596973437652
+I1203 15:02:57.439924 137274321021824 utils.py:1231] [78750] epoch = 62.94261403860699
+I1203 15:02:57.439999 137274321021824 utils.py:1231] [78750] img/sec/core = 164.21498009357668
+I1203 15:02:57.440077 137274321021824 utils.py:1231] [78750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.17868296645
+I1203 15:02:57.440139 137274321021824 utils.py:1231] [78750] core_hours = 137.17868296645
+I1203 15:02:57.440218 137274321021824 train.py:125] NOTE: Steps:78750/112603 [69.9%]
+Walltime:5d17h12m (0s eval)
+ETA:2d10h58m
+Total train time:8d4h9m
+I1203 15:08:09.236570 137274321021824 utils.py:1231] [78800] l2_params = 262.34872617063337
+I1203 15:08:09.236821 137274321021824 utils.py:1231] [78800] train/loss = 3.632328599691391
+I1203 15:08:09.236951 137274321021824 utils.py:1231] [78800] l2_grads = 1.994191288948059
+I1203 15:08:09.237044 137274321021824 utils.py:1231] [78800] lr = 0.0002447550228515409
+I1203 15:08:09.237127 137274321021824 utils.py:1231] [78800] uptime = 494278.59948834695
+I1203 15:08:09.237198 137274321021824 utils.py:1231] [78800] examples_seen = 80691200.0
+I1203 15:08:09.237265 137274321021824 utils.py:1231] [78800] progress = 0.6998037352468407
+I1203 15:08:09.237327 137274321021824 utils.py:1231] [78800] epoch = 62.982577603075946
+I1203 15:08:09.237394 137274321021824 utils.py:1231] [78800] img/sec/core = 164.20917341213777
+I1203 15:08:09.237469 137274321021824 utils.py:1231] [78800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.26529336662858
+I1203 15:08:09.237528 137274321021824 utils.py:1231] [78800] core_hours = 137.26529336662858
+I1203 15:08:09.237615 137274321021824 train.py:125] NOTE: Steps:78800/112603 [70.0%]
+Walltime:5d17h17m (0s eval)
+ETA:2d10h53m
+Total train time:8d4h9m
+I1203 15:13:20.996174 137274321021824 utils.py:1231] [78850] l2_params = 262.2637098772558
+I1203 15:13:20.996387 137274321021824 utils.py:1231] [78850] train/loss = 1.889158457517624
+I1203 15:13:20.996483 137274321021824 utils.py:1231] [78850] l2_grads = 2.0793817043304443
+I1203 15:13:20.996546 137274321021824 utils.py:1231] [78850] lr = 0.00024409710454394347
+I1203 15:13:20.996600 137274321021824 utils.py:1231] [78850] uptime = 494590.358961342
+I1203 15:13:20.996654 137274321021824 utils.py:1231] [78850] examples_seen = 80742400.0
+I1203 15:13:20.996706 137274321021824 utils.py:1231] [78850] progress = 0.700247773149916
+I1203 15:13:20.996757 137274321021824 utils.py:1231] [78850] epoch = 63.0225411675449
+I1203 15:13:20.996809 137274321021824 utils.py:1231] [78850] img/sec/core = 164.22917163710514
+I1203 15:13:20.996865 137274321021824 utils.py:1231] [78850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.35189322023834
+I1203 15:13:20.996923 137274321021824 utils.py:1231] [78850] core_hours = 137.35189322023834
+I1203 15:13:20.996985 137274321021824 train.py:125] NOTE: Steps:78850/112603 [70.0%]
+Walltime:5d17h23m (0s eval)
+ETA:2d10h47m
+Total train time:8d4h9m
+I1203 15:18:32.792732 137274321021824 utils.py:1231] [78900] l2_params = 262.1853518783359
+I1203 15:18:32.792940 137274321021824 utils.py:1231] [78900] train/loss = 3.7519496977329254
+I1203 15:18:32.793051 137274321021824 utils.py:1231] [78900] l2_grads = 2.0303072929382324
+I1203 15:18:32.793122 137274321021824 utils.py:1231] [78900] lr = 0.0002434397860201725
+I1203 15:18:32.793183 137274321021824 utils.py:1231] [78900] uptime = 494902.15554457204
+I1203 15:18:32.793248 137274321021824 utils.py:1231] [78900] examples_seen = 80793600.0
+I1203 15:18:32.793305 137274321021824 utils.py:1231] [78900] progress = 0.7006918110529915
+I1203 15:18:32.793361 137274321021824 utils.py:1231] [78900] epoch = 63.062504732013856
+I1203 15:18:32.793417 137274321021824 utils.py:1231] [78900] img/sec/core = 164.20962497279086
+I1203 15:18:32.793479 137274321021824 utils.py:1231] [78900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.4385033822467
+I1203 15:18:32.793542 137274321021824 utils.py:1231] [78900] core_hours = 137.4385033822467
+I1203 15:18:32.793608 137274321021824 train.py:125] NOTE: Steps:78900/112603 [70.1%]
+Walltime:5d17h28m (0s eval)
+ETA:2d10h42m
+Total train time:8d4h9m
+I1203 15:23:44.590924 137274321021824 utils.py:1231] [78950] l2_params = 262.10499338754374
+I1203 15:23:44.591184 137274321021824 utils.py:1231] [78950] train/loss = 2.0248575806617737
+I1203 15:23:44.591364 137274321021824 utils.py:1231] [78950] l2_grads = 2.1114120483398438
+I1203 15:23:44.591457 137274321021824 utils.py:1231] [78950] lr = 0.00024278306882084724
+I1203 15:23:44.591539 137274321021824 utils.py:1231] [78950] uptime = 495213.95388855296
+I1203 15:23:44.591607 137274321021824 utils.py:1231] [78950] examples_seen = 80844800.0
+I1203 15:23:44.591665 137274321021824 utils.py:1231] [78950] progress = 0.7011358489560668
+I1203 15:23:44.591723 137274321021824 utils.py:1231] [78950] epoch = 63.10246829648282
+I1203 15:23:44.591784 137274321021824 utils.py:1231] [78950] img/sec/core = 164.20869766753307
+I1203 15:23:44.591847 137274321021824 utils.py:1231] [78950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.52511403335248
+I1203 15:23:44.591910 137274321021824 utils.py:1231] [78950] core_hours = 137.52511403335248
+I1203 15:23:44.591978 137274321021824 train.py:125] NOTE: Steps:78950/112603 [70.1%]
+Walltime:5d17h33m (0s eval)
+ETA:2d10h37m
+Total train time:8d4h9m
+I1203 15:28:56.328552 137274321021824 utils.py:1231] [79000] l2_params = 262.0231685900354
+I1203 15:28:56.328785 137274321021824 utils.py:1231] [79000] train/loss = 2.67590269446373
+I1203 15:28:56.328913 137274321021824 utils.py:1231] [79000] l2_grads = 1.8681517839431763
+I1203 15:28:56.328993 137274321021824 utils.py:1231] [79000] lr = 0.00024212695448517824
+I1203 15:28:56.329055 137274321021824 utils.py:1231] [79000] uptime = 495525.691415839
+I1203 15:28:56.329116 137274321021824 utils.py:1231] [79000] examples_seen = 80896000.0
+I1203 15:28:56.329182 137274321021824 utils.py:1231] [79000] progress = 0.7015798868591423
+I1203 15:28:56.329258 137274321021824 utils.py:1231] [79000] epoch = 63.142431860951774
+I1203 15:28:56.329324 137274321021824 utils.py:1231] [79000] img/sec/core = 164.2407330478972
+I1203 15:28:56.329404 137274321021824 utils.py:1231] [79000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.61170779093194
+I1203 15:28:56.329469 137274321021824 utils.py:1231] [79000] core_hours = 137.61170779093194
+I1203 15:28:56.329545 137274321021824 train.py:125] NOTE: Steps:79000/112603 [70.2%]
+Walltime:5d17h38m (0s eval)
+ETA:2d10h32m
+Total train time:8d4h9m
+I1203 15:34:08.415601 137274321021824 utils.py:1231] [79050] l2_params = 261.9424280926365
+I1203 15:34:08.415848 137274321021824 utils.py:1231] [79050] train/loss = 1.8341010510921478
+I1203 15:34:08.415984 137274321021824 utils.py:1231] [79050] l2_grads = 2.023517370223999
+I1203 15:34:08.416061 137274321021824 utils.py:1231] [79050] lr = 0.00024147144455096304
+I1203 15:34:08.416127 137274321021824 utils.py:1231] [79050] uptime = 495837.77848888596
+I1203 15:34:08.416182 137274321021824 utils.py:1231] [79050] examples_seen = 80947200.0
+I1203 15:34:08.416232 137274321021824 utils.py:1231] [79050] progress = 0.7020239247622178
+I1203 15:34:08.416281 137274321021824 utils.py:1231] [79050] epoch = 63.18239542542073
+I1203 15:34:08.416332 137274321021824 utils.py:1231] [79050] img/sec/core = 164.05677909093671
+I1203 15:34:08.416407 137274321021824 utils.py:1231] [79050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.6983986445561
+I1203 15:34:08.416457 137274321021824 utils.py:1231] [79050] core_hours = 137.6983986445561
+I1203 15:34:08.416517 137274321021824 train.py:125] NOTE: Steps:79050/112603 [70.2%]
+Walltime:5d17h43m (0s eval)
+ETA:2d10h26m
+Total train time:8d4h8m
+I1203 15:39:20.159661 137274321021824 utils.py:1231] [79100] l2_params = 261.8636677891641
+I1203 15:39:20.159908 137274321021824 utils.py:1231] [79100] train/loss = 1.890823245048523
+I1203 15:39:20.160010 137274321021824 utils.py:1231] [79100] l2_grads = 2.1555893421173096
+I1203 15:39:20.160071 137274321021824 utils.py:1231] [79100] lr = 0.0002408165405545818
+I1203 15:39:20.160121 137274321021824 utils.py:1231] [79100] uptime = 496149.52248368494
+I1203 15:39:20.160174 137274321021824 utils.py:1231] [79100] examples_seen = 80998400.0
+I1203 15:39:20.160221 137274321021824 utils.py:1231] [79100] progress = 0.7024679626652931
+I1203 15:39:20.160268 137274321021824 utils.py:1231] [79100] epoch = 63.222358989889685
+I1203 15:39:20.160317 137274321021824 utils.py:1231] [79100] img/sec/core = 164.2373256717118
+I1203 15:39:20.160377 137274321021824 utils.py:1231] [79100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.78499419866694
+I1203 15:39:20.160428 137274321021824 utils.py:1231] [79100] core_hours = 137.78499419866694
+I1203 15:39:20.160488 137274321021824 train.py:125] NOTE: Steps:79100/112603 [70.2%]
+Walltime:5d17h49m (0s eval)
+ETA:2d10h21m
+Total train time:8d4h8m
+I1203 15:44:31.948716 137274321021824 utils.py:1231] [79150] l2_params = 261.78533613252085
+I1203 15:44:31.948934 137274321021824 utils.py:1231] [79150] train/loss = 1.8389790654182434
+I1203 15:44:31.949040 137274321021824 utils.py:1231] [79150] l2_grads = 2.217482328414917
+I1203 15:44:31.949102 137274321021824 utils.py:1231] [79150] lr = 0.00024016224403099526
+I1203 15:44:31.949170 137274321021824 utils.py:1231] [79150] uptime = 496461.311527977
+I1203 15:44:31.949246 137274321021824 utils.py:1231] [79150] examples_seen = 81049600.0
+I1203 15:44:31.949304 137274321021824 utils.py:1231] [79150] progress = 0.7029120005683686
+I1203 15:44:31.949355 137274321021824 utils.py:1231] [79150] epoch = 63.26232255435864
+I1203 15:44:31.949407 137274321021824 utils.py:1231] [79150] img/sec/core = 164.21359549772865
+I1203 15:44:31.949464 137274321021824 utils.py:1231] [79150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.87160226652583
+I1203 15:44:31.949515 137274321021824 utils.py:1231] [79150] core_hours = 137.87160226652583
+I1203 15:44:31.949579 137274321021824 train.py:125] NOTE: Steps:79150/112603 [70.3%]
+Walltime:5d17h54m (0s eval)
+ETA:2d10h16m
+Total train time:8d4h8m
+I1203 15:49:43.730911 137274321021824 utils.py:1231] [79200] l2_params = 261.7095762116226
+I1203 15:49:43.731132 137274321021824 utils.py:1231] [79200] train/loss = 2.525950849056244
+I1203 15:49:43.731236 137274321021824 utils.py:1231] [79200] l2_grads = 2.0506796836853027
+I1203 15:49:43.731315 137274321021824 utils.py:1231] [79200] lr = 0.00023950855651373991
+I1203 15:49:43.731408 137274321021824 utils.py:1231] [79200] uptime = 496773.09376657894
+I1203 15:49:43.731480 137274321021824 utils.py:1231] [79200] examples_seen = 81100800.0
+I1203 15:49:43.731539 137274321021824 utils.py:1231] [79200] progress = 0.7033560384714439
+I1203 15:49:43.731597 137274321021824 utils.py:1231] [79200] epoch = 63.3022861188276
+I1203 15:49:43.731656 137274321021824 utils.py:1231] [79200] img/sec/core = 164.2171800086673
+I1203 15:49:43.731719 137274321021824 utils.py:1231] [79200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 137.95820844391525
+I1203 15:49:43.731777 137274321021824 utils.py:1231] [79200] core_hours = 137.95820844391525
+I1203 15:49:43.731845 137274321021824 train.py:125] NOTE: Steps:79200/112603 [70.3%]
+Walltime:5d17h59m (0s eval)
+ETA:2d10h11m
+Total train time:8d4h8m
+I1203 15:54:55.462790 137274321021824 utils.py:1231] [79250] l2_params = 261.6317026206518
+I1203 15:54:55.463010 137274321021824 utils.py:1231] [79250] train/loss = 1.8909723907709122
+I1203 15:54:55.463117 137274321021824 utils.py:1231] [79250] l2_grads = 2.1095616817474365
+I1203 15:54:55.463196 137274321021824 utils.py:1231] [79250] lr = 0.00023885547953492506
+I1203 15:54:55.463254 137274321021824 utils.py:1231] [79250] uptime = 497084.82561596495
+I1203 15:54:55.463326 137274321021824 utils.py:1231] [79250] examples_seen = 81152000.0
+I1203 15:54:55.463392 137274321021824 utils.py:1231] [79250] progress = 0.7038000763745194
+I1203 15:54:55.463464 137274321021824 utils.py:1231] [79250] epoch = 63.34224968329656
+I1203 15:54:55.463552 137274321021824 utils.py:1231] [79250] img/sec/core = 164.2437245371123
+I1203 15:54:55.463634 137274321021824 utils.py:1231] [79250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.04480062430028
+I1203 15:54:55.463701 137274321021824 utils.py:1231] [79250] core_hours = 138.04480062430028
+I1203 15:54:55.463786 137274321021824 train.py:125] NOTE: Steps:79250/112603 [70.4%]
+Walltime:5d18h4m (0s eval)
+ETA:2d10h5m
+Total train time:8d4h8m
+I1203 16:00:07.266049 137274321021824 utils.py:1231] [79300] l2_params = 261.55234526163133
+I1203 16:00:07.266276 137274321021824 utils.py:1231] [79300] train/loss = 1.7671802192926407
+I1203 16:00:07.266422 137274321021824 utils.py:1231] [79300] l2_grads = 2.0694284439086914
+I1203 16:00:07.266525 137274321021824 utils.py:1231] [79300] lr = 0.00023820301462522918
+I1203 16:00:07.266617 137274321021824 utils.py:1231] [79300] uptime = 497396.628976276
+I1203 16:00:07.266688 137274321021824 utils.py:1231] [79300] examples_seen = 81203200.0
+I1203 16:00:07.266756 137274321021824 utils.py:1231] [79300] progress = 0.7042441142775947
+I1203 16:00:07.266823 137274321021824 utils.py:1231] [79300] epoch = 63.38221324776551
+I1203 16:00:07.266905 137274321021824 utils.py:1231] [79300] img/sec/core = 164.20605585814891
+I1203 16:00:07.266984 137274321021824 utils.py:1231] [79300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.13141266883113
+I1203 16:00:07.267057 137274321021824 utils.py:1231] [79300] core_hours = 138.13141266883113
+I1203 16:00:07.267129 137274321021824 train.py:125] NOTE: Steps:79300/112603 [70.4%]
+Walltime:5d18h9m (0s eval)
+ETA:2d10h0m
+Total train time:8d4h8m
+I1203 16:05:19.055266 137274321021824 utils.py:1231] [79350] l2_params = 261.4695681594152
+I1203 16:05:19.055523 137274321021824 utils.py:1231] [79350] train/loss = 2.4019422829151154
+I1203 16:05:19.055641 137274321021824 utils.py:1231] [79350] l2_grads = 1.9061111211776733
+I1203 16:05:19.055724 137274321021824 utils.py:1231] [79350] lr = 0.00023755116331389576
+I1203 16:05:19.055789 137274321021824 utils.py:1231] [79350] uptime = 497708.418147595
+I1203 16:05:19.055850 137274321021824 utils.py:1231] [79350] examples_seen = 81254400.0
+I1203 16:05:19.055905 137274321021824 utils.py:1231] [79350] progress = 0.7046881521806702
+I1203 16:05:19.055968 137274321021824 utils.py:1231] [79350] epoch = 63.42217681223447
+I1203 16:05:19.056021 137274321021824 utils.py:1231] [79350] img/sec/core = 164.21352859499297
+I1203 16:05:19.056087 137274321021824 utils.py:1231] [79350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.2180207719753
+I1203 16:05:19.056147 137274321021824 utils.py:1231] [79350] core_hours = 138.2180207719753
+I1203 16:05:19.056213 137274321021824 train.py:125] NOTE: Steps:79350/112603 [70.5%]
+Walltime:5d18h15m (0s eval)
+ETA:2d9h55m
+Total train time:8d4h8m
+I1203 16:10:30.709618 137274321021824 utils.py:1231] [79400] l2_params = 261.391591375798
+I1203 16:10:30.709856 137274321021824 utils.py:1231] [79400] train/loss = 3.7900404036045074
+I1203 16:10:30.710005 137274321021824 utils.py:1231] [79400] l2_grads = 1.9370566606521606
+I1203 16:10:30.710126 137274321021824 utils.py:1231] [79400] lr = 0.00023689992712873064
+I1203 16:10:30.710224 137274321021824 utils.py:1231] [79400] uptime = 498020.07257203397
+I1203 16:10:30.710325 137274321021824 utils.py:1231] [79400] examples_seen = 81305600.0
+I1203 16:10:30.710411 137274321021824 utils.py:1231] [79400] progress = 0.7051321900837455
+I1203 16:10:30.710505 137274321021824 utils.py:1231] [79400] epoch = 63.46214037670343
+I1203 16:10:30.710589 137274321021824 utils.py:1231] [79400] img/sec/core = 164.28452794203574
+I1203 16:10:30.710669 137274321021824 utils.py:1231] [79400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.30459144543053
+I1203 16:10:30.710735 137274321021824 utils.py:1231] [79400] core_hours = 138.30459144543053
+I1203 16:10:30.710808 137274321021824 train.py:125] NOTE: Steps:79400/112603 [70.5%]
+Walltime:5d18h20m (0s eval)
+ETA:2d9h50m
+Total train time:8d4h8m
+I1203 16:15:42.501005 137274321021824 utils.py:1231] [79450] l2_params = 261.3166407250462
+I1203 16:15:42.501238 137274321021824 utils.py:1231] [79450] train/loss = 1.8507986813783646
+I1203 16:15:42.501363 137274321021824 utils.py:1231] [79450] l2_grads = 2.0684895515441895
+I1203 16:15:42.501457 137274321021824 utils.py:1231] [79450] lr = 0.00023624930759609736
+I1203 16:15:42.501523 137274321021824 utils.py:1231] [79450] uptime = 498331.86388497404
+I1203 16:15:42.501597 137274321021824 utils.py:1231] [79450] examples_seen = 81356800.0
+I1203 16:15:42.501653 137274321021824 utils.py:1231] [79450] progress = 0.705576227986821
+I1203 16:15:42.501708 137274321021824 utils.py:1231] [79450] epoch = 63.502103941172386
+I1203 16:15:42.501764 137274321021824 utils.py:1231] [79450] img/sec/core = 164.21240065094867
+I1203 16:15:42.501823 137274321021824 utils.py:1231] [79450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.39120014346946
+I1203 16:15:42.501878 137274321021824 utils.py:1231] [79450] core_hours = 138.39120014346946
+I1203 16:15:42.501948 137274321021824 train.py:125] NOTE: Steps:79450/112603 [70.6%]
+Walltime:5d18h25m (0s eval)
+ETA:2d9h44m
+Total train time:8d4h8m
+I1203 16:20:54.292189 137274321021824 utils.py:1231] [79500] l2_params = 261.23722697054853
+I1203 16:20:54.292441 137274321021824 utils.py:1231] [79500] train/loss = 1.6922021508216858
+I1203 16:20:54.292569 137274321021824 utils.py:1231] [79500] l2_grads = 2.0528578758239746
+I1203 16:20:54.292662 137274321021824 utils.py:1231] [79500] lr = 0.00023559930624091473
+I1203 16:20:54.292739 137274321021824 utils.py:1231] [79500] uptime = 498643.65509547794
+I1203 16:20:54.292810 137274321021824 utils.py:1231] [79500] examples_seen = 81408000.0
+I1203 16:20:54.292897 137274321021824 utils.py:1231] [79500] progress = 0.7060202658898964
+I1203 16:20:54.292959 137274321021824 utils.py:1231] [79500] epoch = 63.54206750564134
+I1203 16:20:54.293020 137274321021824 utils.py:1231] [79500] img/sec/core = 164.21245460144124
+I1203 16:20:54.293087 137274321021824 utils.py:1231] [79500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.47780881305385
+I1203 16:20:54.293145 137274321021824 utils.py:1231] [79500] core_hours = 138.47780881305385
+I1203 16:20:54.293212 137274321021824 train.py:125] NOTE: Steps:79500/112603 [70.6%]
+Walltime:5d18h30m (0s eval)
+ETA:2d9h39m
+Total train time:8d4h8m
+I1203 16:26:06.081641 137274321021824 utils.py:1231] [79550] l2_params = 261.15659616061
+I1203 16:26:06.081919 137274321021824 utils.py:1231] [79550] train/loss = 1.9505027681589127
+I1203 16:26:06.082113 137274321021824 utils.py:1231] [79550] l2_grads = 2.119974136352539
+I1203 16:26:06.082205 137274321021824 utils.py:1231] [79550] lr = 0.00023494992458665244
+I1203 16:26:06.082277 137274321021824 utils.py:1231] [79550] uptime = 498955.44463601196
+I1203 16:26:06.082353 137274321021824 utils.py:1231] [79550] examples_seen = 81459200.0
+I1203 16:26:06.082415 137274321021824 utils.py:1231] [79550] progress = 0.7064643037929718
+I1203 16:26:06.082478 137274321021824 utils.py:1231] [79550] epoch = 63.5820310701103
+I1203 16:26:06.082546 137274321021824 utils.py:1231] [79550] img/sec/core = 164.2133341365657
+I1203 16:26:06.082628 137274321021824 utils.py:1231] [79550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.56441701875778
+I1203 16:26:06.082684 137274321021824 utils.py:1231] [79550] core_hours = 138.56441701875778
+I1203 16:26:06.082761 137274321021824 train.py:125] NOTE: Steps:79550/112603 [70.6%]
+Walltime:5d18h35m (0s eval)
+ETA:2d9h34m
+Total train time:8d4h8m
+I1203 16:31:17.880250 137274321021824 utils.py:1231] [79600] l2_params = 261.075057507975
+I1203 16:31:17.880457 137274321021824 utils.py:1231] [79600] train/loss = 4.113889932632446
+I1203 16:31:17.880554 137274321021824 utils.py:1231] [79600] l2_grads = 2.1112074851989746
+I1203 16:31:17.880614 137274321021824 utils.py:1231] [79600] lr = 0.00023430116415532753
+I1203 16:31:17.880677 137274321021824 utils.py:1231] [79600] uptime = 499267.24303893896
+I1203 16:31:17.880731 137274321021824 utils.py:1231] [79600] examples_seen = 81510400.0
+I1203 16:31:17.880780 137274321021824 utils.py:1231] [79600] progress = 0.7069083416960472
+I1203 16:31:17.880826 137274321021824 utils.py:1231] [79600] epoch = 63.62199463457925
+I1203 16:31:17.880876 137274321021824 utils.py:1231] [79600] img/sec/core = 164.2086666235653
+I1203 16:31:17.880937 137274321021824 utils.py:1231] [79600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.65102768623748
+I1203 16:31:17.880989 137274321021824 utils.py:1231] [79600] core_hours = 138.65102768623748
+I1203 16:31:17.881047 137274321021824 train.py:125] NOTE: Steps:79600/112603 [70.7%]
+Walltime:5d18h41m (0s eval)
+ETA:2d9h29m
+Total train time:8d4h8m
+I1203 16:36:29.685253 137274321021824 utils.py:1231] [79650] l2_params = 260.9917174150732
+I1203 16:36:29.685457 137274321021824 utils.py:1231] [79650] train/loss = 2.188215419650078
+I1203 16:36:29.685555 137274321021824 utils.py:1231] [79650] l2_grads = 2.070688486099243
+I1203 16:36:29.685622 137274321021824 utils.py:1231] [79650] lr = 0.00023365302646750138
+I1203 16:36:29.685676 137274321021824 utils.py:1231] [79650] uptime = 499579.04803731595
+I1203 16:36:29.685729 137274321021824 utils.py:1231] [79650] examples_seen = 81561600.0
+I1203 16:36:29.685784 137274321021824 utils.py:1231] [79650] progress = 0.7073523795991226
+I1203 16:36:29.685835 137274321021824 utils.py:1231] [79650] epoch = 63.661958199048215
+I1203 16:36:29.685891 137274321021824 utils.py:1231] [79650] img/sec/core = 164.20519320250435
+I1203 16:36:29.685950 137274321021824 utils.py:1231] [79650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.73764018578666
+I1203 16:36:29.686002 137274321021824 utils.py:1231] [79650] core_hours = 138.73764018578666
+I1203 16:36:29.686068 137274321021824 train.py:125] NOTE: Steps:79650/112603 [70.7%]
+Walltime:5d18h46m (0s eval)
+ETA:2d9h24m
+Total train time:8d4h8m
+I1203 16:41:41.621718 137274321021824 utils.py:1231] [79700] l2_params = 260.91394599699265
+I1203 16:41:41.621937 137274321021824 utils.py:1231] [79700] train/loss = 4.120353311300278
+I1203 16:41:41.622043 137274321021824 utils.py:1231] [79700] l2_grads = 2.0148351192474365
+I1203 16:41:41.622113 137274321021824 utils.py:1231] [79700] lr = 0.0002330055130422755
+I1203 16:41:41.622184 137274321021824 utils.py:1231] [79700] uptime = 499890.98454490094
+I1203 16:41:41.622246 137274321021824 utils.py:1231] [79700] examples_seen = 81612800.0
+I1203 16:41:41.622307 137274321021824 utils.py:1231] [79700] progress = 0.707796417502198
+I1203 16:41:41.622365 137274321021824 utils.py:1231] [79700] epoch = 63.70192176351717
+I1203 16:41:41.622428 137274321021824 utils.py:1231] [79700] img/sec/core = 164.13596598996872
+I1203 16:41:41.622493 137274321021824 utils.py:1231] [79700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.82428921567137
+I1203 16:41:41.622590 137274321021824 utils.py:1231] [79700] core_hours = 138.82428921567137
+I1203 16:41:41.622665 137274321021824 train.py:125] NOTE: Steps:79700/112603 [70.8%]
+Walltime:5d18h51m (0s eval)
+ETA:2d9h18m
+Total train time:8d4h8m
+I1203 16:46:53.350227 137274321021824 utils.py:1231] [79750] l2_params = 260.844451089641
+I1203 16:46:53.350477 137274321021824 utils.py:1231] [79750] train/loss = 3.6524171233177185
+I1203 16:46:53.350612 137274321021824 utils.py:1231] [79750] l2_grads = 2.0398964881896973
+I1203 16:46:53.350694 137274321021824 utils.py:1231] [79750] lr = 0.00023235862539728848
+I1203 16:46:53.350756 137274321021824 utils.py:1231] [79750] uptime = 500202.71311752393
+I1203 16:46:53.350817 137274321021824 utils.py:1231] [79750] examples_seen = 81664000.0
+I1203 16:46:53.350874 137274321021824 utils.py:1231] [79750] progress = 0.7082404554052734
+I1203 16:46:53.350947 137274321021824 utils.py:1231] [79750] epoch = 63.741885327986125
+I1203 16:46:53.351005 137274321021824 utils.py:1231] [79750] img/sec/core = 164.2454509998424
+I1203 16:46:53.351063 137274321021824 utils.py:1231] [79750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.91088048584442
+I1203 16:46:53.351118 137274321021824 utils.py:1231] [79750] core_hours = 138.91088048584442
+I1203 16:46:53.351183 137274321021824 train.py:125] NOTE: Steps:79750/112603 [70.8%]
+Walltime:5d18h56m (0s eval)
+ETA:2d9h13m
+Total train time:8d4h8m
+I1203 16:52:05.134975 137274321021824 utils.py:1231] [79800] l2_params = 260.76314619695233
+I1203 16:52:05.135281 137274321021824 utils.py:1231] [79800] train/loss = 1.8961880058050156
+I1203 16:52:05.135437 137274321021824 utils.py:1231] [79800] l2_grads = 2.0145814418792725
+I1203 16:52:05.135502 137274321021824 utils.py:1231] [79800] lr = 0.00023171236504871163
+I1203 16:52:05.135571 137274321021824 utils.py:1231] [79800] uptime = 500514.497933259
+I1203 16:52:05.135630 137274321021824 utils.py:1231] [79800] examples_seen = 81715200.0
+I1203 16:52:05.135684 137274321021824 utils.py:1231] [79800] progress = 0.7086844933083488
+I1203 16:52:05.135737 137274321021824 utils.py:1231] [79800] epoch = 63.78184889245508
+I1203 16:52:05.135797 137274321021824 utils.py:1231] [79800] img/sec/core = 164.21582263167733
+I1203 16:52:05.135857 137274321021824 utils.py:1231] [79800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 138.99748737910417
+I1203 16:52:05.135916 137274321021824 utils.py:1231] [79800] core_hours = 138.99748737910417
+I1203 16:52:05.135980 137274321021824 train.py:125] NOTE: Steps:79800/112603 [70.9%]
+Walltime:5d19h1m (0s eval)
+ETA:2d9h8m
+Total train time:8d4h8m
+I1203 16:57:16.928794 137274321021824 utils.py:1231] [79850] l2_params = 260.68637875370916
+I1203 16:57:16.929166 137274321021824 utils.py:1231] [79850] train/loss = 2.0827097594738007
+I1203 16:57:16.929326 137274321021824 utils.py:1231] [79850] l2_grads = 2.160465955734253
+I1203 16:57:16.929400 137274321021824 utils.py:1231] [79850] lr = 0.0002310667335112469
+I1203 16:57:16.929466 137274321021824 utils.py:1231] [79850] uptime = 500826.29182763794
+I1203 16:57:16.929531 137274321021824 utils.py:1231] [79850] examples_seen = 81766400.0
+I1203 16:57:16.929590 137274321021824 utils.py:1231] [79850] progress = 0.7091285312114242
+I1203 16:57:16.929648 137274321021824 utils.py:1231] [79850] epoch = 63.821812456924036
+I1203 16:57:16.929708 137274321021824 utils.py:1231] [79850] img/sec/core = 164.2110410852845
+I1203 16:57:16.929773 137274321021824 utils.py:1231] [79850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.08409679420942
+I1203 16:57:16.929832 137274321021824 utils.py:1231] [79850] core_hours = 139.08409679420942
+I1203 16:57:16.929907 137274321021824 train.py:125] NOTE: Steps:79850/112603 [70.9%]
+Walltime:5d19h7m (0s eval)
+ETA:2d9h3m
+Total train time:8d4h8m
+I1203 17:02:28.725317 137274321021824 utils.py:1231] [79900] l2_params = 260.60052958538586
+I1203 17:02:28.725538 137274321021824 utils.py:1231] [79900] train/loss = 1.8140412271022797
+I1203 17:02:28.725671 137274321021824 utils.py:1231] [79900] l2_grads = 2.0232577323913574
+I1203 17:02:28.725762 137274321021824 utils.py:1231] [79900] lr = 0.0002304217322981221
+I1203 17:02:28.725841 137274321021824 utils.py:1231] [79900] uptime = 501138.08819845994
+I1203 17:02:28.725925 137274321021824 utils.py:1231] [79900] examples_seen = 81817600.0
+I1203 17:02:28.725998 137274321021824 utils.py:1231] [79900] progress = 0.7095725691144996
+I1203 17:02:28.726071 137274321021824 utils.py:1231] [79900] epoch = 63.861776021393
+I1203 17:02:28.726131 137274321021824 utils.py:1231] [79900] img/sec/core = 164.20973683888232
+I1203 17:02:28.726203 137274321021824 utils.py:1231] [79900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.17070689721552
+I1203 17:02:28.726260 137274321021824 utils.py:1231] [79900] core_hours = 139.17070689721552
+I1203 17:02:28.726332 137274321021824 train.py:125] NOTE: Steps:79900/112603 [71.0%]
+Walltime:5d19h12m (0s eval)
+ETA:2d8h57m
+Total train time:8d4h8m
+I1203 17:07:40.512488 137274321021824 utils.py:1231] [79950] l2_params = 260.5190980358639
+I1203 17:07:40.512701 137274321021824 utils.py:1231] [79950] train/loss = 1.9796302020549774
+I1203 17:07:40.512817 137274321021824 utils.py:1231] [79950] l2_grads = 1.9904347658157349
+I1203 17:07:40.512898 137274321021824 utils.py:1231] [79950] lr = 0.00022977736292108738
+I1203 17:07:40.512959 137274321021824 utils.py:1231] [79950] uptime = 501449.87532089604
+I1203 17:07:40.513021 137274321021824 utils.py:1231] [79950] examples_seen = 81868800.0
+I1203 17:07:40.513078 137274321021824 utils.py:1231] [79950] progress = 0.7100166070175751
+I1203 17:07:40.513134 137274321021824 utils.py:1231] [79950] epoch = 63.901739585861954
+I1203 17:07:40.513203 137274321021824 utils.py:1231] [79950] img/sec/core = 164.21460771040412
+I1203 17:07:40.513273 137274321021824 utils.py:1231] [79950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.25731443122555
+I1203 17:07:40.513329 137274321021824 utils.py:1231] [79950] core_hours = 139.25731443122555
+I1203 17:07:40.513406 137274321021824 train.py:125] NOTE: Steps:79950/112603 [71.0%]
+Walltime:5d19h17m (0s eval)
+ETA:2d8h52m
+Total train time:8d4h8m
+I1203 17:12:52.314792 137274321021824 utils.py:1231] [80000] l2_params = 260.44170012427514
+I1203 17:12:52.315036 137274321021824 utils.py:1231] [80000] train/loss = 4.257516324520111
+I1203 17:12:52.315206 137274321021824 utils.py:1231] [80000] l2_grads = 2.170921564102173
+I1203 17:12:52.315293 137274321021824 utils.py:1231] [80000] lr = 0.0002291336268904125
+I1203 17:12:52.315368 137274321021824 utils.py:1231] [80000] uptime = 501761.677725701
+I1203 17:12:52.315447 137274321021824 utils.py:1231] [80000] examples_seen = 81920000.0
+I1203 17:12:52.315520 137274321021824 utils.py:1231] [80000] progress = 0.7104606449206504
+I1203 17:12:52.315592 137274321021824 utils.py:1231] [80000] epoch = 63.94170315033091
+I1203 17:12:52.315673 137274321021824 utils.py:1231] [80000] img/sec/core = 164.20655906109891
+I1203 17:12:52.315745 137274321021824 utils.py:1231] [80000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.34392621033805
+I1203 17:12:52.315808 137274321021824 utils.py:1231] [80000] core_hours = 139.34392621033805
+I1203 17:12:52.315889 137274321021824 train.py:125] NOTE: Steps:80000/112603 [71.0%]
+Walltime:5d19h22m (0s eval)
+ETA:2d8h47m
+Total train time:8d4h8m
+I1203 17:12:52.677161 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:80000/112603 [71.0%]
+Walltime:5d19h22m (0s eval)
+ETA:2d8h47m
+Total train time:8d4h8m
+I1203 17:14:30.265570 137274321021824 utils.py:1231] [80000] val/acc@1 = 0.7196867028061225
+I1203 17:14:30.265902 137274321021824 utils.py:1231] [80000] val/loss = 1.1079247362759648
+I1203 17:14:30.266165 137274321021824 utils.py:1231] [80000] z/secs/eval/val = 97.58872587105725
+I1203 17:14:30.266294 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.58872587105725
+I1203 17:19:42.047993 137274321021824 utils.py:1231] [80050] l2_params = 260.3680747287152
+I1203 17:19:42.048231 137274321021824 utils.py:1231] [80050] train/loss = 1.8473752290010452
+I1203 17:19:42.048337 137274321021824 utils.py:1231] [80050] l2_grads = 2.288043737411499
+I1203 17:19:42.048409 137274321021824 utils.py:1231] [80050] lr = 0.00022849052571488222
+I1203 17:19:42.048469 137274321021824 utils.py:1231] [80050] uptime = 502171.410830299
+I1203 17:19:42.048533 137274321021824 utils.py:1231] [80050] examples_seen = 81971200.0
+I1203 17:19:42.048590 137274321021824 utils.py:1231] [80050] progress = 0.7109046828237259
+I1203 17:19:42.048646 137274321021824 utils.py:1231] [80050] epoch = 63.981666714799864
+I1203 17:19:42.048705 137274321021824 utils.py:1231] [80050] img/sec/core = 124.95939289609618
+I1203 17:19:42.048775 137274321021824 utils.py:1231] [80050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.4577409616153
+I1203 17:19:42.048830 137274321021824 utils.py:1231] [80050] core_hours = 139.4577409616153
+I1203 17:19:42.048901 137274321021824 train.py:125] NOTE: Steps:80050/112603 [71.1%]
+Walltime:5d19h29m (0s eval)
+ETA:2d8h42m
+Total train time:8d4h10m
+I1203 17:24:53.847430 137274321021824 utils.py:1231] [80100] l2_params = 260.288145821492
+I1203 17:24:53.847712 137274321021824 utils.py:1231] [80100] train/loss = 1.8855964243412018
+I1203 17:24:53.847901 137274321021824 utils.py:1231] [80100] l2_grads = 2.122713088989258
+I1203 17:24:53.847990 137274321021824 utils.py:1231] [80100] lr = 0.00022784806090179338
+I1203 17:24:53.848057 137274321021824 utils.py:1231] [80100] uptime = 502483.21041819896
+I1203 17:24:53.848129 137274321021824 utils.py:1231] [80100] examples_seen = 82022400.0
+I1203 17:24:53.848201 137274321021824 utils.py:1231] [80100] progress = 0.7113487207268012
+I1203 17:24:53.848268 137274321021824 utils.py:1231] [80100] epoch = 64.02163027926882
+I1203 17:24:53.848341 137274321021824 utils.py:1231] [80100] img/sec/core = 164.20804255979775
+I1203 17:24:53.848449 137274321021824 utils.py:1231] [80100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.54435195825414
+I1203 17:24:53.848520 137274321021824 utils.py:1231] [80100] core_hours = 139.54435195825414
+I1203 17:24:53.848615 137274321021824 train.py:125] NOTE: Steps:80100/112603 [71.1%]
+Walltime:5d19h34m (0s eval)
+ETA:2d8h37m
+Total train time:8d4h10m
+I1203 17:30:05.642307 137274321021824 utils.py:1231] [80150] l2_params = 260.2090014602344
+I1203 17:30:05.642645 137274321021824 utils.py:1231] [80150] train/loss = 2.802168130874634
+I1203 17:30:05.642870 137274321021824 utils.py:1231] [80150] l2_grads = 1.9123265743255615
+I1203 17:30:05.643013 137274321021824 utils.py:1231] [80150] lr = 0.0002272062339569521
+I1203 17:30:05.643097 137274321021824 utils.py:1231] [80150] uptime = 502795.005455601
+I1203 17:30:05.643191 137274321021824 utils.py:1231] [80150] examples_seen = 82073600.0
+I1203 17:30:05.643268 137274321021824 utils.py:1231] [80150] progress = 0.7117927586298767
+I1203 17:30:05.643356 137274321021824 utils.py:1231] [80150] epoch = 64.06159384373778
+I1203 17:30:05.643436 137274321021824 utils.py:1231] [80150] img/sec/core = 164.2104390968078
+I1203 17:30:05.643513 137274321021824 utils.py:1231] [80150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.63096169086583
+I1203 17:30:05.643577 137274321021824 utils.py:1231] [80150] core_hours = 139.63096169086583
+I1203 17:30:05.643662 137274321021824 train.py:125] NOTE: Steps:80150/112603 [71.2%]
+Walltime:5d19h39m (0s eval)
+ETA:2d8h32m
+Total train time:8d4h10m
+I1203 17:35:17.432418 137274321021824 utils.py:1231] [80200] l2_params = 260.1348013659347
+I1203 17:35:17.432662 137274321021824 utils.py:1231] [80200] train/loss = 3.290058970451355
+I1203 17:35:17.432780 137274321021824 utils.py:1231] [80200] l2_grads = 1.8870372772216797
+I1203 17:35:17.432863 137274321021824 utils.py:1231] [80200] lr = 0.0002265650463846686
+I1203 17:35:17.432932 137274321021824 utils.py:1231] [80200] uptime = 503106.795292853
+I1203 17:35:17.432986 137274321021824 utils.py:1231] [80200] examples_seen = 82124800.0
+I1203 17:35:17.433037 137274321021824 utils.py:1231] [80200] progress = 0.712236796532952
+I1203 17:35:17.433087 137274321021824 utils.py:1231] [80200] epoch = 64.10155740820673
+I1203 17:35:17.433140 137274321021824 utils.py:1231] [80200] img/sec/core = 164.21317786127318
+I1203 17:35:17.433198 137274321021824 utils.py:1231] [80200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.7175699789914
+I1203 17:35:17.433250 137274321021824 utils.py:1231] [80200] core_hours = 139.7175699789914
+I1203 17:35:17.433313 137274321021824 train.py:125] NOTE: Steps:80200/112603 [71.2%]
+Walltime:5d19h45m (0s eval)
+ETA:2d8h27m
+Total train time:8d4h10m
+I1203 17:40:29.222328 137274321021824 utils.py:1231] [80250] l2_params = 260.06315690670186
+I1203 17:40:29.222594 137274321021824 utils.py:1231] [80250] train/loss = 1.6333881467580795
+I1203 17:40:29.222719 137274321021824 utils.py:1231] [80250] l2_grads = 2.083259105682373
+I1203 17:40:29.222804 137274321021824 utils.py:1231] [80250] lr = 0.00022592449968775468
+I1203 17:40:29.222863 137274321021824 utils.py:1231] [80250] uptime = 503418.585225024
+I1203 17:40:29.222927 137274321021824 utils.py:1231] [80250] examples_seen = 82176000.0
+I1203 17:40:29.222975 137274321021824 utils.py:1231] [80250] progress = 0.7126808344360275
+I1203 17:40:29.223023 137274321021824 utils.py:1231] [80250] epoch = 64.14152097267569
+I1203 17:40:29.223073 137274321021824 utils.py:1231] [80250] img/sec/core = 164.21312786943898
+I1203 17:40:29.223130 137274321021824 utils.py:1231] [80250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.80417829348335
+I1203 17:40:29.223180 137274321021824 utils.py:1231] [80250] core_hours = 139.80417829348335
+I1203 17:40:29.223239 137274321021824 train.py:125] NOTE: Steps:80250/112603 [71.3%]
+Walltime:5d19h50m (0s eval)
+ETA:2d8h21m
+Total train time:8d4h10m
+I1203 17:45:41.011288 137274321021824 utils.py:1231] [80300] l2_params = 259.9837514612036
+I1203 17:45:41.011540 137274321021824 utils.py:1231] [80300] train/loss = 3.014646351337433
+I1203 17:45:41.011677 137274321021824 utils.py:1231] [80300] l2_grads = 1.993574619293213
+I1203 17:45:41.011764 137274321021824 utils.py:1231] [80300] lr = 0.0002252845953675204
+I1203 17:45:41.011841 137274321021824 utils.py:1231] [80300] uptime = 503730.37420187297
+I1203 17:45:41.011910 137274321021824 utils.py:1231] [80300] examples_seen = 82227200.0
+I1203 17:45:41.011968 137274321021824 utils.py:1231] [80300] progress = 0.7131248723391028
+I1203 17:45:41.012025 137274321021824 utils.py:1231] [80300] epoch = 64.18148453714466
+I1203 17:45:41.012091 137274321021824 utils.py:1231] [80300] img/sec/core = 164.21363101879658
+I1203 17:45:41.012158 137274321021824 utils.py:1231] [80300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.89078634260804
+I1203 17:45:41.012236 137274321021824 utils.py:1231] [80300] core_hours = 139.89078634260804
+I1203 17:45:41.012307 137274321021824 train.py:125] NOTE: Steps:80300/112603 [71.3%]
+Walltime:5d19h55m (0s eval)
+ETA:2d8h16m
+Total train time:8d4h10m
+I1203 17:50:52.799065 137274321021824 utils.py:1231] [80350] l2_params = 259.9093822497135
+I1203 17:50:52.799284 137274321021824 utils.py:1231] [80350] train/loss = 1.8306463807821274
+I1203 17:50:52.799391 137274321021824 utils.py:1231] [80350] l2_grads = 2.067176103591919
+I1203 17:50:52.799463 137274321021824 utils.py:1231] [80350] lr = 0.0002246453349237701
+I1203 17:50:52.799550 137274321021824 utils.py:1231] [80350] uptime = 504042.161896953
+I1203 17:50:52.799618 137274321021824 utils.py:1231] [80350] examples_seen = 82278400.0
+I1203 17:50:52.799676 137274321021824 utils.py:1231] [80350] progress = 0.7135689102421783
+I1203 17:50:52.799748 137274321021824 utils.py:1231] [80350] epoch = 64.2214481016136
+I1203 17:50:52.799819 137274321021824 utils.py:1231] [80350] img/sec/core = 164.2143061061341
+I1203 17:50:52.799910 137274321021824 utils.py:1231] [80350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 139.97739403568585
+I1203 17:50:52.799973 137274321021824 utils.py:1231] [80350] core_hours = 139.97739403568585
+I1203 17:50:52.800037 137274321021824 train.py:125] NOTE: Steps:80350/112603 [71.4%]
+Walltime:5d20h0m (0s eval)
+ETA:2d8h11m
+Total train time:8d4h10m
+I1203 17:56:04.586016 137274321021824 utils.py:1231] [80400] l2_params = 259.828047729945
+I1203 17:56:04.586224 137274321021824 utils.py:1231] [80400] train/loss = 1.7580437064170837
+I1203 17:56:04.586324 137274321021824 utils.py:1231] [80400] l2_grads = 2.0012288093566895
+I1203 17:56:04.586397 137274321021824 utils.py:1231] [80400] lr = 0.00022400671985479918
+I1203 17:56:04.586480 137274321021824 utils.py:1231] [80400] uptime = 504353.94883161003
+I1203 17:56:04.586571 137274321021824 utils.py:1231] [80400] examples_seen = 82329600.0
+I1203 17:56:04.586643 137274321021824 utils.py:1231] [80400] progress = 0.7140129481452537
+I1203 17:56:04.586709 137274321021824 utils.py:1231] [80400] epoch = 64.26141166608257
+I1203 17:56:04.586786 137274321021824 utils.py:1231] [80400] img/sec/core = 164.21470661149408
+I1203 17:56:04.586859 137274321021824 utils.py:1231] [80400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.064001517535
+I1203 17:56:04.586933 137274321021824 utils.py:1231] [80400] core_hours = 140.064001517535
+I1203 17:56:04.587008 137274321021824 train.py:125] NOTE: Steps:80400/112603 [71.4%]
+Walltime:5d20h5m (0s eval)
+ETA:2d8h6m
+Total train time:8d4h10m
+I1203 18:01:16.374843 137274321021824 utils.py:1231] [80450] l2_params = 259.7489716795869
+I1203 18:01:16.375077 137274321021824 utils.py:1231] [80450] train/loss = 1.8181100934743881
+I1203 18:01:16.375184 137274321021824 utils.py:1231] [80450] l2_grads = 2.2100188732147217
+I1203 18:01:16.375263 137274321021824 utils.py:1231] [80450] lr = 0.0002233687516573901
+I1203 18:01:16.375360 137274321021824 utils.py:1231] [80450] uptime = 504665.737713109
+I1203 18:01:16.375449 137274321021824 utils.py:1231] [80450] examples_seen = 82380800.0
+I1203 18:01:16.375509 137274321021824 utils.py:1231] [80450] progress = 0.7144569860483291
+I1203 18:01:16.375565 137274321021824 utils.py:1231] [80450] epoch = 64.30137523055151
+I1203 18:01:16.375630 137274321021824 utils.py:1231] [80450] img/sec/core = 164.21368123792115
+I1203 18:01:16.375707 137274321021824 utils.py:1231] [80450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.1506095401736
+I1203 18:01:16.375762 137274321021824 utils.py:1231] [80450] core_hours = 140.1506095401736
+I1203 18:01:16.375841 137274321021824 train.py:125] NOTE: Steps:80450/112603 [71.4%]
+Walltime:5d20h11m (0s eval)
+ETA:2d8h0m
+Total train time:8d4h10m
+I1203 18:06:28.172691 137274321021824 utils.py:1231] [80500] l2_params = 259.6819787858128
+I1203 18:06:28.172913 137274321021824 utils.py:1231] [80500] train/loss = 1.7919449657201767
+I1203 18:06:28.173016 137274321021824 utils.py:1231] [80500] l2_grads = 2.098679304122925
+I1203 18:06:28.173089 137274321021824 utils.py:1231] [80500] lr = 0.00022273143182680884
+I1203 18:06:28.173144 137274321021824 utils.py:1231] [80500] uptime = 504977.535505875
+I1203 18:06:28.173202 137274321021824 utils.py:1231] [80500] examples_seen = 82432000.0
+I1203 18:06:28.173255 137274321021824 utils.py:1231] [80500] progress = 0.7149010239514045
+I1203 18:06:28.173308 137274321021824 utils.py:1231] [80500] epoch = 64.34133879502048
+I1203 18:06:28.173365 137274321021824 utils.py:1231] [80500] img/sec/core = 164.2089879655537
+I1203 18:06:28.173426 137274321021824 utils.py:1231] [80500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.23722003816417
+I1203 18:06:28.173481 137274321021824 utils.py:1231] [80500] core_hours = 140.23722003816417
+I1203 18:06:28.173546 137274321021824 train.py:125] NOTE: Steps:80500/112603 [71.5%]
+Walltime:5d20h16m (0s eval)
+ETA:2d7h55m
+Total train time:8d4h10m
+I1203 18:11:39.961328 137274321021824 utils.py:1231] [80550] l2_params = 259.6005458266004
+I1203 18:11:39.961579 137274321021824 utils.py:1231] [80550] train/loss = 1.7883732467889786
+I1203 18:11:39.961716 137274321021824 utils.py:1231] [80550] l2_grads = 2.1500515937805176
+I1203 18:11:39.961805 137274321021824 utils.py:1231] [80550] lr = 0.00022209476185680246
+I1203 18:11:39.961870 137274321021824 utils.py:1231] [80550] uptime = 505289.324232541
+I1203 18:11:39.961935 137274321021824 utils.py:1231] [80550] examples_seen = 82483200.0
+I1203 18:11:39.962004 137274321021824 utils.py:1231] [80550] progress = 0.7153450618544799
+I1203 18:11:39.962058 137274321021824 utils.py:1231] [80550] epoch = 64.38130235948944
+I1203 18:11:39.962111 137274321021824 utils.py:1231] [80550] img/sec/core = 164.21376278573922
+I1203 18:11:39.962177 137274321021824 utils.py:1231] [80550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.3238280177936
+I1203 18:11:39.962227 137274321021824 utils.py:1231] [80550] core_hours = 140.3238280177936
+I1203 18:11:39.962285 137274321021824 train.py:125] NOTE: Steps:80550/112603 [71.5%]
+Walltime:5d20h21m (0s eval)
+ETA:2d7h50m
+Total train time:8d4h10m
+I1203 18:16:51.659236 137274321021824 utils.py:1231] [80600] l2_params = 259.5187301916212
+I1203 18:16:51.659462 137274321021824 utils.py:1231] [80600] train/loss = 1.8934387266635895
+I1203 18:16:51.659605 137274321021824 utils.py:1231] [80600] l2_grads = 2.1741795539855957
+I1203 18:16:51.659708 137274321021824 utils.py:1231] [80600] lr = 0.00022145874323959466
+I1203 18:16:51.659765 137274321021824 utils.py:1231] [80600] uptime = 505601.02212688094
+I1203 18:16:51.659820 137274321021824 utils.py:1231] [80600] examples_seen = 82534400.0
+I1203 18:16:51.659871 137274321021824 utils.py:1231] [80600] progress = 0.7157890997575553
+I1203 18:16:51.659935 137274321021824 utils.py:1231] [80600] epoch = 64.42126592395839
+I1203 18:16:51.659988 137274321021824 utils.py:1231] [80600] img/sec/core = 164.26161655158617
+I1203 18:16:51.660045 137274321021824 utils.py:1231] [80600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.4104107662214
+I1203 18:16:51.660097 137274321021824 utils.py:1231] [80600] core_hours = 140.4104107662214
+I1203 18:16:51.660158 137274321021824 train.py:125] NOTE: Steps:80600/112603 [71.6%]
+Walltime:5d20h26m (0s eval)
+ETA:2d7h45m
+Total train time:8d4h9m
+I1203 18:22:03.452526 137274321021824 utils.py:1231] [80650] l2_params = 259.4459295410512
+I1203 18:22:03.452736 137274321021824 utils.py:1231] [80650] train/loss = 2.876444488763809
+I1203 18:22:03.452842 137274321021824 utils.py:1231] [80650] l2_grads = 1.9329735040664673
+I1203 18:22:03.452956 137274321021824 utils.py:1231] [80650] lr = 0.00022082337746588262
+I1203 18:22:03.453056 137274321021824 utils.py:1231] [80650] uptime = 505912.81540917105
+I1203 18:22:03.453168 137274321021824 utils.py:1231] [80650] examples_seen = 82585600.0
+I1203 18:22:03.453249 137274321021824 utils.py:1231] [80650] progress = 0.7162331376606307
+I1203 18:22:03.453331 137274321021824 utils.py:1231] [80650] epoch = 64.46122948842735
+I1203 18:22:03.453400 137274321021824 utils.py:1231] [80650] img/sec/core = 164.2113634518955
+I1203 18:22:03.453486 137274321021824 utils.py:1231] [80650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.49702001130197
+I1203 18:22:03.453561 137274321021824 utils.py:1231] [80650] core_hours = 140.49702001130197
+I1203 18:22:03.453667 137274321021824 train.py:125] NOTE: Steps:80650/112603 [71.6%]
+Walltime:5d20h31m (0s eval)
+ETA:2d7h39m
+Total train time:8d4h9m
+I1203 18:27:15.250562 137274321021824 utils.py:1231] [80700] l2_params = 259.36724948368123
+I1203 18:27:15.250768 137274321021824 utils.py:1231] [80700] train/loss = 3.6193292140960693
+I1203 18:27:15.250865 137274321021824 utils.py:1231] [80700] l2_grads = 2.1155781745910645
+I1203 18:27:15.250946 137274321021824 utils.py:1231] [80700] lr = 0.00022018866602483293
+I1203 18:27:15.251006 137274321021824 utils.py:1231] [80700] uptime = 506224.613367136
+I1203 18:27:15.251065 137274321021824 utils.py:1231] [80700] examples_seen = 82636800.0
+I1203 18:27:15.251122 137274321021824 utils.py:1231] [80700] progress = 0.7166771755637061
+I1203 18:27:15.251183 137274321021824 utils.py:1231] [80700] epoch = 64.5011930528963
+I1203 18:27:15.251241 137274321021824 utils.py:1231] [80700] img/sec/core = 164.208900963223
+I1203 18:27:15.251302 137274321021824 utils.py:1231] [80700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.58363055518112
+I1203 18:27:15.251360 137274321021824 utils.py:1231] [80700] core_hours = 140.58363055518112
+I1203 18:27:15.251430 137274321021824 train.py:125] NOTE: Steps:80700/112603 [71.7%]
+Walltime:5d20h37m (0s eval)
+ETA:2d7h34m
+Total train time:8d4h9m
+I1203 18:32:27.031328 137274321021824 utils.py:1231] [80750] l2_params = 259.29375680529495
+I1203 18:32:27.031610 137274321021824 utils.py:1231] [80750] train/loss = 1.9588608145713806
+I1203 18:32:27.031834 137274321021824 utils.py:1231] [80750] l2_grads = 2.0098042488098145
+I1203 18:32:27.031972 137274321021824 utils.py:1231] [80750] lr = 0.00021955461040407874
+I1203 18:32:27.032068 137274321021824 utils.py:1231] [80750] uptime = 506536.39442539704
+I1203 18:32:27.032178 137274321021824 utils.py:1231] [80750] examples_seen = 82688000.0
+I1203 18:32:27.032262 137274321021824 utils.py:1231] [80750] progress = 0.7171212134667815
+I1203 18:32:27.032350 137274321021824 utils.py:1231] [80750] epoch = 64.54115661736526
+I1203 18:32:27.032444 137274321021824 utils.py:1231] [80750] img/sec/core = 164.21780170214893
+I1203 18:32:27.032525 137274321021824 utils.py:1231] [80750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.67023640469807
+I1203 18:32:27.032608 137274321021824 utils.py:1231] [80750] core_hours = 140.67023640469807
+I1203 18:32:27.032699 137274321021824 train.py:125] NOTE: Steps:80750/112603 [71.7%]
+Walltime:5d20h42m (0s eval)
+ETA:2d7h29m
+Total train time:8d4h9m
+I1203 18:37:38.823674 137274321021824 utils.py:1231] [80800] l2_params = 259.21536498225095
+I1203 18:37:38.823946 137274321021824 utils.py:1231] [80800] train/loss = 2.8639156818389893
+I1203 18:37:38.824181 137274321021824 utils.py:1231] [80800] l2_grads = 1.8276501893997192
+I1203 18:37:38.824314 137274321021824 utils.py:1231] [80800] lr = 0.0002189212120897165
+I1203 18:37:38.824410 137274321021824 utils.py:1231] [80800] uptime = 506848.186761188
+I1203 18:37:38.824501 137274321021824 utils.py:1231] [80800] examples_seen = 82739200.0
+I1203 18:37:38.824594 137274321021824 utils.py:1231] [80800] progress = 0.7175652513698569
+I1203 18:37:38.824680 137274321021824 utils.py:1231] [80800] epoch = 64.58112018183422
+I1203 18:37:38.824745 137274321021824 utils.py:1231] [80800] img/sec/core = 164.21186194368667
+I1203 18:37:38.824822 137274321021824 utils.py:1231] [80800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.7568453868622
+I1203 18:37:38.824890 137274321021824 utils.py:1231] [80800] core_hours = 140.7568453868622
+I1203 18:37:38.824977 137274321021824 train.py:125] NOTE: Steps:80800/112603 [71.8%]
+Walltime:5d20h47m (0s eval)
+ETA:2d7h24m
+Total train time:8d4h9m
+I1203 18:42:50.625871 137274321021824 utils.py:1231] [80850] l2_params = 259.1417475571698
+I1203 18:42:50.626162 137274321021824 utils.py:1231] [80850] train/loss = 1.8787714391946793
+I1203 18:42:50.626399 137274321021824 utils.py:1231] [80850] l2_grads = 2.17073392868042
+I1203 18:42:50.626499 137274321021824 utils.py:1231] [80850] lr = 0.0002182884725663018
+I1203 18:42:50.626570 137274321021824 utils.py:1231] [80850] uptime = 507159.98893093795
+I1203 18:42:50.626635 137274321021824 utils.py:1231] [80850] examples_seen = 82790400.0
+I1203 18:42:50.626692 137274321021824 utils.py:1231] [80850] progress = 0.7180092892729323
+I1203 18:42:50.626747 137274321021824 utils.py:1231] [80850] epoch = 64.62108374630317
+I1203 18:42:50.626797 137274321021824 utils.py:1231] [80850] img/sec/core = 164.20668284975415
+I1203 18:42:50.626868 137274321021824 utils.py:1231] [80850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.84345710068163
+I1203 18:42:50.626944 137274321021824 utils.py:1231] [80850] core_hours = 140.84345710068163
+I1203 18:42:50.627025 137274321021824 train.py:125] NOTE: Steps:80850/112603 [71.8%]
+Walltime:5d20h52m (0s eval)
+ETA:2d7h18m
+Total train time:8d4h9m
+I1203 18:48:02.331700 137274321021824 utils.py:1231] [80900] l2_params = 259.06885936349425
+I1203 18:48:02.331958 137274321021824 utils.py:1231] [80900] train/loss = 3.485304832458496
+I1203 18:48:02.332051 137274321021824 utils.py:1231] [80900] l2_grads = 2.037235736846924
+I1203 18:48:02.332125 137274321021824 utils.py:1231] [80900] lr = 0.00021765639331684593
+I1203 18:48:02.332177 137274321021824 utils.py:1231] [80900] uptime = 507471.69453943195
+I1203 18:48:02.332226 137274321021824 utils.py:1231] [80900] examples_seen = 82841600.0
+I1203 18:48:02.332272 137274321021824 utils.py:1231] [80900] progress = 0.7184533271760077
+I1203 18:48:02.332319 137274321021824 utils.py:1231] [80900] epoch = 64.66104731077213
+I1203 18:48:02.332367 137274321021824 utils.py:1231] [80900] img/sec/core = 164.2575513715401
+I1203 18:48:02.332422 137274321021824 utils.py:1231] [80900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 140.93004199192998
+I1203 18:48:02.332471 137274321021824 utils.py:1231] [80900] core_hours = 140.93004199192998
+I1203 18:48:02.332530 137274321021824 train.py:125] NOTE: Steps:80900/112603 [71.8%]
+Walltime:5d20h57m (0s eval)
+ETA:2d7h13m
+Total train time:8d4h9m
+I1203 18:53:14.059888 137274321021824 utils.py:1231] [80950] l2_params = 258.99342129058954
+I1203 18:53:14.060117 137274321021824 utils.py:1231] [80950] train/loss = 1.8026368468999863
+I1203 18:53:14.060233 137274321021824 utils.py:1231] [80950] l2_grads = 2.0595669746398926
+I1203 18:53:14.060307 137274321021824 utils.py:1231] [80950] lr = 0.00021702497582281287
+I1203 18:53:14.060369 137274321021824 utils.py:1231] [80950] uptime = 507783.42272992595
+I1203 18:53:14.060437 137274321021824 utils.py:1231] [80950] examples_seen = 82892800.0
+I1203 18:53:14.060496 137274321021824 utils.py:1231] [80950] progress = 0.7188973650790832
+I1203 18:53:14.060553 137274321021824 utils.py:1231] [80950] epoch = 64.7010108752411
+I1203 18:53:14.060635 137274321021824 utils.py:1231] [80950] img/sec/core = 164.24565233853951
+I1203 18:53:14.060699 137274321021824 utils.py:1231] [80950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.0166331559561
+I1203 18:53:14.060756 137274321021824 utils.py:1231] [80950] core_hours = 141.0166331559561
+I1203 18:53:14.060824 137274321021824 train.py:125] NOTE: Steps:80950/112603 [71.9%]
+Walltime:5d21h3m (0s eval)
+ETA:2d7h8m
+Total train time:8d4h9m
+I1203 18:58:25.837231 137274321021824 utils.py:1231] [81000] l2_params = 258.91538483856476
+I1203 18:58:25.837445 137274321021824 utils.py:1231] [81000] train/loss = 3.622480481863022
+I1203 18:58:25.837546 137274321021824 utils.py:1231] [81000] l2_grads = 1.907812237739563
+I1203 18:58:25.837615 137274321021824 utils.py:1231] [81000] lr = 0.00021639422156411538
+I1203 18:58:25.837673 137274321021824 utils.py:1231] [81000] uptime = 508095.20003501803
+I1203 18:58:25.837733 137274321021824 utils.py:1231] [81000] examples_seen = 82944000.0
+I1203 18:58:25.837790 137274321021824 utils.py:1231] [81000] progress = 0.7193414029821585
+I1203 18:58:25.837845 137274321021824 utils.py:1231] [81000] epoch = 64.74097443971004
+I1203 18:58:25.837907 137274321021824 utils.py:1231] [81000] img/sec/core = 164.21977855276722
+I1203 18:58:25.837970 137274321021824 utils.py:1231] [81000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.1032379629261
+I1203 18:58:25.838024 137274321021824 utils.py:1231] [81000] core_hours = 141.1032379629261
+I1203 18:58:25.838091 137274321021824 train.py:125] NOTE: Steps:81000/112603 [71.9%]
+Walltime:5d21h8m (0s eval)
+ETA:2d7h3m
+Total train time:8d4h9m
+I1203 19:03:37.975037 137274321021824 utils.py:1231] [81050] l2_params = 258.8381885298155
+I1203 19:03:37.975239 137274321021824 utils.py:1231] [81050] train/loss = 1.7793197631835938
+I1203 19:03:37.975338 137274321021824 utils.py:1231] [81050] l2_grads = 2.15633225440979
+I1203 19:03:37.975410 137274321021824 utils.py:1231] [81050] lr = 0.00021576413201911205
+I1203 19:03:37.975467 137274321021824 utils.py:1231] [81050] uptime = 508407.33782857493
+I1203 19:03:37.975527 137274321021824 utils.py:1231] [81050] examples_seen = 82995200.0
+I1203 19:03:37.975601 137274321021824 utils.py:1231] [81050] progress = 0.719785440885234
+I1203 19:03:37.975652 137274321021824 utils.py:1231] [81050] epoch = 64.780938004179
+I1203 19:03:37.975705 137274321021824 utils.py:1231] [81050] img/sec/core = 164.0301208532318
+I1203 19:03:37.975762 137274321021824 utils.py:1231] [81050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.1899429055808
+I1203 19:03:37.975815 137274321021824 utils.py:1231] [81050] core_hours = 141.1899429055808
+I1203 19:03:37.975876 137274321021824 train.py:125] NOTE: Steps:81050/112603 [72.0%]
+Walltime:5d21h13m (0s eval)
+ETA:2d6h58m
+Total train time:8d4h9m
+I1203 19:08:49.757546 137274321021824 utils.py:1231] [81100] l2_params = 258.7683805947005
+I1203 19:08:49.757816 137274321021824 utils.py:1231] [81100] train/loss = 3.534298539161682
+I1203 19:08:49.758022 137274321021824 utils.py:1231] [81100] l2_grads = 1.9240235090255737
+I1203 19:08:49.758137 137274321021824 utils.py:1231] [81100] lr = 0.00021513470866460363
+I1203 19:08:49.758190 137274321021824 utils.py:1231] [81100] uptime = 508719.12055191
+I1203 19:08:49.758244 137274321021824 utils.py:1231] [81100] examples_seen = 83046400.0
+I1203 19:08:49.758294 137274321021824 utils.py:1231] [81100] progress = 0.7202294787883093
+I1203 19:08:49.758344 137274321021824 utils.py:1231] [81100] epoch = 64.82090156864795
+I1203 19:08:49.758397 137274321021824 utils.py:1231] [81100] img/sec/core = 164.21692469782428
+I1203 19:08:49.758451 137274321021824 utils.py:1231] [81100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.2765492176183
+I1203 19:08:49.758500 137274321021824 utils.py:1231] [81100] core_hours = 141.2765492176183
+I1203 19:08:49.758563 137274321021824 train.py:125] NOTE: Steps:81100/112603 [72.0%]
+Walltime:5d21h18m (0s eval)
+ETA:2d6h52m
+Total train time:8d4h9m
+I1203 19:14:01.551123 137274321021824 utils.py:1231] [81150] l2_params = 258.6989508682222
+I1203 19:14:01.551361 137274321021824 utils.py:1231] [81150] train/loss = 1.8567090928554535
+I1203 19:14:01.551459 137274321021824 utils.py:1231] [81150] l2_grads = 2.1717052459716797
+I1203 19:14:01.551551 137274321021824 utils.py:1231] [81150] lr = 0.00021450595297582866
+I1203 19:14:01.551608 137274321021824 utils.py:1231] [81150] uptime = 509030.913965615
+I1203 19:14:01.551661 137274321021824 utils.py:1231] [81150] examples_seen = 83097600.0
+I1203 19:14:01.551713 137274321021824 utils.py:1231] [81150] progress = 0.7206735166913848
+I1203 19:14:01.551763 137274321021824 utils.py:1231] [81150] epoch = 64.86086513311692
+I1203 19:14:01.551815 137274321021824 utils.py:1231] [81150] img/sec/core = 164.21129424000262
+I1203 19:14:01.551874 137274321021824 utils.py:1231] [81150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.36315849920305
+I1203 19:14:01.551932 137274321021824 utils.py:1231] [81150] core_hours = 141.36315849920305
+I1203 19:14:01.551994 137274321021824 train.py:125] NOTE: Steps:81150/112603 [72.1%]
+Walltime:5d21h23m (0s eval)
+ETA:2d6h47m
+Total train time:8d4h9m
+I1203 19:19:13.350919 137274321021824 utils.py:1231] [81200] l2_params = 258.61731442257445
+I1203 19:19:13.351163 137274321021824 utils.py:1231] [81200] train/loss = 1.8229375034570694
+I1203 19:19:13.351269 137274321021824 utils.py:1231] [81200] l2_grads = 2.182213544845581
+I1203 19:19:13.351340 137274321021824 utils.py:1231] [81200] lr = 0.0002138778664264616
+I1203 19:19:13.351409 137274321021824 utils.py:1231] [81200] uptime = 509342.71377017896
+I1203 19:19:13.351474 137274321021824 utils.py:1231] [81200] examples_seen = 83148800.0
+I1203 19:19:13.351532 137274321021824 utils.py:1231] [81200] progress = 0.7211175545944601
+I1203 19:19:13.351587 137274321021824 utils.py:1231] [81200] epoch = 64.90082869758588
+I1203 19:19:13.351646 137274321021824 utils.py:1231] [81200] img/sec/core = 164.2079284545973
+I1203 19:19:13.351729 137274321021824 utils.py:1231] [81200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.44976955602635
+I1203 19:19:13.351787 137274321021824 utils.py:1231] [81200] core_hours = 141.44976955602635
+I1203 19:19:13.351858 137274321021824 train.py:125] NOTE: Steps:81200/112603 [72.1%]
+Walltime:5d21h29m (0s eval)
+ETA:2d6h42m
+Total train time:8d4h9m
+I1203 19:24:25.128068 137274321021824 utils.py:1231] [81250] l2_params = 258.5482474882159
+I1203 19:24:25.128295 137274321021824 utils.py:1231] [81250] train/loss = 4.138487219810486
+I1203 19:24:25.128413 137274321021824 utils.py:1231] [81250] l2_grads = 2.1254324913024902
+I1203 19:24:25.128502 137274321021824 utils.py:1231] [81250] lr = 0.00021325045048860853
+I1203 19:24:25.128566 137274321021824 utils.py:1231] [81250] uptime = 509654.490927133
+I1203 19:24:25.128631 137274321021824 utils.py:1231] [81250] examples_seen = 83200000.0
+I1203 19:24:25.128691 137274321021824 utils.py:1231] [81250] progress = 0.7215615924975356
+I1203 19:24:25.128752 137274321021824 utils.py:1231] [81250] epoch = 64.94079226205483
+I1203 19:24:25.128814 137274321021824 utils.py:1231] [81250] img/sec/core = 164.21985658027552
+I1203 19:24:25.408593 137274321021824 utils.py:1231] [81250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.53637432184692
+I1203 19:24:25.408890 137274321021824 utils.py:1231] [81250] core_hours = 141.53637432184692
+I1203 19:24:25.408997 137274321021824 train.py:125] NOTE: Steps:81250/112603 [72.2%]
+Walltime:5d21h34m (0s eval)
+ETA:2d6h37m
+Total train time:8d4h9m
+I1203 19:29:37.208930 137274321021824 utils.py:1231] [81300] l2_params = 258.47308328266354
+I1203 19:29:37.209180 137274321021824 utils.py:1231] [81300] train/loss = 1.9500934928655624
+I1203 19:29:37.209295 137274321021824 utils.py:1231] [81300] l2_grads = 2.1531476974487305
+I1203 19:29:37.209369 137274321021824 utils.py:1231] [81300] lr = 0.0002126237066328029
+I1203 19:29:37.209438 137274321021824 utils.py:1231] [81300] uptime = 509966.57179903495
+I1203 19:29:37.209493 137274321021824 utils.py:1231] [81300] examples_seen = 83251200.0
+I1203 19:29:37.209543 137274321021824 utils.py:1231] [81300] progress = 0.7220056304006109
+I1203 19:29:37.209591 137274321021824 utils.py:1231] [81300] epoch = 64.98075582652379
+I1203 19:29:37.209643 137274321021824 utils.py:1231] [81300] img/sec/core = 164.06003895069748
+I1203 19:29:37.209705 137274321021824 utils.py:1231] [81300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.6230634529308
+I1203 19:29:37.209760 137274321021824 utils.py:1231] [81300] core_hours = 141.6230634529308
+I1203 19:29:37.209837 137274321021824 train.py:125] NOTE: Steps:81300/112603 [72.2%]
+Walltime:5d21h39m (0s eval)
+ETA:2d6h31m
+Total train time:8d4h9m
+I1203 19:34:48.906607 137274321021824 utils.py:1231] [81350] l2_params = 258.3953950359434
+I1203 19:34:48.906820 137274321021824 utils.py:1231] [81350] train/loss = 4.153364419937134
+I1203 19:34:48.906919 137274321021824 utils.py:1231] [81350] l2_grads = 2.165001392364502
+I1203 19:34:48.906979 137274321021824 utils.py:1231] [81350] lr = 0.00021199763632800432
+I1203 19:34:48.907030 137274321021824 utils.py:1231] [81350] uptime = 510278.269391924
+I1203 19:34:48.907082 137274321021824 utils.py:1231] [81350] examples_seen = 83302400.0
+I1203 19:34:48.907130 137274321021824 utils.py:1231] [81350] progress = 0.7224496683036864
+I1203 19:34:48.907177 137274321021824 utils.py:1231] [81350] epoch = 65.02071939099274
+I1203 19:34:48.907228 137274321021824 utils.py:1231] [81350] img/sec/core = 164.26177541326317
+I1203 19:34:48.907285 137274321021824 utils.py:1231] [81350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.7096461176222
+I1203 19:34:48.907335 137274321021824 utils.py:1231] [81350] core_hours = 141.7096461176222
+I1203 19:34:48.907399 137274321021824 train.py:125] NOTE: Steps:81350/112603 [72.2%]
+Walltime:5d21h44m (0s eval)
+ETA:2d6h26m
+Total train time:8d4h9m
+I1203 19:40:00.656113 137274321021824 utils.py:1231] [81400] l2_params = 258.3250713858953
+I1203 19:40:00.656383 137274321021824 utils.py:1231] [81400] train/loss = 1.7638083398342133
+I1203 19:40:00.656584 137274321021824 utils.py:1231] [81400] l2_grads = 2.177612543106079
+I1203 19:40:00.656705 137274321021824 utils.py:1231] [81400] lr = 0.00021137224104159243
+I1203 19:40:00.656792 137274321021824 utils.py:1231] [81400] uptime = 510590.01915025
+I1203 19:40:00.656877 137274321021824 utils.py:1231] [81400] examples_seen = 83353600.0
+I1203 19:40:00.656973 137274321021824 utils.py:1231] [81400] progress = 0.7228937062067619
+I1203 19:40:00.657063 137274321021824 utils.py:1231] [81400] epoch = 65.0606829554617
+I1203 19:40:00.657135 137274321021824 utils.py:1231] [81400] img/sec/core = 164.23428930604797
+I1203 19:40:00.657224 137274321021824 utils.py:1231] [81400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.79624327271276
+I1203 19:40:00.657297 137274321021824 utils.py:1231] [81400] core_hours = 141.79624327271276
+I1203 19:40:00.657384 137274321021824 train.py:125] NOTE: Steps:81400/112603 [72.3%]
+Walltime:5d21h49m (0s eval)
+ETA:2d6h21m
+Total train time:8d4h9m
+I1203 19:45:12.436027 137274321021824 utils.py:1231] [81450] l2_params = 258.2490680993734
+I1203 19:45:12.436299 137274321021824 utils.py:1231] [81450] train/loss = 3.6958396434783936
+I1203 19:45:12.436434 137274321021824 utils.py:1231] [81450] l2_grads = 1.9453994035720825
+I1203 19:45:12.436522 137274321021824 utils.py:1231] [81450] lr = 0.00021074752223936564
+I1203 19:45:12.436595 137274321021824 utils.py:1231] [81450] uptime = 510901.79895669094
+I1203 19:45:12.436675 137274321021824 utils.py:1231] [81450] examples_seen = 83404800.0
+I1203 19:45:12.436741 137274321021824 utils.py:1231] [81450] progress = 0.7233377441098372
+I1203 19:45:12.436810 137274321021824 utils.py:1231] [81450] epoch = 65.10064651993066
+I1203 19:45:12.436895 137274321021824 utils.py:1231] [81450] img/sec/core = 164.2184610493706
+I1203 19:45:12.436958 137274321021824 utils.py:1231] [81450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.88284877450192
+I1203 19:45:12.437014 137274321021824 utils.py:1231] [81450] core_hours = 141.88284877450192
+I1203 19:45:12.437079 137274321021824 train.py:125] NOTE: Steps:81450/112603 [72.3%]
+Walltime:5d21h55m (0s eval)
+ETA:2d6h16m
+Total train time:8d4h9m
+I1203 19:50:24.224880 137274321021824 utils.py:1231] [81500] l2_params = 258.17679120957195
+I1203 19:50:24.225086 137274321021824 utils.py:1231] [81500] train/loss = 4.1904381811618805
+I1203 19:50:24.225181 137274321021824 utils.py:1231] [81500] l2_grads = 2.2520194053649902
+I1203 19:50:24.225247 137274321021824 utils.py:1231] [81500] lr = 0.0002101234813855365
+I1203 19:50:24.225302 137274321021824 utils.py:1231] [81500] uptime = 511213.587664532
+I1203 19:50:24.225359 137274321021824 utils.py:1231] [81500] examples_seen = 83456000.0
+I1203 19:50:24.225416 137274321021824 utils.py:1231] [81500] progress = 0.7237817820129127
+I1203 19:50:24.225470 137274321021824 utils.py:1231] [81500] epoch = 65.14061008439961
+I1203 19:50:24.225525 137274321021824 utils.py:1231] [81500] img/sec/core = 164.21377270051192
+I1203 19:50:24.225584 137274321021824 utils.py:1231] [81500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 141.96945674890222
+I1203 19:50:24.225639 137274321021824 utils.py:1231] [81500] core_hours = 141.96945674890222
+I1203 19:50:24.225721 137274321021824 train.py:125] NOTE: Steps:81500/112603 [72.4%]
+Walltime:5d22h0m (0s eval)
+ETA:2d6h10m
+Total train time:8d4h9m
+I1203 19:55:36.012008 137274321021824 utils.py:1231] [81550] l2_params = 258.1008147645605
+I1203 19:55:36.012256 137274321021824 utils.py:1231] [81550] train/loss = 3.4922173619270325
+I1203 19:55:36.012367 137274321021824 utils.py:1231] [81550] l2_grads = 1.916327714920044
+I1203 19:55:36.012454 137274321021824 utils.py:1231] [81550] lr = 0.0002095001199427287
+I1203 19:55:36.012518 137274321021824 utils.py:1231] [81550] uptime = 511525.37487879803
+I1203 19:55:36.012582 137274321021824 utils.py:1231] [81550] examples_seen = 83507200.0
+I1203 19:55:36.012642 137274321021824 utils.py:1231] [81550] progress = 0.724225819915988
+I1203 19:55:36.012701 137274321021824 utils.py:1231] [81550] epoch = 65.18057364886857
+I1203 19:55:36.012761 137274321021824 utils.py:1231] [81550] img/sec/core = 164.2145593446807
+I1203 19:55:36.012826 137274321021824 utils.py:1231] [81550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.05606430842056
+I1203 19:55:36.012890 137274321021824 utils.py:1231] [81550] core_hours = 142.05606430842056
+I1203 19:55:36.012962 137274321021824 train.py:125] NOTE: Steps:81550/112603 [72.4%]
+Walltime:5d22h5m (0s eval)
+ETA:2d6h5m
+Total train time:8d4h9m
+I1203 20:00:47.805263 137274321021824 utils.py:1231] [81600] l2_params = 258.0283851106446
+I1203 20:00:47.805511 137274321021824 utils.py:1231] [81600] train/loss = 2.1157850176095963
+I1203 20:00:47.805644 137274321021824 utils.py:1231] [81600] l2_grads = 2.0857913494110107
+I1203 20:00:47.805703 137274321021824 utils.py:1231] [81600] lr = 0.00020887743937197297
+I1203 20:00:47.805757 137274321021824 utils.py:1231] [81600] uptime = 511837.168119733
+I1203 20:00:47.805818 137274321021824 utils.py:1231] [81600] examples_seen = 83558400.0
+I1203 20:00:47.805867 137274321021824 utils.py:1231] [81600] progress = 0.7246698578190635
+I1203 20:00:47.805927 137274321021824 utils.py:1231] [81600] epoch = 65.22053721333752
+I1203 20:00:47.805979 137274321021824 utils.py:1231] [81600] img/sec/core = 164.21138523230647
+I1203 20:00:47.806035 137274321021824 utils.py:1231] [81600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.1426735420136
+I1203 20:00:47.806088 137274321021824 utils.py:1231] [81600] core_hours = 142.1426735420136
+I1203 20:00:47.806154 137274321021824 train.py:125] NOTE: Steps:81600/112603 [72.5%]
+Walltime:5d22h10m (0s eval)
+ETA:2d6h0m
+Total train time:8d4h9m
+I1203 20:05:59.580326 137274321021824 utils.py:1231] [81650] l2_params = 257.95186529684383
+I1203 20:05:59.580579 137274321021824 utils.py:1231] [81650] train/loss = 1.8626775294542313
+I1203 20:05:59.580701 137274321021824 utils.py:1231] [81650] l2_grads = 2.1971805095672607
+I1203 20:05:59.580770 137274321021824 utils.py:1231] [81650] lr = 0.0002082554411327049
+I1203 20:05:59.804372 137274321021824 utils.py:1231] [81650] uptime = 512149.166684891
+I1203 20:05:59.804607 137274321021824 utils.py:1231] [81650] examples_seen = 83609600.0
+I1203 20:05:59.804674 137274321021824 utils.py:1231] [81650] progress = 0.7251138957221388
+I1203 20:05:59.804729 137274321021824 utils.py:1231] [81650] epoch = 65.26050077780648
+I1203 20:05:59.804804 137274321021824 utils.py:1231] [81650] img/sec/core = 164.1033187895368
+I1203 20:05:59.804900 137274321021824 utils.py:1231] [81650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.22933981011303
+I1203 20:05:59.804976 137274321021824 utils.py:1231] [81650] core_hours = 142.22933981011303
+I1203 20:05:59.805043 137274321021824 train.py:125] NOTE: Steps:81650/112603 [72.5%]
+Walltime:5d22h15m (0s eval)
+ETA:2d5h55m
+Total train time:8d4h9m
+I1203 20:11:11.581253 137274321021824 utils.py:1231] [81700] l2_params = 257.86935062317696
+I1203 20:11:11.581464 137274321021824 utils.py:1231] [81700] train/loss = 3.6276052594184875
+I1203 20:11:11.581574 137274321021824 utils.py:1231] [81700] l2_grads = 2.0047171115875244
+I1203 20:11:11.581675 137274321021824 utils.py:1231] [81700] lr = 0.0002076341266827607
+I1203 20:11:11.581773 137274321021824 utils.py:1231] [81700] uptime = 512460.94411383
+I1203 20:11:11.581892 137274321021824 utils.py:1231] [81700] examples_seen = 83660800.0
+I1203 20:11:11.581964 137274321021824 utils.py:1231] [81700] progress = 0.7255579336252143
+I1203 20:11:11.582026 137274321021824 utils.py:1231] [81700] epoch = 65.30046434227545
+I1203 20:11:11.582084 137274321021824 utils.py:1231] [81700] img/sec/core = 164.21971331996653
+I1203 20:11:11.582147 137274321021824 utils.py:1231] [81700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.315944651485
+I1203 20:11:11.582214 137274321021824 utils.py:1231] [81700] core_hours = 142.315944651485
+I1203 20:11:11.582298 137274321021824 train.py:125] NOTE: Steps:81700/112603 [72.6%]
+Walltime:5d22h21m (0s eval)
+ETA:2d5h49m
+Total train time:8d4h9m
+I1203 20:16:23.358175 137274321021824 utils.py:1231] [81750] l2_params = 257.79655226309177
+I1203 20:16:23.358452 137274321021824 utils.py:1231] [81750] train/loss = 1.9728519767522812
+I1203 20:16:23.358615 137274321021824 utils.py:1231] [81750] l2_grads = 2.101475238800049
+I1203 20:16:23.358692 137274321021824 utils.py:1231] [81750] lr = 0.00020701349747837415
+I1203 20:16:23.358753 137274321021824 utils.py:1231] [81750] uptime = 512772.721114082
+I1203 20:16:23.358818 137274321021824 utils.py:1231] [81750] examples_seen = 83712000.0
+I1203 20:16:23.358898 137274321021824 utils.py:1231] [81750] progress = 0.7260019715282896
+I1203 20:16:23.358964 137274321021824 utils.py:1231] [81750] epoch = 65.3404279067444
+I1203 20:16:23.359028 137274321021824 utils.py:1231] [81750] img/sec/core = 164.2199391186965
+I1203 20:16:23.359101 137274321021824 utils.py:1231] [81750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.40254937377722
+I1203 20:16:23.359210 137274321021824 utils.py:1231] [81750] core_hours = 142.40254937377722
+I1203 20:16:23.359301 137274321021824 train.py:125] NOTE: Steps:81750/112603 [72.6%]
+Walltime:5d22h26m (0s eval)
+ETA:2d5h44m
+Total train time:8d4h9m
+I1203 20:21:35.155209 137274321021824 utils.py:1231] [81800] l2_params = 257.71981318231275
+I1203 20:21:35.155470 137274321021824 utils.py:1231] [81800] train/loss = 2.360058158636093
+I1203 20:21:35.155595 137274321021824 utils.py:1231] [81800] l2_grads = 2.0263946056365967
+I1203 20:21:35.155681 137274321021824 utils.py:1231] [81800] lr = 0.00020639355497417228
+I1203 20:21:35.155753 137274321021824 utils.py:1231] [81800] uptime = 513084.51811307797
+I1203 20:21:35.155828 137274321021824 utils.py:1231] [81800] examples_seen = 83763200.0
+I1203 20:21:35.155906 137274321021824 utils.py:1231] [81800] progress = 0.7264460094313651
+I1203 20:21:35.155967 137274321021824 utils.py:1231] [81800] epoch = 65.38039147121336
+I1203 20:21:35.156026 137274321021824 utils.py:1231] [81800] img/sec/core = 164.20940600735852
+I1203 20:21:35.156092 137274321021824 utils.py:1231] [81800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.4891596512761
+I1203 20:21:35.156146 137274321021824 utils.py:1231] [81800] core_hours = 142.4891596512761
+I1203 20:21:35.156213 137274321021824 train.py:125] NOTE: Steps:81800/112603 [72.6%]
+Walltime:5d22h31m (0s eval)
+ETA:2d5h39m
+Total train time:8d4h9m
+I1203 20:26:46.937398 137274321021824 utils.py:1231] [81850] l2_params = 257.6467234550072
+I1203 20:26:46.937613 137274321021824 utils.py:1231] [81850] train/loss = 1.8689833134412766
+I1203 20:26:46.937730 137274321021824 utils.py:1231] [81850] l2_grads = 2.1952779293060303
+I1203 20:26:46.937808 137274321021824 utils.py:1231] [81850] lr = 0.00020577430062317247
+I1203 20:26:46.937875 137274321021824 utils.py:1231] [81850] uptime = 513396.30023515795
+I1203 20:26:46.937968 137274321021824 utils.py:1231] [81850] examples_seen = 83814400.0
+I1203 20:26:46.938029 137274321021824 utils.py:1231] [81850] progress = 0.7268900473344405
+I1203 20:26:46.938088 137274321021824 utils.py:1231] [81850] epoch = 65.4203550356823
+I1203 20:26:46.938147 137274321021824 utils.py:1231] [81850] img/sec/core = 164.21724138135687
+I1203 20:26:46.938210 137274321021824 utils.py:1231] [81850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.5757657962983
+I1203 20:26:46.938269 137274321021824 utils.py:1231] [81850] core_hours = 142.5757657962983
+I1203 20:26:46.938338 137274321021824 train.py:125] NOTE: Steps:81850/112603 [72.7%]
+Walltime:5d22h36m (0s eval)
+ETA:2d5h34m
+Total train time:8d4h8m
+I1203 20:31:58.727784 137274321021824 utils.py:1231] [81900] l2_params = 257.57831438860006
+I1203 20:31:58.727992 137274321021824 utils.py:1231] [81900] train/loss = 1.7514901012182236
+I1203 20:31:58.728095 137274321021824 utils.py:1231] [81900] l2_grads = 2.2656495571136475
+I1203 20:31:58.728156 137274321021824 utils.py:1231] [81900] lr = 0.0002051557358767805
+I1203 20:31:58.728208 137274321021824 utils.py:1231] [81900] uptime = 513708.090570244
+I1203 20:31:58.728262 137274321021824 utils.py:1231] [81900] examples_seen = 83865600.0
+I1203 20:31:58.728316 137274321021824 utils.py:1231] [81900] progress = 0.7273340852375159
+I1203 20:31:58.728366 137274321021824 utils.py:1231] [81900] epoch = 65.46031860015127
+I1203 20:31:58.728417 137274321021824 utils.py:1231] [81900] img/sec/core = 164.21291566291873
+I1203 20:31:58.728473 137274321021824 utils.py:1231] [81900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.66237422271112
+I1203 20:31:58.728523 137274321021824 utils.py:1231] [81900] core_hours = 142.66237422271112
+I1203 20:31:58.728588 137274321021824 train.py:125] NOTE: Steps:81900/112603 [72.7%]
+Walltime:5d22h41m (0s eval)
+ETA:2d5h28m
+Total train time:8d4h8m
+I1203 20:37:10.506117 137274321021824 utils.py:1231] [81950] l2_params = 257.50525457671216
+I1203 20:37:10.506332 137274321021824 utils.py:1231] [81950] train/loss = 1.7965411394834518
+I1203 20:37:10.506435 137274321021824 utils.py:1231] [81950] l2_grads = 2.188347339630127
+I1203 20:37:10.506516 137274321021824 utils.py:1231] [81950] lr = 0.00020453786218478493
+I1203 20:37:10.506578 137274321021824 utils.py:1231] [81950] uptime = 514019.86893965595
+I1203 20:37:10.506640 137274321021824 utils.py:1231] [81950] examples_seen = 83916800.0
+I1203 20:37:10.506699 137274321021824 utils.py:1231] [81950] progress = 0.7277781231405913
+I1203 20:37:10.506761 137274321021824 utils.py:1231] [81950] epoch = 65.50028216462023
+I1203 20:37:10.506821 137274321021824 utils.py:1231] [81950] img/sec/core = 164.2192179546334
+I1203 20:37:10.506890 137274321021824 utils.py:1231] [81950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.74897932532556
+I1203 20:37:10.506950 137274321021824 utils.py:1231] [81950] core_hours = 142.74897932532556
+I1203 20:37:10.507018 137274321021824 train.py:125] NOTE: Steps:81950/112603 [72.8%]
+Walltime:5d22h46m (0s eval)
+ETA:2d5h23m
+Total train time:8d4h8m
+I1203 20:42:22.288989 137274321021824 utils.py:1231] [82000] l2_params = 257.43300650498253
+I1203 20:42:22.289214 137274321021824 utils.py:1231] [82000] train/loss = 2.124313399195671
+I1203 20:42:22.289313 137274321021824 utils.py:1231] [82000] l2_grads = 2.067612409591675
+I1203 20:42:22.289385 137274321021824 utils.py:1231] [82000] lr = 0.00020392068099535452
+I1203 20:42:22.289438 137274321021824 utils.py:1231] [82000] uptime = 514331.65180053504
+I1203 20:42:22.289493 137274321021824 utils.py:1231] [82000] examples_seen = 83968000.0
+I1203 20:42:22.289543 137274321021824 utils.py:1231] [82000] progress = 0.7282221610436667
+I1203 20:42:22.289593 137274321021824 utils.py:1231] [82000] epoch = 65.54024572908918
+I1203 20:42:22.289647 137274321021824 utils.py:1231] [82000] img/sec/core = 164.2168522530038
+I1203 20:42:22.289705 137274321021824 utils.py:1231] [82000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.83558567556972
+I1203 20:42:22.289769 137274321021824 utils.py:1231] [82000] core_hours = 142.83558567556972
+I1203 20:42:22.289842 137274321021824 train.py:125] NOTE: Steps:82000/112603 [72.8%]
+Walltime:5d22h52m (0s eval)
+ETA:2d5h18m
+Total train time:8d4h8m
+I1203 20:47:34.545818 137274321021824 utils.py:1231] [82050] l2_params = 257.36312319358507
+I1203 20:47:34.546109 137274321021824 utils.py:1231] [82050] train/loss = 1.865881323814392
+I1203 20:47:34.546294 137274321021824 utils.py:1231] [82050] l2_grads = 2.169956922531128
+I1203 20:47:34.546387 137274321021824 utils.py:1231] [82050] lr = 0.00020330419375503563
+I1203 20:47:34.546459 137274321021824 utils.py:1231] [82050] uptime = 514643.90882027
+I1203 20:47:34.546523 137274321021824 utils.py:1231] [82050] examples_seen = 84019200.0
+I1203 20:47:34.546582 137274321021824 utils.py:1231] [82050] progress = 0.7286661989467421
+I1203 20:47:34.546642 137274321021824 utils.py:1231] [82050] epoch = 65.58020929355814
+I1203 20:47:34.546701 137274321021824 utils.py:1231] [82050] img/sec/core = 163.96749076597538
+I1203 20:47:34.546761 137274321021824 utils.py:1231] [82050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 142.9223237366072
+I1203 20:47:34.546817 137274321021824 utils.py:1231] [82050] core_hours = 142.9223237366072
+I1203 20:47:34.546894 137274321021824 train.py:125] NOTE: Steps:82050/112603 [72.9%]
+Walltime:5d22h57m (0s eval)
+ETA:2d5h13m
+Total train time:8d4h8m
+I1203 20:52:46.322412 137274321021824 utils.py:1231] [82100] l2_params = 257.2918620288097
+I1203 20:52:46.322623 137274321021824 utils.py:1231] [82100] train/loss = 2.593282014131546
+I1203 20:52:46.322731 137274321021824 utils.py:1231] [82100] l2_grads = 2.0795605182647705
+I1203 20:52:46.322803 137274321021824 utils.py:1231] [82100] lr = 0.00020268840190874727
+I1203 20:52:46.322864 137274321021824 utils.py:1231] [82100] uptime = 514955.685224939
+I1203 20:52:46.322930 137274321021824 utils.py:1231] [82100] examples_seen = 84070400.0
+I1203 20:52:46.322988 137274321021824 utils.py:1231] [82100] progress = 0.7291102368498175
+I1203 20:52:46.323045 137274321021824 utils.py:1231] [82100] epoch = 65.62017285802709
+I1203 20:52:46.323103 137274321021824 utils.py:1231] [82100] img/sec/core = 164.2202528262198
+I1203 20:52:46.323169 137274321021824 utils.py:1231] [82100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.00892829345972
+I1203 20:52:46.323227 137274321021824 utils.py:1231] [82100] core_hours = 143.00892829345972
+I1203 20:52:46.323297 137274321021824 train.py:125] NOTE: Steps:82100/112603 [72.9%]
+Walltime:5d23h2m (0s eval)
+ETA:2d5h8m
+Total train time:8d4h8m
+I1203 20:57:58.060091 137274321021824 utils.py:1231] [82150] l2_params = 257.2168855374765
+I1203 20:57:58.060302 137274321021824 utils.py:1231] [82150] train/loss = 3.5076777935028076
+I1203 20:57:58.060407 137274321021824 utils.py:1231] [82150] l2_grads = 1.9909127950668335
+I1203 20:57:58.060469 137274321021824 utils.py:1231] [82150] lr = 0.00020207330689977934
+I1203 20:57:58.060521 137274321021824 utils.py:1231] [82150] uptime = 515267.422883529
+I1203 20:57:58.060575 137274321021824 utils.py:1231] [82150] examples_seen = 84121600.0
+I1203 20:57:58.060625 137274321021824 utils.py:1231] [82150] progress = 0.7295542747528929
+I1203 20:57:58.060680 137274321021824 utils.py:1231] [82150] epoch = 65.66013642249605
+I1203 20:57:58.060731 137274321021824 utils.py:1231] [82150] img/sec/core = 164.24066386969605
+I1203 20:57:58.060787 137274321021824 utils.py:1231] [82150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.0955220875125
+I1203 20:57:58.060837 137274321021824 utils.py:1231] [82150] core_hours = 143.0955220875125
+I1203 20:57:58.060909 137274321021824 train.py:125] NOTE: Steps:82150/112603 [73.0%]
+Walltime:5d23h7m (0s eval)
+ETA:2d5h2m
+Total train time:8d4h8m
+I1203 21:03:09.858397 137274321021824 utils.py:1231] [82200] l2_params = 257.1453896587743
+I1203 21:03:09.858680 137274321021824 utils.py:1231] [82200] train/loss = 1.7288058251142502
+I1203 21:03:09.858844 137274321021824 utils.py:1231] [82200] l2_grads = 2.0633137226104736
+I1203 21:03:09.858928 137274321021824 utils.py:1231] [82200] lr = 0.00020145891016978826
+I1203 21:03:09.858992 137274321021824 utils.py:1231] [82200] uptime = 515579.22135316604
+I1203 21:03:09.859057 137274321021824 utils.py:1231] [82200] examples_seen = 84172800.0
+I1203 21:03:09.859117 137274321021824 utils.py:1231] [82200] progress = 0.7299983126559683
+I1203 21:03:09.859181 137274321021824 utils.py:1231] [82200] epoch = 65.70009998696501
+I1203 21:03:09.859243 137274321021824 utils.py:1231] [82200] img/sec/core = 164.20863149071485
+I1203 21:03:09.859311 137274321021824 utils.py:1231] [82200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.1821327735228
+I1203 21:03:09.859367 137274321021824 utils.py:1231] [82200] core_hours = 143.1821327735228
+I1203 21:03:09.859437 137274321021824 train.py:125] NOTE: Steps:82200/112603 [73.0%]
+Walltime:5d23h12m (0s eval)
+ETA:2d4h57m
+Total train time:8d4h8m
+I1203 21:08:21.650464 137274321021824 utils.py:1231] [82250] l2_params = 257.07308775362065
+I1203 21:08:21.650676 137274321021824 utils.py:1231] [82250] train/loss = 1.8240244537591934
+I1203 21:08:21.650771 137274321021824 utils.py:1231] [82250] l2_grads = 2.278759241104126
+I1203 21:08:21.650838 137274321021824 utils.py:1231] [82250] lr = 0.00020084521315879346
+I1203 21:08:21.650899 137274321021824 utils.py:1231] [82250] uptime = 515891.01325972297
+I1203 21:08:21.650962 137274321021824 utils.py:1231] [82250] examples_seen = 84224000.0
+I1203 21:08:21.651015 137274321021824 utils.py:1231] [82250] progress = 0.7304423505590437
+I1203 21:08:21.651067 137274321021824 utils.py:1231] [82250] epoch = 65.74006355143396
+I1203 21:08:21.651127 137274321021824 utils.py:1231] [82250] img/sec/core = 164.21208800893456
+I1203 21:08:21.651185 137274321021824 utils.py:1231] [82250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.26874163645525
+I1203 21:08:21.651241 137274321021824 utils.py:1231] [82250] core_hours = 143.26874163645525
+I1203 21:08:21.651304 137274321021824 train.py:125] NOTE: Steps:82250/112603 [73.0%]
+Walltime:5d23h18m (0s eval)
+ETA:2d4h52m
+Total train time:8d4h8m
+I1203 21:13:33.460494 137274321021824 utils.py:1231] [82300] l2_params = 257.0029022163324
+I1203 21:13:33.460704 137274321021824 utils.py:1231] [82300] train/loss = 1.7877484112977982
+I1203 21:13:33.460799 137274321021824 utils.py:1231] [82300] l2_grads = 2.204293727874756
+I1203 21:13:33.460870 137274321021824 utils.py:1231] [82300] lr = 0.00020023221730517497
+I1203 21:13:33.460933 137274321021824 utils.py:1231] [82300] uptime = 516202.823294627
+I1203 21:13:33.460994 137274321021824 utils.py:1231] [82300] examples_seen = 84275200.0
+I1203 21:13:33.461055 137274321021824 utils.py:1231] [82300] progress = 0.7308863884621192
+I1203 21:13:33.461109 137274321021824 utils.py:1231] [82300] epoch = 65.78002711590293
+I1203 21:13:33.461167 137274321021824 utils.py:1231] [82300] img/sec/core = 164.20254086997468
+I1203 21:13:33.461229 137274321021824 utils.py:1231] [82300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.3553555350397
+I1203 21:13:33.461293 137274321021824 utils.py:1231] [82300] core_hours = 143.3553555350397
+I1203 21:13:33.461358 137274321021824 train.py:125] NOTE: Steps:82300/112603 [73.1%]
+Walltime:5d23h23m (0s eval)
+ETA:2d4h47m
+Total train time:8d4h8m
+I1203 21:18:45.259441 137274321021824 utils.py:1231] [82350] l2_params = 256.9280819225773
+I1203 21:18:45.259651 137274321021824 utils.py:1231] [82350] train/loss = 1.951987624168396
+I1203 21:18:45.259752 137274321021824 utils.py:1231] [82350] l2_grads = 2.0057995319366455
+I1203 21:18:45.259825 137274321021824 utils.py:1231] [82350] lr = 0.00019961992404566892
+I1203 21:18:45.259889 137274321021824 utils.py:1231] [82350] uptime = 516514.62224550894
+I1203 21:18:45.259953 137274321021824 utils.py:1231] [82350] examples_seen = 84326400.0
+I1203 21:18:45.260009 137274321021824 utils.py:1231] [82350] progress = 0.7313304263651945
+I1203 21:18:45.260063 137274321021824 utils.py:1231] [82350] epoch = 65.81999068037187
+I1203 21:18:45.260121 137274321021824 utils.py:1231] [82350] img/sec/core = 164.20837804354278
+I1203 21:18:45.260181 137274321021824 utils.py:1231] [82350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.44196635472915
+I1203 21:18:45.260248 137274321021824 utils.py:1231] [82350] core_hours = 143.44196635472915
+I1203 21:18:45.260318 137274321021824 train.py:125] NOTE: Steps:82350/112603 [73.1%]
+Walltime:5d23h28m (0s eval)
+ETA:2d4h41m
+Total train time:8d4h8m
+I1203 21:23:57.058979 137274321021824 utils.py:1231] [82400] l2_params = 256.8571177046422
+I1203 21:23:57.059200 137274321021824 utils.py:1231] [82400] train/loss = 1.8827508240938187
+I1203 21:23:57.059321 137274321021824 utils.py:1231] [82400] l2_grads = 2.3060567378997803
+I1203 21:23:57.059401 137274321021824 utils.py:1231] [82400] lr = 0.0001990083348153652
+I1203 21:23:57.059468 137274321021824 utils.py:1231] [82400] uptime = 516826.421828664
+I1203 21:23:57.059537 137274321021824 utils.py:1231] [82400] examples_seen = 84377600.0
+I1203 21:23:57.059602 137274321021824 utils.py:1231] [82400] progress = 0.73177446426827
+I1203 21:23:57.059664 137274321021824 utils.py:1231] [82400] epoch = 65.85995424484084
+I1203 21:23:57.059726 137274321021824 utils.py:1231] [82400] img/sec/core = 164.20804505865763
+I1203 21:23:57.059795 137274321021824 utils.py:1231] [82400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.52857735005
+I1203 21:23:57.059866 137274321021824 utils.py:1231] [82400] core_hours = 143.52857735005
+I1203 21:23:57.059949 137274321021824 train.py:125] NOTE: Steps:82400/112603 [73.2%]
+Walltime:5d23h33m (0s eval)
+ETA:2d4h36m
+Total train time:8d4h8m
+I1203 21:29:08.852011 137274321021824 utils.py:1231] [82450] l2_params = 256.78097594354364
+I1203 21:29:08.852212 137274321021824 utils.py:1231] [82450] train/loss = 3.7068368792533875
+I1203 21:29:08.852303 137274321021824 utils.py:1231] [82450] l2_grads = 2.037637948989868
+I1203 21:29:08.852364 137274321021824 utils.py:1231] [82450] lr = 0.00019839745104770332
+I1203 21:29:08.852417 137274321021824 utils.py:1231] [82450] uptime = 517138.21477854997
+I1203 21:29:08.852472 137274321021824 utils.py:1231] [82450] examples_seen = 84428800.0
+I1203 21:29:08.852522 137274321021824 utils.py:1231] [82450] progress = 0.7322185021713453
+I1203 21:29:08.852572 137274321021824 utils.py:1231] [82450] epoch = 65.8999178093098
+I1203 21:29:08.852626 137274321021824 utils.py:1231] [82450] img/sec/core = 164.21153851852384
+I1203 21:29:08.852684 137274321021824 utils.py:1231] [82450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.6151865027961
+I1203 21:29:08.852736 137274321021824 utils.py:1231] [82450] core_hours = 143.6151865027961
+I1203 21:29:08.852798 137274321021824 train.py:125] NOTE: Steps:82450/112603 [73.2%]
+Walltime:5d23h38m (0s eval)
+ETA:2d4h31m
+Total train time:8d4h8m
+I1203 21:34:20.646365 137274321021824 utils.py:1231] [82500] l2_params = 256.7121353733276
+I1203 21:34:20.646638 137274321021824 utils.py:1231] [82500] train/loss = 2.3221717178821564
+I1203 21:34:20.646749 137274321021824 utils.py:1231] [82500] l2_grads = 2.1738007068634033
+I1203 21:34:20.646829 137274321021824 utils.py:1231] [82500] lr = 0.00019778727417446907
+I1203 21:34:20.646897 137274321021824 utils.py:1231] [82500] uptime = 517450.009258062
+I1203 21:34:20.646953 137274321021824 utils.py:1231] [82500] examples_seen = 84480000.0
+I1203 21:34:20.647006 137274321021824 utils.py:1231] [82500] progress = 0.7326625400744208
+I1203 21:34:20.647055 137274321021824 utils.py:1231] [82500] epoch = 65.93988137377875
+I1203 21:34:20.647107 137274321021824 utils.py:1231] [82500] img/sec/core = 164.21073291653633
+I1203 21:34:20.647164 137274321021824 utils.py:1231] [82500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.70179608043833
+I1203 21:34:20.647215 137274321021824 utils.py:1231] [82500] core_hours = 143.70179608043833
+I1203 21:34:20.647284 137274321021824 train.py:125] NOTE: Steps:82500/112603 [73.3%]
+Walltime:5d23h44m (0s eval)
+ETA:2d4h26m
+Total train time:8d4h8m
+I1203 21:34:20.647380 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:82500/112603 [73.3%]
+Walltime:5d23h44m (0s eval)
+ETA:2d4h26m
+Total train time:8d4h8m
+I1203 21:35:58.661605 137274321021824 utils.py:1231] [82500] val/acc@1 = 0.7265824298469388
+I1203 21:35:58.661812 137274321021824 utils.py:1231] [82500] val/loss = 1.0882979637506056
+I1203 21:35:58.661966 137274321021824 utils.py:1231] [82500] z/secs/eval/val = 98.01451567502227
+I1203 21:35:58.662040 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.01451567502227
+I1203 21:41:10.449568 137274321021824 utils.py:1231] [82550] l2_params = 256.63408016865014
+I1203 21:41:10.449902 137274321021824 utils.py:1231] [82550] train/loss = 3.4414034485816956
+I1203 21:41:10.450121 137274321021824 utils.py:1231] [82550] l2_grads = 2.1129209995269775
+I1203 21:41:10.450230 137274321021824 utils.py:1231] [82550] lr = 0.00019717780562579216
+I1203 21:41:10.450330 137274321021824 utils.py:1231] [82550] uptime = 517859.81268795393
+I1203 21:41:10.450417 137274321021824 utils.py:1231] [82550] examples_seen = 84531200.0
+I1203 21:41:10.450495 137274321021824 utils.py:1231] [82550] progress = 0.7331065779774961
+I1203 21:41:10.450575 137274321021824 utils.py:1231] [82550] epoch = 65.97984493824771
+I1203 21:41:10.450660 137274321021824 utils.py:1231] [82550] img/sec/core = 124.93794894176732
+I1203 21:41:10.450755 137274321021824 utils.py:1231] [82550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.81563036651943
+I1203 21:41:10.450834 137274321021824 utils.py:1231] [82550] core_hours = 143.81563036651943
+I1203 21:41:10.450923 137274321021824 train.py:125] NOTE: Steps:82550/112603 [73.3%]
+Walltime:5d23h50m (0s eval)
+ETA:2d4h21m
+Total train time:8d4h10m
+I1203 21:46:22.225913 137274321021824 utils.py:1231] [82600] l2_params = 256.56661110370896
+I1203 21:46:22.226127 137274321021824 utils.py:1231] [82600] train/loss = 4.150286912918091
+I1203 21:46:22.226221 137274321021824 utils.py:1231] [82600] l2_grads = 2.160444736480713
+I1203 21:46:22.226281 137274321021824 utils.py:1231] [82600] lr = 0.00019656904683014117
+I1203 21:46:22.226356 137274321021824 utils.py:1231] [82600] uptime = 518171.58871346596
+I1203 21:46:22.226428 137274321021824 utils.py:1231] [82600] examples_seen = 84582400.0
+I1203 21:46:22.226480 137274321021824 utils.py:1231] [82600] progress = 0.7335506158805716
+I1203 21:46:22.226530 137274321021824 utils.py:1231] [82600] epoch = 66.01980850271666
+I1203 21:46:22.226584 137274321021824 utils.py:1231] [82600] img/sec/core = 164.2204525377324
+I1203 21:46:22.226647 137274321021824 utils.py:1231] [82600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.90223481805054
+I1203 21:46:22.226696 137274321021824 utils.py:1231] [82600] core_hours = 143.90223481805054
+I1203 21:46:22.226756 137274321021824 train.py:125] NOTE: Steps:82600/112603 [73.4%]
+Walltime:5d23h56m (0s eval)
+ETA:2d4h16m
+Total train time:8d4h10m
+I1203 21:51:34.000392 137274321021824 utils.py:1231] [82650] l2_params = 256.4975790871365
+I1203 21:51:34.000626 137274321021824 utils.py:1231] [82650] train/loss = 1.8509526997804642
+I1203 21:51:34.000753 137274321021824 utils.py:1231] [82650] l2_grads = 2.1907246112823486
+I1203 21:51:34.000834 137274321021824 utils.py:1231] [82650] lr = 0.00019596099921432227
+I1203 21:51:34.000916 137274321021824 utils.py:1231] [82650] uptime = 518483.36327707605
+I1203 21:51:34.001026 137274321021824 utils.py:1231] [82650] examples_seen = 84633600.0
+I1203 21:51:34.001097 137274321021824 utils.py:1231] [82650] progress = 0.7339946537836469
+I1203 21:51:34.001171 137274321021824 utils.py:1231] [82650] epoch = 66.05977206718562
+I1203 21:51:34.001227 137274321021824 utils.py:1231] [82650] img/sec/core = 164.22122256269697
+I1203 21:51:34.001295 137274321021824 utils.py:1231] [82650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 143.9888388634978
+I1203 21:51:34.001349 137274321021824 utils.py:1231] [82650] core_hours = 143.9888388634978
+I1203 21:51:34.001432 137274321021824 train.py:125] NOTE: Steps:82650/112603 [73.4%]
+Walltime:6d0h1m (0s eval)
+ETA:2d4h11m
+Total train time:8d4h10m
+I1203 21:56:45.787768 137274321021824 utils.py:1231] [82700] l2_params = 256.42176776412106
+I1203 21:56:45.787992 137274321021824 utils.py:1231] [82700] train/loss = 1.8907054364681244
+I1203 21:56:45.788099 137274321021824 utils.py:1231] [82700] l2_grads = 2.2481675148010254
+I1203 21:56:45.788199 137274321021824 utils.py:1231] [82700] lr = 0.00019535366420347355
+I1203 21:56:45.788275 137274321021824 utils.py:1231] [82700] uptime = 518795.15063685493
+I1203 21:56:45.788334 137274321021824 utils.py:1231] [82700] examples_seen = 84684800.0
+I1203 21:56:45.788393 137274321021824 utils.py:1231] [82700] progress = 0.7344386916867224
+I1203 21:56:45.788451 137274321021824 utils.py:1231] [82700] epoch = 66.09973563165458
+I1203 21:56:45.788514 137274321021824 utils.py:1231] [82700] img/sec/core = 164.21448270484782
+I1203 21:56:45.788575 137274321021824 utils.py:1231] [82700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.07544646343638
+I1203 21:56:45.788629 137274321021824 utils.py:1231] [82700] core_hours = 144.07544646343638
+I1203 21:56:45.788689 137274321021824 train.py:125] NOTE: Steps:82700/112603 [73.4%]
+Walltime:6d0h6m (0s eval)
+ETA:2d4h5m
+Total train time:8d4h10m
+I1203 22:01:57.573521 137274321021824 utils.py:1231] [82750] l2_params = 256.35198832672336
+I1203 22:01:57.573722 137274321021824 utils.py:1231] [82750] train/loss = 2.967708468437195
+I1203 22:01:57.573817 137274321021824 utils.py:1231] [82750] l2_grads = 1.9581122398376465
+I1203 22:01:57.573878 137274321021824 utils.py:1231] [82750] lr = 0.00019474704322106393
+I1203 22:01:57.573937 137274321021824 utils.py:1231] [82750] uptime = 519106.936298872
+I1203 22:01:57.573990 137274321021824 utils.py:1231] [82750] examples_seen = 84736000.0
+I1203 22:01:57.574041 137274321021824 utils.py:1231] [82750] progress = 0.7348827295897978
+I1203 22:01:57.574091 137274321021824 utils.py:1231] [82750] epoch = 66.13969919612353
+I1203 22:01:57.574143 137274321021824 utils.py:1231] [82750] img/sec/core = 164.21537689952464
+I1203 22:01:57.574201 137274321021824 utils.py:1231] [82750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.16205359177445
+I1203 22:01:57.574252 137274321021824 utils.py:1231] [82750] core_hours = 144.16205359177445
+I1203 22:01:57.574314 137274321021824 train.py:125] NOTE: Steps:82750/112603 [73.5%]
+Walltime:6d0h11m (0s eval)
+ETA:2d4h0m
+Total train time:8d4h10m
+I1203 22:07:09.365310 137274321021824 utils.py:1231] [82800] l2_params = 256.28490812239653
+I1203 22:07:09.365533 137274321021824 utils.py:1231] [82800] train/loss = 2.3793870508670807
+I1203 22:07:09.365638 137274321021824 utils.py:1231] [82800] l2_grads = 2.074446678161621
+I1203 22:07:09.365715 137274321021824 utils.py:1231] [82800] lr = 0.00019414113768888858
+I1203 22:07:09.365777 137274321021824 utils.py:1231] [82800] uptime = 519418.728138576
+I1203 22:07:09.365838 137274321021824 utils.py:1231] [82800] examples_seen = 84787200.0
+I1203 22:07:09.365918 137274321021824 utils.py:1231] [82800] progress = 0.7353267674928732
+I1203 22:07:09.365978 137274321021824 utils.py:1231] [82800] epoch = 66.17966276059249
+I1203 22:07:09.366039 137274321021824 utils.py:1231] [82800] img/sec/core = 164.2121232184948
+I1203 22:07:09.366098 137274321021824 utils.py:1231] [82800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.24866243613667
+I1203 22:07:09.366153 137274321021824 utils.py:1231] [82800] core_hours = 144.24866243613667
+I1203 22:07:09.366223 137274321021824 train.py:125] NOTE: Steps:82800/112603 [73.5%]
+Walltime:6d0h16m (0s eval)
+ETA:2d3h55m
+Total train time:8d4h10m
+I1203 22:12:21.089271 137274321021824 utils.py:1231] [82850] l2_params = 256.20950448946513
+I1203 22:12:21.089490 137274321021824 utils.py:1231] [82850] train/loss = 1.6926035732030869
+I1203 22:12:21.089596 137274321021824 utils.py:1231] [82850] l2_grads = 2.113391876220703
+I1203 22:12:21.089656 137274321021824 utils.py:1231] [82850] lr = 0.00019353594902706518
+I1203 22:12:21.089714 137274321021824 utils.py:1231] [82850] uptime = 519730.45207097405
+I1203 22:12:21.089776 137274321021824 utils.py:1231] [82850] examples_seen = 84838400.0
+I1203 22:12:21.089829 137274321021824 utils.py:1231] [82850] progress = 0.7357708053959486
+I1203 22:12:21.089889 137274321021824 utils.py:1231] [82850] epoch = 66.21962632506145
+I1203 22:12:21.089941 137274321021824 utils.py:1231] [82850] img/sec/core = 164.24789590626125
+I1203 22:12:21.089998 137274321021824 utils.py:1231] [82850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.33525241735833
+I1203 22:12:21.090048 137274321021824 utils.py:1231] [82850] core_hours = 144.33525241735833
+I1203 22:12:21.090108 137274321021824 train.py:125] NOTE: Steps:82850/112603 [73.6%]
+Walltime:6d0h22m (0s eval)
+ETA:2d3h50m
+Total train time:8d4h10m
+I1203 22:17:32.874959 137274321021824 utils.py:1231] [82900] l2_params = 256.13549849541863
+I1203 22:17:32.875161 137274321021824 utils.py:1231] [82900] train/loss = 3.9539983570575714
+I1203 22:17:32.875259 137274321021824 utils.py:1231] [82900] l2_grads = 2.211660385131836
+I1203 22:17:32.875327 137274321021824 utils.py:1231] [82900] lr = 0.00019293147865403202
+I1203 22:17:32.875401 137274321021824 utils.py:1231] [82900] uptime = 520042.23774769995
+I1203 22:17:32.875463 137274321021824 utils.py:1231] [82900] examples_seen = 84889600.0
+I1203 22:17:32.875519 137274321021824 utils.py:1231] [82900] progress = 0.736214843299024
+I1203 22:17:32.875576 137274321021824 utils.py:1231] [82900] epoch = 66.2595898895304
+I1203 22:17:32.875636 137274321021824 utils.py:1231] [82900] img/sec/core = 164.2153691524779
+I1203 22:17:32.875696 137274321021824 utils.py:1231] [82900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.42185954978223
+I1203 22:17:32.875756 137274321021824 utils.py:1231] [82900] core_hours = 144.42185954978223
+I1203 22:17:32.875817 137274321021824 train.py:125] NOTE: Steps:82900/112603 [73.6%]
+Walltime:6d0h27m (0s eval)
+ETA:2d3h44m
+Total train time:8d4h10m
+I1203 22:22:44.672640 137274321021824 utils.py:1231] [82950] l2_params = 256.0690015585937
+I1203 22:22:44.672914 137274321021824 utils.py:1231] [82950] train/loss = 3.8328438103199005
+I1203 22:22:44.673112 137274321021824 utils.py:1231] [82950] l2_grads = 2.0549392700195312
+I1203 22:22:44.673232 137274321021824 utils.py:1231] [82950] lr = 0.00019232772798654312
+I1203 22:22:44.673330 137274321021824 utils.py:1231] [82950] uptime = 520354.035678469
+I1203 22:22:44.673406 137274321021824 utils.py:1231] [82950] examples_seen = 84940800.0
+I1203 22:22:44.673466 137274321021824 utils.py:1231] [82950] progress = 0.7366588812020994
+I1203 22:22:44.673550 137274321021824 utils.py:1231] [82950] epoch = 66.29955345399937
+I1203 22:22:44.673620 137274321021824 utils.py:1231] [82950] img/sec/core = 164.208915285989
+I1203 22:22:44.673707 137274321021824 utils.py:1231] [82950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.50847008610694
+I1203 22:22:44.673780 137274321021824 utils.py:1231] [82950] core_hours = 144.50847008610694
+I1203 22:22:44.673872 137274321021824 train.py:125] NOTE: Steps:82950/112603 [73.7%]
+Walltime:6d0h32m (0s eval)
+ETA:2d3h39m
+Total train time:8d4h10m
+I1203 22:27:56.403483 137274321021824 utils.py:1231] [83000] l2_params = 256.0049159167751
+I1203 22:27:56.403771 137274321021824 utils.py:1231] [83000] train/loss = 2.622436046600342
+I1203 22:27:56.403934 137274321021824 utils.py:1231] [83000] l2_grads = 1.9680837392807007
+I1203 22:27:56.404010 137274321021824 utils.py:1231] [83000] lr = 0.00019172469843966626
+I1203 22:27:56.404084 137274321021824 utils.py:1231] [83000] uptime = 520665.76643635496
+I1203 22:27:56.404144 137274321021824 utils.py:1231] [83000] examples_seen = 84992000.0
+I1203 22:27:56.404208 137274321021824 utils.py:1231] [83000] progress = 0.7371029191051748
+I1203 22:27:56.404267 137274321021824 utils.py:1231] [83000] epoch = 66.33951701846831
+I1203 22:27:56.404340 137274321021824 utils.py:1231] [83000] img/sec/core = 164.2442996232349
+I1203 22:27:56.404410 137274321021824 utils.py:1231] [83000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.5950619632975
+I1203 22:27:56.404469 137274321021824 utils.py:1231] [83000] core_hours = 144.5950619632975
+I1203 22:27:56.404536 137274321021824 train.py:125] NOTE: Steps:83000/112603 [73.7%]
+Walltime:6d0h37m (0s eval)
+ETA:2d3h34m
+Total train time:8d4h10m
+I1203 22:33:08.401011 137274321021824 utils.py:1231] [83050] l2_params = 255.9323902088146
+I1203 22:33:08.401237 137274321021824 utils.py:1231] [83050] train/loss = 2.4674819707870483
+I1203 22:33:08.401350 137274321021824 utils.py:1231] [83050] l2_grads = 2.1198151111602783
+I1203 22:33:08.401423 137274321021824 utils.py:1231] [83050] lr = 0.00019112239142677901
+I1203 22:33:08.401480 137274321021824 utils.py:1231] [83050] uptime = 520977.76384211995
+I1203 22:33:08.401531 137274321021824 utils.py:1231] [83050] examples_seen = 85043200.0
+I1203 22:33:08.401580 137274321021824 utils.py:1231] [83050] progress = 0.7375469570082502
+I1203 22:33:08.401627 137274321021824 utils.py:1231] [83050] epoch = 66.37948058293728
+I1203 22:33:08.401678 137274321021824 utils.py:1231] [83050] img/sec/core = 164.10392860306368
+I1203 22:33:08.401734 137274321021824 utils.py:1231] [83050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.68172790934332
+I1203 22:33:08.401792 137274321021824 utils.py:1231] [83050] core_hours = 144.68172790934332
+I1203 22:33:08.401852 137274321021824 train.py:125] NOTE: Steps:83050/112603 [73.8%]
+Walltime:6d0h42m (0s eval)
+ETA:2d3h29m
+Total train time:8d4h10m
+I1203 22:38:20.194213 137274321021824 utils.py:1231] [83100] l2_params = 255.8664267566145
+I1203 22:38:20.194442 137274321021824 utils.py:1231] [83100] train/loss = 1.6842930316925049
+I1203 22:38:20.194569 137274321021824 utils.py:1231] [83100] l2_grads = 2.2321274280548096
+I1203 22:38:20.194647 137274321021824 utils.py:1231] [83100] lr = 0.00019052080835956506
+I1203 22:38:20.194710 137274321021824 utils.py:1231] [83100] uptime = 521289.557071512
+I1203 22:38:20.194779 137274321021824 utils.py:1231] [83100] examples_seen = 85094400.0
+I1203 22:38:20.194838 137274321021824 utils.py:1231] [83100] progress = 0.7379909949113256
+I1203 22:38:20.194922 137274321021824 utils.py:1231] [83100] epoch = 66.41944414740624
+I1203 22:38:20.194978 137274321021824 utils.py:1231] [83100] img/sec/core = 164.21139131159262
+I1203 22:38:20.195040 137274321021824 utils.py:1231] [83100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.76833713973
+I1203 22:38:20.195096 137274321021824 utils.py:1231] [83100] core_hours = 144.76833713973
+I1203 22:38:20.195162 137274321021824 train.py:125] NOTE: Steps:83100/112603 [73.8%]
+Walltime:6d0h48m (0s eval)
+ETA:2d3h23m
+Total train time:8d4h10m
+I1203 22:43:31.987051 137274321021824 utils.py:1231] [83150] l2_params = 255.7919365396431
+I1203 22:43:31.987281 137274321021824 utils.py:1231] [83150] train/loss = 3.3983557522296906
+I1203 22:43:31.987375 137274321021824 utils.py:1231] [83150] l2_grads = 2.0013954639434814
+I1203 22:43:31.987437 137274321021824 utils.py:1231] [83150] lr = 0.00018991995064801176
+I1203 22:43:31.987488 137274321021824 utils.py:1231] [83150] uptime = 521601.349849954
+I1203 22:43:31.987550 137274321021824 utils.py:1231] [83150] examples_seen = 85145600.0
+I1203 22:43:31.987600 137274321021824 utils.py:1231] [83150] progress = 0.738435032814401
+I1203 22:43:31.987647 137274321021824 utils.py:1231] [83150] epoch = 66.45940771187519
+I1203 22:43:31.987696 137274321021824 utils.py:1231] [83150] img/sec/core = 164.21162881268936
+I1203 22:43:31.987751 137274321021824 utils.py:1231] [83150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.85494624485278
+I1203 22:43:31.987799 137274321021824 utils.py:1231] [83150] core_hours = 144.85494624485278
+I1203 22:43:31.987859 137274321021824 train.py:125] NOTE: Steps:83150/112603 [73.8%]
+Walltime:6d0h53m (0s eval)
+ETA:2d3h18m
+Total train time:8d4h10m
+I1203 22:48:43.781398 137274321021824 utils.py:1231] [83200] l2_params = 255.71840931171488
+I1203 22:48:43.781617 137274321021824 utils.py:1231] [83200] train/loss = 2.878326565027237
+I1203 22:48:43.781724 137274321021824 utils.py:1231] [83200] l2_grads = 1.973752737045288
+I1203 22:48:43.781798 137274321021824 utils.py:1231] [83200] lr = 0.00018931981970040583
+I1203 22:48:43.781865 137274321021824 utils.py:1231] [83200] uptime = 521913.144222631
+I1203 22:48:43.781934 137274321021824 utils.py:1231] [83200] examples_seen = 85196800.0
+I1203 22:48:43.781992 137274321021824 utils.py:1231] [83200] progress = 0.7388790707174765
+I1203 22:48:43.782051 137274321021824 utils.py:1231] [83200] epoch = 66.49937127634415
+I1203 22:48:43.782109 137274321021824 utils.py:1231] [83200] img/sec/core = 164.210789182658
+I1203 22:48:43.782182 137274321021824 utils.py:1231] [83200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 144.9415557928186
+I1203 22:48:43.782238 137274321021824 utils.py:1231] [83200] core_hours = 144.9415557928186
+I1203 22:48:43.782306 137274321021824 train.py:125] NOTE: Steps:83200/112603 [73.9%]
+Walltime:6d0h58m (0s eval)
+ETA:2d3h13m
+Total train time:8d4h10m
+I1203 22:53:55.669668 137274321021824 utils.py:1231] [83250] l2_params = 255.64877129622633
+I1203 22:53:55.669917 137274321021824 utils.py:1231] [83250] train/loss = 4.041491270065308
+I1203 22:53:55.670081 137274321021824 utils.py:1231] [83250] l2_grads = 2.203723192214966
+I1203 22:53:55.670220 137274321021824 utils.py:1231] [83250] lr = 0.00018872041692333095
+I1203 22:53:55.670311 137274321021824 utils.py:1231] [83250] uptime = 522225.032667413
+I1203 22:53:55.670408 137274321021824 utils.py:1231] [83250] examples_seen = 85248000.0
+I1203 22:53:55.670509 137274321021824 utils.py:1231] [83250] progress = 0.7393231086205518
+I1203 22:53:55.670591 137274321021824 utils.py:1231] [83250] epoch = 66.5393348408131
+I1203 22:53:55.670674 137274321021824 utils.py:1231] [83250] img/sec/core = 164.161259759998
+I1203 22:53:55.670761 137274321021824 utils.py:1231] [83250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.02819147192474
+I1203 22:53:55.670832 137274321021824 utils.py:1231] [83250] core_hours = 145.02819147192474
+I1203 22:53:55.670932 137274321021824 train.py:125] NOTE: Steps:83250/112603 [73.9%]
+Walltime:6d1h3m (0s eval)
+ETA:2d3h8m
+Total train time:8d4h10m
+I1203 22:59:07.454135 137274321021824 utils.py:1231] [83300] l2_params = 255.58282930541384
+I1203 22:59:07.454365 137274321021824 utils.py:1231] [83300] train/loss = 1.6908614486455917
+I1203 22:59:07.454486 137274321021824 utils.py:1231] [83300] l2_grads = 2.1693296432495117
+I1203 22:59:07.454576 137274321021824 utils.py:1231] [83300] lr = 0.00018812174372166463
+I1203 22:59:07.454653 137274321021824 utils.py:1231] [83300] uptime = 522536.81701469293
+I1203 22:59:07.454718 137274321021824 utils.py:1231] [83300] examples_seen = 85299200.0
+I1203 22:59:07.454780 137274321021824 utils.py:1231] [83300] progress = 0.7397671465236273
+I1203 22:59:07.454832 137274321021824 utils.py:1231] [83300] epoch = 66.57929840528206
+I1203 22:59:07.454903 137274321021824 utils.py:1231] [83300] img/sec/core = 164.21606936550674
+I1203 22:59:07.454962 137274321021824 utils.py:1231] [83300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.11479823505803
+I1203 22:59:07.455013 137274321021824 utils.py:1231] [83300] core_hours = 145.11479823505803
+I1203 22:59:07.455073 137274321021824 train.py:125] NOTE: Steps:83300/112603 [74.0%]
+Walltime:6d1h8m (0s eval)
+ETA:2d3h2m
+Total train time:8d4h10m
+I1203 23:04:19.241149 137274321021824 utils.py:1231] [83350] l2_params = 255.51009429113392
+I1203 23:04:19.241351 137274321021824 utils.py:1231] [83350] train/loss = 3.9748773276805878
+I1203 23:04:19.241451 137274321021824 utils.py:1231] [83350] l2_grads = 2.083564043045044
+I1203 23:04:19.241514 137274321021824 utils.py:1231] [83350] lr = 0.00018752380149857342
+I1203 23:04:19.241568 137274321021824 utils.py:1231] [83350] uptime = 522848.60392912396
+I1203 23:04:19.241622 137274321021824 utils.py:1231] [83350] examples_seen = 85350400.0
+I1203 23:04:19.241673 137274321021824 utils.py:1231] [83350] progress = 0.7402111844267026
+I1203 23:04:19.241726 137274321021824 utils.py:1231] [83350] epoch = 66.61926196975102
+I1203 23:04:19.241778 137274321021824 utils.py:1231] [83350] img/sec/core = 164.2147172643027
+I1203 23:04:19.241837 137274321021824 utils.py:1231] [83350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.20140571128886
+I1203 23:04:19.241893 137274321021824 utils.py:1231] [83350] core_hours = 145.20140571128886
+I1203 23:04:19.241965 137274321021824 train.py:125] NOTE: Steps:83350/112603 [74.0%]
+Walltime:6d1h14m (0s eval)
+ETA:2d2h57m
+Total train time:8d4h10m
+I1203 23:09:31.022361 137274321021824 utils.py:1231] [83400] l2_params = 255.43843873756632
+I1203 23:09:31.022573 137274321021824 utils.py:1231] [83400] train/loss = 1.7373394668102264
+I1203 23:09:31.022668 137274321021824 utils.py:1231] [83400] l2_grads = 2.288088321685791
+I1203 23:09:31.022729 137274321021824 utils.py:1231] [83400] lr = 0.0001869265916555113
+I1203 23:09:31.022797 137274321021824 utils.py:1231] [83400] uptime = 523160.385145282
+I1203 23:09:31.022856 137274321021824 utils.py:1231] [83400] examples_seen = 85401600.0
+I1203 23:09:31.022912 137274321021824 utils.py:1231] [83400] progress = 0.7406552223297781
+I1203 23:09:31.022962 137274321021824 utils.py:1231] [83400] epoch = 66.65922553421997
+I1203 23:09:31.023015 137274321021824 utils.py:1231] [83400] img/sec/core = 164.2177185364585
+I1203 23:09:31.023072 137274321021824 utils.py:1231] [83400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.28801160466614
+I1203 23:09:31.023123 137274321021824 utils.py:1231] [83400] core_hours = 145.28801160466614
+I1203 23:09:31.023185 137274321021824 train.py:125] NOTE: Steps:83400/112603 [74.1%]
+Walltime:6d1h19m (0s eval)
+ETA:2d2h52m
+Total train time:8d4h9m
+I1203 23:14:42.807399 137274321021824 utils.py:1231] [83450] l2_params = 255.36988043858275
+I1203 23:14:42.807629 137274321021824 utils.py:1231] [83450] train/loss = 2.8390143513679504
+I1203 23:14:42.807750 137274321021824 utils.py:1231] [83450] l2_grads = 2.073019027709961
+I1203 23:14:42.807846 137274321021824 utils.py:1231] [83450] lr = 0.00018633011559221536
+I1203 23:14:42.807917 137274321021824 utils.py:1231] [83450] uptime = 523472.17027833697
+I1203 23:14:42.807977 137274321021824 utils.py:1231] [83450] examples_seen = 85452800.0
+I1203 23:14:42.808032 137274321021824 utils.py:1231] [83450] progress = 0.7410992602328534
+I1203 23:14:42.808087 137274321021824 utils.py:1231] [83450] epoch = 66.69918909868893
+I1203 23:14:42.808142 137274321021824 utils.py:1231] [83450] img/sec/core = 164.2156555007293
+I1203 23:14:42.808201 137274321021824 utils.py:1231] [83450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.37461858607028
+I1203 23:14:42.808255 137274321021824 utils.py:1231] [83450] core_hours = 145.37461858607028
+I1203 23:14:42.808318 137274321021824 train.py:125] NOTE: Steps:83450/112603 [74.1%]
+Walltime:6d1h24m (0s eval)
+ETA:2d2h47m
+Total train time:8d4h9m
+I1203 23:19:54.599277 137274321021824 utils.py:1231] [83500] l2_params = 255.2995106996277
+I1203 23:19:54.599539 137274321021824 utils.py:1231] [83500] train/loss = 3.2166771292686462
+I1203 23:19:54.599673 137274321021824 utils.py:1231] [83500] l2_grads = 2.1026759147644043
+I1203 23:19:54.599759 137274321021824 utils.py:1231] [83500] lr = 0.000185734374706703
+I1203 23:19:54.599836 137274321021824 utils.py:1231] [83500] uptime = 523783.96219762194
+I1203 23:19:54.599914 137274321021824 utils.py:1231] [83500] examples_seen = 85504000.0
+I1203 23:19:54.599987 137274321021824 utils.py:1231] [83500] progress = 0.7415432981359289
+I1203 23:19:54.600049 137274321021824 utils.py:1231] [83500] epoch = 66.73915266315788
+I1203 23:19:54.600131 137274321021824 utils.py:1231] [83500] img/sec/core = 164.2120813054336
+I1203 23:19:54.600197 137274321021824 utils.py:1231] [83500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.4612274525383
+I1203 23:19:54.600259 137274321021824 utils.py:1231] [83500] core_hours = 145.4612274525383
+I1203 23:19:54.600355 137274321021824 train.py:125] NOTE: Steps:83500/112603 [74.2%]
+Walltime:6d1h29m (0s eval)
+ETA:2d2h42m
+Total train time:8d4h9m
+I1203 23:25:06.379102 137274321021824 utils.py:1231] [83550] l2_params = 255.22955869667356
+I1203 23:25:06.379310 137274321021824 utils.py:1231] [83550] train/loss = 2.5931781232357025
+I1203 23:25:06.379422 137274321021824 utils.py:1231] [83550] l2_grads = 1.9096113443374634
+I1203 23:25:06.379493 137274321021824 utils.py:1231] [83550] lr = 0.00018513937039526878
+I1203 23:25:06.379552 137274321021824 utils.py:1231] [83550] uptime = 524095.741913768
+I1203 23:25:06.379616 137274321021824 utils.py:1231] [83550] examples_seen = 85555200.0
+I1203 23:25:06.379680 137274321021824 utils.py:1231] [83550] progress = 0.7419873360390042
+I1203 23:25:06.379738 137274321021824 utils.py:1231] [83550] epoch = 66.77911622762684
+I1203 23:25:06.379797 137274321021824 utils.py:1231] [83550] img/sec/core = 164.2185086088599
+I1203 23:25:06.379860 137274321021824 utils.py:1231] [83550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.54783292924554
+I1203 23:25:06.379920 137274321021824 utils.py:1231] [83550] core_hours = 145.54783292924554
+I1203 23:25:06.379988 137274321021824 train.py:125] NOTE: Steps:83550/112603 [74.2%]
+Walltime:6d1h34m (0s eval)
+ETA:2d2h36m
+Total train time:8d4h9m
+I1203 23:30:18.107742 137274321021824 utils.py:1231] [83600] l2_params = 255.15725755346597
+I1203 23:30:18.107954 137274321021824 utils.py:1231] [83600] train/loss = 1.8163452595472336
+I1203 23:30:18.108050 137274321021824 utils.py:1231] [83600] l2_grads = 2.255070447921753
+I1203 23:30:18.108120 137274321021824 utils.py:1231] [83600] lr = 0.00018454510405248005
+I1203 23:30:18.108179 137274321021824 utils.py:1231] [83600] uptime = 524407.470532477
+I1203 23:30:18.108233 137274321021824 utils.py:1231] [83600] examples_seen = 85606400.0
+I1203 23:30:18.108281 137274321021824 utils.py:1231] [83600] progress = 0.7424313739420797
+I1203 23:30:18.108328 137274321021824 utils.py:1231] [83600] epoch = 66.8190797920958
+I1203 23:30:18.108377 137274321021824 utils.py:1231] [83600] img/sec/core = 164.2454267177565
+I1203 23:30:18.108432 137274321021824 utils.py:1231] [83600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.6344242122203
+I1203 23:30:18.108487 137274321021824 utils.py:1231] [83600] core_hours = 145.6344242122203
+I1203 23:30:18.108546 137274321021824 train.py:125] NOTE: Steps:83600/112603 [74.2%]
+Walltime:6d1h40m (0s eval)
+ETA:2d2h31m
+Total train time:8d4h9m
+I1203 23:35:29.902915 137274321021824 utils.py:1231] [83650] l2_params = 255.08960992611873
+I1203 23:35:29.903184 137274321021824 utils.py:1231] [83650] train/loss = 3.9399044513702393
+I1203 23:35:29.903311 137274321021824 utils.py:1231] [83650] l2_grads = 2.190464496612549
+I1203 23:35:29.903403 137274321021824 utils.py:1231] [83650] lr = 0.00018395157707117534
+I1203 23:35:29.903475 137274321021824 utils.py:1231] [83650] uptime = 524719.26583661
+I1203 23:35:29.903543 137274321021824 utils.py:1231] [83650] examples_seen = 85657600.0
+I1203 23:35:29.903611 137274321021824 utils.py:1231] [83650] progress = 0.7428754118451552
+I1203 23:35:29.903677 137274321021824 utils.py:1231] [83650] epoch = 66.85904335656475
+I1203 23:35:29.903745 137274321021824 utils.py:1231] [83650] img/sec/core = 164.21029862003368
+I1203 23:35:29.903824 137274321021824 utils.py:1231] [83650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.7210340189239
+I1203 23:35:29.903890 137274321021824 utils.py:1231] [83650] core_hours = 145.7210340189239
+I1203 23:35:29.903961 137274321021824 train.py:125] NOTE: Steps:83650/112603 [74.3%]
+Walltime:6d1h45m (0s eval)
+ETA:2d2h26m
+Total train time:8d4h9m
+I1203 23:40:41.694975 137274321021824 utils.py:1231] [83700] l2_params = 255.01739823041748
+I1203 23:40:41.695268 137274321021824 utils.py:1231] [83700] train/loss = 1.887774795293808
+I1203 23:40:41.695436 137274321021824 utils.py:1231] [83700] l2_grads = 2.3493895530700684
+I1203 23:40:41.695514 137274321021824 utils.py:1231] [83700] lr = 0.00018335879084245993
+I1203 23:40:41.695587 137274321021824 utils.py:1231] [83700] uptime = 525031.057949133
+I1203 23:40:41.695651 137274321021824 utils.py:1231] [83700] examples_seen = 85708800.0
+I1203 23:40:41.695714 137274321021824 utils.py:1231] [83700] progress = 0.7433194497482305
+I1203 23:40:41.695771 137274321021824 utils.py:1231] [83700] epoch = 66.89900692103372
+I1203 23:40:41.695830 137274321021824 utils.py:1231] [83700] img/sec/core = 164.21197953243458
+I1203 23:40:41.695907 137274321021824 utils.py:1231] [83700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.80764293906915
+I1203 23:40:41.695960 137274321021824 utils.py:1231] [83700] core_hours = 145.80764293906915
+I1203 23:40:41.696025 137274321021824 train.py:125] NOTE: Steps:83700/112603 [74.3%]
+Walltime:6d1h50m (0s eval)
+ETA:2d2h21m
+Total train time:8d4h9m
+I1203 23:45:53.406979 137274321021824 utils.py:1231] [83750] l2_params = 254.94972170226183
+I1203 23:45:53.407197 137274321021824 utils.py:1231] [83750] train/loss = 1.7799787819385529
+I1203 23:45:53.407302 137274321021824 utils.py:1231] [83750] l2_grads = 2.108846426010132
+I1203 23:45:53.407361 137274321021824 utils.py:1231] [83750] lr = 0.00018276674675570337
+I1203 23:45:53.407410 137274321021824 utils.py:1231] [83750] uptime = 525342.769772966
+I1203 23:45:53.407463 137274321021824 utils.py:1231] [83750] examples_seen = 85760000.0
+I1203 23:45:53.407511 137274321021824 utils.py:1231] [83750] progress = 0.743763487651306
+I1203 23:45:53.407559 137274321021824 utils.py:1231] [83750] epoch = 66.93897048550267
+I1203 23:45:53.407610 137274321021824 utils.py:1231] [83750] img/sec/core = 164.2542761785669
+I1203 23:45:53.407665 137274321021824 utils.py:1231] [83750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.89422955680055
+I1203 23:45:53.407718 137274321021824 utils.py:1231] [83750] core_hours = 145.89422955680055
+I1203 23:45:53.407777 137274321021824 train.py:125] NOTE: Steps:83750/112603 [74.4%]
+Walltime:6d1h55m (0s eval)
+ETA:2d2h15m
+Total train time:8d4h9m
+I1203 23:51:05.133028 137274321021824 utils.py:1231] [83800] l2_params = 254.87231840548952
+I1203 23:51:05.133247 137274321021824 utils.py:1231] [83800] train/loss = 3.1620709598064423
+I1203 23:51:05.133345 137274321021824 utils.py:1231] [83800] l2_grads = 1.929663896560669
+I1203 23:51:05.133418 137274321021824 utils.py:1231] [83800] lr = 0.00018217544619853488
+I1203 23:51:05.133483 137274321021824 utils.py:1231] [83800] uptime = 525654.495844861
+I1203 23:51:05.133560 137274321021824 utils.py:1231] [83800] examples_seen = 85811200.0
+I1203 23:51:05.133616 137274321021824 utils.py:1231] [83800] progress = 0.7442075255543813
+I1203 23:51:05.133670 137274321021824 utils.py:1231] [83800] epoch = 66.97893404997163
+I1203 23:51:05.133728 137274321021824 utils.py:1231] [83800] img/sec/core = 164.2467686092313
+I1203 23:51:05.133820 137274321021824 utils.py:1231] [83800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 145.98082013232695
+I1203 23:51:05.133877 137274321021824 utils.py:1231] [83800] core_hours = 145.98082013232695
+I1203 23:51:05.133959 137274321021824 train.py:125] NOTE: Steps:83800/112603 [74.4%]
+Walltime:6d2h0m (0s eval)
+ETA:2d2h10m
+Total train time:8d4h9m
+I1203 23:56:16.929028 137274321021824 utils.py:1231] [83850] l2_params = 254.80072402374492
+I1203 23:56:16.929239 137274321021824 utils.py:1231] [83850] train/loss = 1.7965342849493027
+I1203 23:56:16.929345 137274321021824 utils.py:1231] [83850] l2_grads = 2.268117666244507
+I1203 23:56:16.929422 137274321021824 utils.py:1231] [83850] lr = 0.00018158489055684183
+I1203 23:56:16.929483 137274321021824 utils.py:1231] [83850] uptime = 525966.291844647
+I1203 23:56:16.929546 137274321021824 utils.py:1231] [83850] examples_seen = 85862400.0
+I1203 23:56:16.929604 137274321021824 utils.py:1231] [83850] progress = 0.7446515634574568
+I1203 23:56:16.929663 137274321021824 utils.py:1231] [83850] epoch = 67.01889761444059
+I1203 23:56:16.929722 137274321021824 utils.py:1231] [83850] img/sec/core = 164.209932247818
+I1203 23:56:16.929782 137274321021824 utils.py:1231] [83850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.0674301322675
+I1203 23:56:16.929840 137274321021824 utils.py:1231] [83850] core_hours = 146.0674301322675
+I1203 23:56:16.929919 137274321021824 train.py:125] NOTE: Steps:83850/112603 [74.5%]
+Walltime:6d2h6m (0s eval)
+ETA:2d2h5m
+Total train time:8d4h9m
+I1204 00:01:28.730776 137274321021824 utils.py:1231] [83900] l2_params = 254.73114229076867
+I1204 00:01:28.731039 137274321021824 utils.py:1231] [83900] train/loss = 1.8632322251796722
+I1204 00:01:28.731174 137274321021824 utils.py:1231] [83900] l2_grads = 2.1310176849365234
+I1204 00:01:28.731241 137274321021824 utils.py:1231] [83900] lr = 0.00018099508121476518
+I1204 00:01:28.731293 137274321021824 utils.py:1231] [83900] uptime = 526278.093655768
+I1204 00:01:28.731347 137274321021824 utils.py:1231] [83900] examples_seen = 85913600.0
+I1204 00:01:28.731396 137274321021824 utils.py:1231] [83900] progress = 0.7450956013605321
+I1204 00:01:28.731445 137274321021824 utils.py:1231] [83900] epoch = 67.05886117890954
+I1204 00:01:28.731495 137274321021824 utils.py:1231] [83900] img/sec/core = 164.2068717174355
+I1204 00:01:28.731551 137274321021824 utils.py:1231] [83900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.15404174646775
+I1204 00:01:28.731601 137274321021824 utils.py:1231] [83900] core_hours = 146.15404174646775
+I1204 00:01:28.731663 137274321021824 train.py:125] NOTE: Steps:83900/112603 [74.5%]
+Walltime:6d2h11m (0s eval)
+ETA:2d2h0m
+Total train time:8d4h9m
+I1204 00:06:40.507553 137274321021824 utils.py:1231] [83950] l2_params = 254.6626875680787
+I1204 00:06:40.507787 137274321021824 utils.py:1231] [83950] train/loss = 1.8311996310949326
+I1204 00:06:40.507921 137274321021824 utils.py:1231] [83950] l2_grads = 2.0936923027038574
+I1204 00:06:40.507998 137274321021824 utils.py:1231] [83950] lr = 0.00018040601955469682
+I1204 00:06:40.508067 137274321021824 utils.py:1231] [83950] uptime = 526589.870428486
+I1204 00:06:40.508127 137274321021824 utils.py:1231] [83950] examples_seen = 85964800.0
+I1204 00:06:40.508183 137274321021824 utils.py:1231] [83950] progress = 0.7455396392636076
+I1204 00:06:40.508238 137274321021824 utils.py:1231] [83950] epoch = 67.0988247433785
+I1204 00:06:40.508295 137274321021824 utils.py:1231] [83950] img/sec/core = 164.2200589660344
+I1204 00:06:40.508352 137274321021824 utils.py:1231] [83950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.2406464055561
+I1204 00:06:40.508406 137274321021824 utils.py:1231] [83950] core_hours = 146.2406464055561
+I1204 00:06:40.508471 137274321021824 train.py:125] NOTE: Steps:83950/112603 [74.6%]
+Walltime:6d2h16m (0s eval)
+ETA:2d1h54m
+Total train time:8d4h9m
+I1204 00:11:52.308506 137274321021824 utils.py:1231] [84000] l2_params = 254.59904218672494
+I1204 00:11:52.308762 137274321021824 utils.py:1231] [84000] train/loss = 4.101737201213837
+I1204 00:11:52.308878 137274321021824 utils.py:1231] [84000] l2_grads = 2.3007352352142334
+I1204 00:11:52.308957 137274321021824 utils.py:1231] [84000] lr = 0.0001798177069572765
+I1204 00:11:52.309007 137274321021824 utils.py:1231] [84000] uptime = 526901.671369512
+I1204 00:11:52.309057 137274321021824 utils.py:1231] [84000] examples_seen = 86016000.0
+I1204 00:11:52.309105 137274321021824 utils.py:1231] [84000] progress = 0.7459836771666829
+I1204 00:11:52.309151 137274321021824 utils.py:1231] [84000] epoch = 67.13878830784745
+I1204 00:11:52.309200 137274321021824 utils.py:1231] [84000] img/sec/core = 164.20732994432512
+I1204 00:11:52.309256 137274321021824 utils.py:1231] [84000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.32725777806334
+I1204 00:11:52.309309 137274321021824 utils.py:1231] [84000] core_hours = 146.32725777806334
+I1204 00:11:52.309371 137274321021824 train.py:125] NOTE: Steps:84000/112603 [74.6%]
+Walltime:6d2h21m (0s eval)
+ETA:2d1h49m
+Total train time:8d4h9m
+I1204 00:17:04.495091 137274321021824 utils.py:1231] [84050] l2_params = 254.53189954516742
+I1204 00:17:04.495343 137274321021824 utils.py:1231] [84050] train/loss = 1.871335357427597
+I1204 00:17:04.495492 137274321021824 utils.py:1231] [84050] l2_grads = 2.2026638984680176
+I1204 00:17:04.495591 137274321021824 utils.py:1231] [84050] lr = 0.00017923014480138803
+I1204 00:17:04.495661 137274321021824 utils.py:1231] [84050] uptime = 527213.8580223899
+I1204 00:17:04.495726 137274321021824 utils.py:1231] [84050] examples_seen = 86067200.0
+I1204 00:17:04.495790 137274321021824 utils.py:1231] [84050] progress = 0.7464277150697584
+I1204 00:17:04.495852 137274321021824 utils.py:1231] [84050] epoch = 67.17875187231641
+I1204 00:17:04.495922 137274321021824 utils.py:1231] [84050] img/sec/core = 164.0044490307429
+I1204 00:17:04.495990 137274321021824 utils.py:1231] [84050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.41397629275164
+I1204 00:17:04.496055 137274321021824 utils.py:1231] [84050] core_hours = 146.41397629275164
+I1204 00:17:04.496121 137274321021824 train.py:125] NOTE: Steps:84050/112603 [74.6%]
+Walltime:6d2h26m (0s eval)
+ETA:2d1h44m
+Total train time:8d4h9m
+I1204 00:22:16.152238 137274321021824 utils.py:1231] [84100] l2_params = 254.46346687087387
+I1204 00:22:16.152507 137274321021824 utils.py:1231] [84100] train/loss = 2.619202136993408
+I1204 00:22:16.152641 137274321021824 utils.py:1231] [84100] l2_grads = 2.1673948764801025
+I1204 00:22:16.152728 137274321021824 utils.py:1231] [84100] lr = 0.00017864333446415654
+I1204 00:22:16.152805 137274321021824 utils.py:1231] [84100] uptime = 527525.515165568
+I1204 00:22:16.152880 137274321021824 utils.py:1231] [84100] examples_seen = 86118400.0
+I1204 00:22:16.152964 137274321021824 utils.py:1231] [84100] progress = 0.7468717529728338
+I1204 00:22:16.153024 137274321021824 utils.py:1231] [84100] epoch = 67.21871543678537
+I1204 00:22:16.153089 137274321021824 utils.py:1231] [84100] img/sec/core = 164.28309480700702
+I1204 00:22:16.153159 137274321021824 utils.py:1231] [84100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.5005477214122
+I1204 00:22:16.153220 137274321021824 utils.py:1231] [84100] core_hours = 146.5005477214122
+I1204 00:22:16.153294 137274321021824 train.py:125] NOTE: Steps:84100/112603 [74.7%]
+Walltime:6d2h32m (0s eval)
+ETA:2d1h39m
+Total train time:8d4h9m
+I1204 00:27:27.944102 137274321021824 utils.py:1231] [84150] l2_params = 254.39210197943592
+I1204 00:27:27.944347 137274321021824 utils.py:1231] [84150] train/loss = 2.564710885286331
+I1204 00:27:27.944472 137274321021824 utils.py:1231] [84150] l2_grads = 2.057534694671631
+I1204 00:27:27.944540 137274321021824 utils.py:1231] [84150] lr = 0.00017805727732094474
+I1204 00:27:27.944611 137274321021824 utils.py:1231] [84150] uptime = 527837.306968071
+I1204 00:27:27.944666 137274321021824 utils.py:1231] [84150] examples_seen = 86169600.0
+I1204 00:27:27.944718 137274321021824 utils.py:1231] [84150] progress = 0.7473157908759092
+I1204 00:27:27.944768 137274321021824 utils.py:1231] [84150] epoch = 67.25867900125432
+I1204 00:27:27.944821 137274321021824 utils.py:1231] [84150] img/sec/core = 164.21214281122352
+I1204 00:27:27.944898 137274321021824 utils.py:1231] [84150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.58715655544083
+I1204 00:27:27.944950 137274321021824 utils.py:1231] [84150] core_hours = 146.58715655544083
+I1204 00:27:27.945012 137274321021824 train.py:125] NOTE: Steps:84150/112603 [74.7%]
+Walltime:6d2h37m (0s eval)
+ETA:2d1h33m
+Total train time:8d4h9m
+I1204 00:32:39.739086 137274321021824 utils.py:1231] [84200] l2_params = 254.3244283018096
+I1204 00:32:39.739277 137274321021824 utils.py:1231] [84200] train/loss = 3.596930503845215
+I1204 00:32:39.739382 137274321021824 utils.py:1231] [84200] l2_grads = 2.0693094730377197
+I1204 00:32:39.739451 137274321021824 utils.py:1231] [84200] lr = 0.0001774719747453501
+I1204 00:32:39.739510 137274321021824 utils.py:1231] [84200] uptime = 528149.10187138
+I1204 00:32:39.739572 137274321021824 utils.py:1231] [84200] examples_seen = 86220800.0
+I1204 00:32:39.739627 137274321021824 utils.py:1231] [84200] progress = 0.7477598287789846
+I1204 00:32:39.739691 137274321021824 utils.py:1231] [84200] epoch = 67.29864256572328
+I1204 00:32:39.739749 137274321021824 utils.py:1231] [84200] img/sec/core = 164.2105097185126
+I1204 00:32:39.739840 137274321021824 utils.py:1231] [84200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.67376625080445
+I1204 00:32:39.739957 137274321021824 utils.py:1231] [84200] core_hours = 146.67376625080445
+I1204 00:32:39.740039 137274321021824 train.py:125] NOTE: Steps:84200/112603 [74.8%]
+Walltime:6d2h42m (0s eval)
+ETA:2d1h28m
+Total train time:8d4h9m
+I1204 00:37:51.526897 137274321021824 utils.py:1231] [84250] l2_params = 254.26109555156015
+I1204 00:37:51.527115 137274321021824 utils.py:1231] [84250] train/loss = 1.7173095792531967
+I1204 00:37:51.527212 137274321021824 utils.py:1231] [84250] l2_grads = 2.2133073806762695
+I1204 00:37:51.527294 137274321021824 utils.py:1231] [84250] lr = 0.00017688742810920214
+I1204 00:37:51.527367 137274321021824 utils.py:1231] [84250] uptime = 528460.88972387
+I1204 00:37:51.527440 137274321021824 utils.py:1231] [84250] examples_seen = 86272000.0
+I1204 00:37:51.527499 137274321021824 utils.py:1231] [84250] progress = 0.74820386668206
+I1204 00:37:51.527562 137274321021824 utils.py:1231] [84250] epoch = 67.33860613019223
+I1204 00:37:51.527613 137274321021824 utils.py:1231] [84250] img/sec/core = 164.21422320052324
+I1204 00:37:51.527669 137274321021824 utils.py:1231] [84250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.7603739876072
+I1204 00:37:51.527719 137274321021824 utils.py:1231] [84250] core_hours = 146.7603739876072
+I1204 00:37:51.527783 137274321021824 train.py:125] NOTE: Steps:84250/112603 [74.8%]
+Walltime:6d2h47m (0s eval)
+ETA:2d1h23m
+Total train time:8d4h9m
+I1204 00:43:03.314257 137274321021824 utils.py:1231] [84300] l2_params = 254.19351199004512
+I1204 00:43:03.314493 137274321021824 utils.py:1231] [84300] train/loss = 1.7395994514226913
+I1204 00:43:03.314589 137274321021824 utils.py:1231] [84300] l2_grads = 2.1494534015655518
+I1204 00:43:03.314655 137274321021824 utils.py:1231] [84300] lr = 0.00017630363878255758
+I1204 00:43:03.314705 137274321021824 utils.py:1231] [84300] uptime = 528772.677067171
+I1204 00:43:03.314757 137274321021824 utils.py:1231] [84300] examples_seen = 86323200.0
+I1204 00:43:03.314805 137274321021824 utils.py:1231] [84300] progress = 0.7486479045851354
+I1204 00:43:03.314853 137274321021824 utils.py:1231] [84300] epoch = 67.3785696946612
+I1204 00:43:03.314913 137274321021824 utils.py:1231] [84300] img/sec/core = 164.21449138354598
+I1204 00:43:03.314969 137274321021824 utils.py:1231] [84300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.8469815829686
+I1204 00:43:03.315018 137274321021824 utils.py:1231] [84300] core_hours = 146.8469815829686
+I1204 00:43:03.315079 137274321021824 train.py:125] NOTE: Steps:84300/112603 [74.9%]
+Walltime:6d2h52m (0s eval)
+ETA:2d1h18m
+Total train time:8d4h9m
+I1204 00:48:15.106009 137274321021824 utils.py:1231] [84350] l2_params = 254.12875041631168
+I1204 00:48:15.106251 137274321021824 utils.py:1231] [84350] train/loss = 1.957698032259941
+I1204 00:48:15.106382 137274321021824 utils.py:1231] [84350] l2_grads = 2.204763174057007
+I1204 00:48:15.106469 137274321021824 utils.py:1231] [84350] lr = 0.000175720608133699
+I1204 00:48:15.106565 137274321021824 utils.py:1231] [84350] uptime = 529084.468922292
+I1204 00:48:15.106643 137274321021824 utils.py:1231] [84350] examples_seen = 86374400.0
+I1204 00:48:15.106708 137274321021824 utils.py:1231] [84350] progress = 0.7490919424882108
+I1204 00:48:15.106764 137274321021824 utils.py:1231] [84350] epoch = 67.41853325913016
+I1204 00:48:15.106832 137274321021824 utils.py:1231] [84350] img/sec/core = 164.21211509878978
+I1204 00:48:15.106906 137274321021824 utils.py:1231] [84350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 146.93359043161334
+I1204 00:48:15.106965 137274321021824 utils.py:1231] [84350] core_hours = 146.93359043161334
+I1204 00:48:15.107033 137274321021824 train.py:125] NOTE: Steps:84350/112603 [74.9%]
+Walltime:6d2h58m (0s eval)
+ETA:2d1h12m
+Total train time:8d4h9m
+I1204 00:53:26.866150 137274321021824 utils.py:1231] [84400] l2_params = 254.05498446624873
+I1204 00:53:26.866381 137274321021824 utils.py:1231] [84400] train/loss = 2.1602950543165207
+I1204 00:53:26.866490 137274321021824 utils.py:1231] [84400] l2_grads = 1.9806580543518066
+I1204 00:53:26.866564 137274321021824 utils.py:1231] [84400] lr = 0.0001751383375291303
+I1204 00:53:26.866630 137274321021824 utils.py:1231] [84400] uptime = 529396.228991109
+I1204 00:53:26.866692 137274321021824 utils.py:1231] [84400] examples_seen = 86425600.0
+I1204 00:53:26.866751 137274321021824 utils.py:1231] [84400] progress = 0.7495359803912862
+I1204 00:53:26.866809 137274321021824 utils.py:1231] [84400] epoch = 67.4584968235991
+I1204 00:53:26.866872 137274321021824 utils.py:1231] [84400] img/sec/core = 164.22885776966666
+I1204 00:53:26.866958 137274321021824 utils.py:1231] [84400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.02019045072913
+I1204 00:53:26.867016 137274321021824 utils.py:1231] [84400] core_hours = 147.02019045072913
+I1204 00:53:26.867080 137274321021824 train.py:125] NOTE: Steps:84400/112603 [75.0%]
+Walltime:6d3h3m (0s eval)
+ETA:2d1h7m
+Total train time:8d4h9m
+I1204 00:58:38.649017 137274321021824 utils.py:1231] [84450] l2_params = 253.9939251081972
+I1204 00:58:38.649221 137274321021824 utils.py:1231] [84450] train/loss = 1.718667909502983
+I1204 00:58:38.649314 137274321021824 utils.py:1231] [84450] l2_grads = 2.197028636932373
+I1204 00:58:38.649376 137274321021824 utils.py:1231] [84450] lr = 0.00017455682833357445
+I1204 00:58:38.649432 137274321021824 utils.py:1231] [84450] uptime = 529708.011793697
+I1204 00:58:38.649488 137274321021824 utils.py:1231] [84450] examples_seen = 86476800.0
+I1204 00:58:38.649539 137274321021824 utils.py:1231] [84450] progress = 0.7499800182943616
+I1204 00:58:38.649590 137274321021824 utils.py:1231] [84450] epoch = 67.49846038806807
+I1204 00:58:38.649643 137274321021824 utils.py:1231] [84450] img/sec/core = 164.21688295503995
+I1204 00:58:38.649700 137274321021824 utils.py:1231] [84450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.10679678478138
+I1204 00:58:38.649753 137274321021824 utils.py:1231] [84450] core_hours = 147.10679678478138
+I1204 00:58:38.649816 137274321021824 train.py:125] NOTE: Steps:84450/112603 [75.0%]
+Walltime:6d3h8m (0s eval)
+ETA:2d1h2m
+Total train time:8d4h9m
+I1204 01:03:50.432989 137274321021824 utils.py:1231] [84500] l2_params = 253.92514178398955
+I1204 01:03:50.433238 137274321021824 utils.py:1231] [84500] train/loss = 2.1011398434638977
+I1204 01:03:50.433362 137274321021824 utils.py:1231] [84500] l2_grads = 2.35939621925354
+I1204 01:03:50.433450 137274321021824 utils.py:1231] [84500] lr = 0.00017397608190996902
+I1204 01:03:50.433505 137274321021824 utils.py:1231] [84500] uptime = 530019.795867686
+I1204 01:03:50.433557 137274321021824 utils.py:1231] [84500] examples_seen = 86528000.0
+I1204 01:03:50.433604 137274321021824 utils.py:1231] [84500] progress = 0.750424056197437
+I1204 01:03:50.433651 137274321021824 utils.py:1231] [84500] epoch = 67.53842395253702
+I1204 01:03:50.433709 137274321021824 utils.py:1231] [84500] img/sec/core = 164.21621330729846
+I1204 01:03:50.433774 137274321021824 utils.py:1231] [84500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.19340347200057
+I1204 01:03:50.433829 137274321021824 utils.py:1231] [84500] core_hours = 147.19340347200057
+I1204 01:03:50.433901 137274321021824 train.py:125] NOTE: Steps:84500/112603 [75.0%]
+Walltime:6d3h13m (0s eval)
+ETA:2d0h57m
+Total train time:8d4h9m
+I1204 01:09:02.175504 137274321021824 utils.py:1231] [84550] l2_params = 253.85545817600905
+I1204 01:09:02.175794 137274321021824 utils.py:1231] [84550] train/loss = 3.114757150411606
+I1204 01:09:02.175966 137274321021824 utils.py:1231] [84550] l2_grads = 1.997761607170105
+I1204 01:09:02.176030 137274321021824 utils.py:1231] [84550] lr = 0.00017339609961946494
+I1204 01:09:02.176082 137274321021824 utils.py:1231] [84550] uptime = 530331.5384441459
+I1204 01:09:02.176136 137274321021824 utils.py:1231] [84550] examples_seen = 86579200.0
+I1204 01:09:02.176186 137274321021824 utils.py:1231] [84550] progress = 0.7508680941005125
+I1204 01:09:02.176236 137274321021824 utils.py:1231] [84550] epoch = 67.57838751700598
+I1204 01:09:02.176291 137274321021824 utils.py:1231] [84550] img/sec/core = 164.23807290432097
+I1204 01:09:02.176347 137274321021824 utils.py:1231] [84550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.27999863212833
+I1204 01:09:02.176399 137274321021824 utils.py:1231] [84550] core_hours = 147.27999863212833
+I1204 01:09:02.176462 137274321021824 train.py:125] NOTE: Steps:84550/112603 [75.1%]
+Walltime:6d3h18m (0s eval)
+ETA:2d0h52m
+Total train time:8d4h9m
+I1204 01:14:13.942716 137274321021824 utils.py:1231] [84600] l2_params = 253.79268223190803
+I1204 01:14:13.943017 137274321021824 utils.py:1231] [84600] train/loss = 3.8515017926692963
+I1204 01:14:13.943198 137274321021824 utils.py:1231] [84600] l2_grads = 2.106050968170166
+I1204 01:14:13.943277 137274321021824 utils.py:1231] [84600] lr = 0.00017281688282142147
+I1204 01:14:13.943357 137274321021824 utils.py:1231] [84600] uptime = 530643.305717947
+I1204 01:14:13.943457 137274321021824 utils.py:1231] [84600] examples_seen = 86630400.0
+I1204 01:14:13.943531 137274321021824 utils.py:1231] [84600] progress = 0.7513121320035878
+I1204 01:14:13.943601 137274321021824 utils.py:1231] [84600] epoch = 67.61835108147494
+I1204 01:14:13.943680 137274321021824 utils.py:1231] [84600] img/sec/core = 164.2250624184369
+I1204 01:14:13.943789 137274321021824 utils.py:1231] [84600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.3666006526286
+I1204 01:14:13.943853 137274321021824 utils.py:1231] [84600] core_hours = 147.3666006526286
+I1204 01:14:13.943950 137274321021824 train.py:125] NOTE: Steps:84600/112603 [75.1%]
+Walltime:6d3h24m (0s eval)
+ETA:2d0h46m
+Total train time:8d4h9m
+I1204 01:19:25.722283 137274321021824 utils.py:1231] [84650] l2_params = 253.72860027673855
+I1204 01:19:25.722508 137274321021824 utils.py:1231] [84650] train/loss = 2.911663204431534
+I1204 01:19:25.722613 137274321021824 utils.py:1231] [84650] l2_grads = 1.9685145616531372
+I1204 01:19:25.722689 137274321021824 utils.py:1231] [84650] lr = 0.00017223843287340345
+I1204 01:19:25.722752 137274321021824 utils.py:1231] [84650] uptime = 530955.085112296
+I1204 01:19:25.722814 137274321021824 utils.py:1231] [84650] examples_seen = 86681600.0
+I1204 01:19:25.722872 137274321021824 utils.py:1231] [84650] progress = 0.7517561699066633
+I1204 01:19:25.722934 137274321021824 utils.py:1231] [84650] epoch = 67.65831464594389
+I1204 01:19:25.722993 137274321021824 utils.py:1231] [84650] img/sec/core = 164.21867810376847
+I1204 01:19:25.723056 137274321021824 utils.py:1231] [84650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.4532060399478
+I1204 01:19:25.723114 137274321021824 utils.py:1231] [84650] core_hours = 147.4532060399478
+I1204 01:19:25.723181 137274321021824 train.py:125] NOTE: Steps:84650/112603 [75.2%]
+Walltime:6d3h29m (0s eval)
+ETA:2d0h41m
+Total train time:8d4h8m
+I1204 01:24:37.351442 137274321021824 utils.py:1231] [84700] l2_params = 253.66027944336162
+I1204 01:24:37.351708 137274321021824 utils.py:1231] [84700] train/loss = 3.2381197214126587
+I1204 01:24:37.351818 137274321021824 utils.py:1231] [84700] l2_grads = 2.037768840789795
+I1204 01:24:37.351916 137274321021824 utils.py:1231] [84700] lr = 0.00017166075113117933
+I1204 01:24:37.351978 137274321021824 utils.py:1231] [84700] uptime = 531266.714338991
+I1204 01:24:37.352039 137274321021824 utils.py:1231] [84700] examples_seen = 86732800.0
+I1204 01:24:37.352095 137274321021824 utils.py:1231] [84700] progress = 0.7522002078097386
+I1204 01:24:37.352150 137274321021824 utils.py:1231] [84700] epoch = 67.69827821041285
+I1204 01:24:37.352206 137274321021824 utils.py:1231] [84700] img/sec/core = 164.2978116751614
+I1204 01:24:37.352266 137274321021824 utils.py:1231] [84700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.53976971402972
+I1204 01:24:37.352316 137274321021824 utils.py:1231] [84700] core_hours = 147.53976971402972
+I1204 01:24:37.352384 137274321021824 train.py:125] NOTE: Steps:84700/112603 [75.2%]
+Walltime:6d3h34m (0s eval)
+ETA:2d0h36m
+Total train time:8d4h8m
+I1204 01:29:49.128359 137274321021824 utils.py:1231] [84750] l2_params = 253.5945033960674
+I1204 01:29:49.128567 137274321021824 utils.py:1231] [84750] train/loss = 1.7538829147815704
+I1204 01:29:49.128663 137274321021824 utils.py:1231] [84750] l2_grads = 2.285231113433838
+I1204 01:29:49.128733 137274321021824 utils.py:1231] [84750] lr = 0.00017108383894871582
+I1204 01:29:49.128785 137274321021824 utils.py:1231] [84750] uptime = 531578.4911473889
+I1204 01:29:49.128837 137274321021824 utils.py:1231] [84750] examples_seen = 86784000.0
+I1204 01:29:49.128891 137274321021824 utils.py:1231] [84750] progress = 0.7526442457128141
+I1204 01:29:49.128944 137274321021824 utils.py:1231] [84750] epoch = 67.7382417748818
+I1204 01:29:49.128996 137274321021824 utils.py:1231] [84750] img/sec/core = 164.22004017260596
+I1204 01:29:49.129051 137274321021824 utils.py:1231] [84750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.62637438302914
+I1204 01:29:49.129102 137274321021824 utils.py:1231] [84750] core_hours = 147.62637438302914
+I1204 01:29:49.129163 137274321021824 train.py:125] NOTE: Steps:84750/112603 [75.3%]
+Walltime:6d3h39m (0s eval)
+ETA:2d0h31m
+Total train time:8d4h8m
+I1204 01:35:00.918467 137274321021824 utils.py:1231] [84800] l2_params = 253.52252070613332
+I1204 01:35:00.918731 137274321021824 utils.py:1231] [84800] train/loss = 3.9198046028614044
+I1204 01:35:00.918852 137274321021824 utils.py:1231] [84800] l2_grads = 2.2048237323760986
+I1204 01:35:00.918936 137274321021824 utils.py:1231] [84800] lr = 0.00017050769767817703
+I1204 01:35:00.918988 137274321021824 utils.py:1231] [84800] uptime = 531890.281350254
+I1204 01:35:00.919043 137274321021824 utils.py:1231] [84800] examples_seen = 86835200.0
+I1204 01:35:00.919094 137274321021824 utils.py:1231] [84800] progress = 0.7530882836158894
+I1204 01:35:00.919143 137274321021824 utils.py:1231] [84800] epoch = 67.77820533935076
+I1204 01:35:00.919196 137274321021824 utils.py:1231] [84800] img/sec/core = 164.21298530077348
+I1204 01:35:00.919254 137274321021824 utils.py:1231] [84800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.71298277271387
+I1204 01:35:00.919306 137274321021824 utils.py:1231] [84800] core_hours = 147.71298277271387
+I1204 01:35:00.919367 137274321021824 train.py:125] NOTE: Steps:84800/112603 [75.3%]
+Walltime:6d3h44m (0s eval)
+ETA:2d0h25m
+Total train time:8d4h8m
+I1204 01:40:12.788678 137274321021824 utils.py:1231] [84850] l2_params = 253.45586079352717
+I1204 01:40:12.788935 137274321021824 utils.py:1231] [84850] train/loss = 1.9136294722557068
+I1204 01:40:12.789082 137274321021824 utils.py:1231] [84850] l2_grads = 2.215712308883667
+I1204 01:40:12.789196 137274321021824 utils.py:1231] [84850] lr = 0.00016993232866991984
+I1204 01:40:12.789283 137274321021824 utils.py:1231] [84850] uptime = 532202.151644902
+I1204 01:40:12.789353 137274321021824 utils.py:1231] [84850] examples_seen = 86886400.0
+I1204 01:40:12.789422 137274321021824 utils.py:1231] [84850] progress = 0.7535323215189649
+I1204 01:40:12.789490 137274321021824 utils.py:1231] [84850] epoch = 67.81816890381972
+I1204 01:40:12.789556 137274321021824 utils.py:1231] [84850] img/sec/core = 164.1708135678321
+I1204 01:40:12.789634 137274321021824 utils.py:1231] [84850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.7996134101161
+I1204 01:40:12.789695 137274321021824 utils.py:1231] [84850] core_hours = 147.7996134101161
+I1204 01:40:12.789765 137274321021824 train.py:125] NOTE: Steps:84850/112603 [75.4%]
+Walltime:6d3h50m (0s eval)
+ETA:2d0h20m
+Total train time:8d4h8m
+I1204 01:45:24.589964 137274321021824 utils.py:1231] [84900] l2_params = 253.38580164411937
+I1204 01:45:24.590201 137274321021824 utils.py:1231] [84900] train/loss = 2.4975059032440186
+I1204 01:45:24.590357 137274321021824 utils.py:1231] [84900] l2_grads = 1.9910582304000854
+I1204 01:45:24.590462 137274321021824 utils.py:1231] [84900] lr = 0.0001693577332724906
+I1204 01:45:24.590538 137274321021824 utils.py:1231] [84900] uptime = 532513.952899458
+I1204 01:45:24.590612 137274321021824 utils.py:1231] [84900] examples_seen = 86937600.0
+I1204 01:45:24.590671 137274321021824 utils.py:1231] [84900] progress = 0.7539763594220402
+I1204 01:45:24.590730 137274321021824 utils.py:1231] [84900] epoch = 67.85813246828867
+I1204 01:45:24.590792 137274321021824 utils.py:1231] [84900] img/sec/core = 164.2071648265174
+I1204 01:45:24.590874 137274321021824 utils.py:1231] [84900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.886224869715
+I1204 01:45:24.590944 137274321021824 utils.py:1231] [84900] core_hours = 147.886224869715
+I1204 01:45:24.591013 137274321021824 train.py:125] NOTE: Steps:84900/112603 [75.4%]
+Walltime:6d3h55m (0s eval)
+ETA:2d0h15m
+Total train time:8d4h8m
+I1204 01:50:36.399404 137274321021824 utils.py:1231] [84950] l2_params = 253.31942519762674
+I1204 01:50:36.399636 137274321021824 utils.py:1231] [84950] train/loss = 1.894696056842804
+I1204 01:50:36.399795 137274321021824 utils.py:1231] [84950] l2_grads = 2.2700552940368652
+I1204 01:50:36.399876 137274321021824 utils.py:1231] [84950] lr = 0.00016878391283262344
+I1204 01:50:36.399942 137274321021824 utils.py:1231] [84950] uptime = 532825.762303461
+I1204 01:50:36.400002 137274321021824 utils.py:1231] [84950] examples_seen = 86988800.0
+I1204 01:50:36.400058 137274321021824 utils.py:1231] [84950] progress = 0.7544203973251157
+I1204 01:50:36.400112 137274321021824 utils.py:1231] [84950] epoch = 67.89809603275764
+I1204 01:50:36.400173 137274321021824 utils.py:1231] [84950] img/sec/core = 164.2028731099868
+I1204 01:50:36.400230 137274321021824 utils.py:1231] [84950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 147.97283859304918
+I1204 01:50:36.400284 137274321021824 utils.py:1231] [84950] core_hours = 147.97283859304918
+I1204 01:50:36.400348 137274321021824 train.py:125] NOTE: Steps:84950/112603 [75.4%]
+Walltime:6d4h0m (0s eval)
+ETA:2d0h10m
+Total train time:8d4h8m
+I1204 01:55:48.201010 137274321021824 utils.py:1231] [85000] l2_params = 253.25479680917184
+I1204 01:55:48.201202 137274321021824 utils.py:1231] [85000] train/loss = 1.746158480644226
+I1204 01:55:48.201298 137274321021824 utils.py:1231] [85000] l2_grads = 2.2329275608062744
+I1204 01:55:48.201356 137274321021824 utils.py:1231] [85000] lr = 0.000168210868695235
+I1204 01:55:48.201407 137274321021824 utils.py:1231] [85000] uptime = 533137.563769698
+I1204 01:55:48.201461 137274321021824 utils.py:1231] [85000] examples_seen = 87040000.0
+I1204 01:55:48.201513 137274321021824 utils.py:1231] [85000] progress = 0.7548644352281911
+I1204 01:55:48.201562 137274321021824 utils.py:1231] [85000] epoch = 67.9380595972266
+I1204 01:55:48.201613 137274321021824 utils.py:1231] [85000] img/sec/core = 164.20705334680457
+I1204 01:55:48.201670 137274321021824 utils.py:1231] [85000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.05945011144834
+I1204 01:55:48.201721 137274321021824 utils.py:1231] [85000] core_hours = 148.05945011144834
+I1204 01:55:48.201781 137274321021824 train.py:125] NOTE: Steps:85000/112603 [75.5%]
+Walltime:6d4h5m (0s eval)
+ETA:2d0h4m
+Total train time:8d4h8m
+I1204 01:55:48.555376 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:85000/112603 [75.5%]
+Walltime:6d4h5m (0s eval)
+ETA:2d0h4m
+Total train time:8d4h8m
+I1204 01:57:23.876039 137274321021824 utils.py:1231] [85000] val/acc@1 = 0.7327008928571429
+I1204 01:57:23.876297 137274321021824 utils.py:1231] [85000] val/loss = 1.0706414431333542
+I1204 01:57:23.876482 137274321021824 utils.py:1231] [85000] z/secs/eval/val = 95.32086947804783
+I1204 01:57:23.876583 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 95.32086947804783
+I1204 02:02:34.375715 137274321021824 utils.py:1231] [85050] l2_params = 253.1872492343807
+I1204 02:02:34.375980 137274321021824 utils.py:1231] [85050] train/loss = 2.987929105758667
+I1204 02:02:34.376117 137274321021824 utils.py:1231] [85050] l2_grads = 2.062725067138672
+I1204 02:02:34.376218 137274321021824 utils.py:1231] [85050] lr = 0.00016763860220342334
+I1204 02:02:34.376289 137274321021824 utils.py:1231] [85050] uptime = 533543.738650234
+I1204 02:02:34.376361 137274321021824 utils.py:1231] [85050] examples_seen = 87091200.0
+I1204 02:02:34.376436 137274321021824 utils.py:1231] [85050] progress = 0.7553084731312665
+I1204 02:02:34.376493 137274321021824 utils.py:1231] [85050] epoch = 67.97802316169555
+I1204 02:02:34.376554 137274321021824 utils.py:1231] [85050] img/sec/core = 126.05407782097359
+I1204 02:02:34.376618 137274321021824 utils.py:1231] [85050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.17227646715276
+I1204 02:02:34.376686 137274321021824 utils.py:1231] [85050] core_hours = 148.17227646715276
+I1204 02:02:34.376764 137274321021824 train.py:125] NOTE: Steps:85050/112603 [75.5%]
+Walltime:6d4h12m (0s eval)
+ETA:2d0h0m
+Total train time:8d4h10m
+I1204 02:07:46.161035 137274321021824 utils.py:1231] [85100] l2_params = 253.11716319927402
+I1204 02:07:46.161281 137274321021824 utils.py:1231] [85100] train/loss = 4.05662214756012
+I1204 02:07:46.161391 137274321021824 utils.py:1231] [85100] l2_grads = 2.271146297454834
+I1204 02:07:46.161464 137274321021824 utils.py:1231] [85100] lr = 0.00016706711469846372
+I1204 02:07:46.161524 137274321021824 utils.py:1231] [85100] uptime = 533855.523885291
+I1204 02:07:46.161597 137274321021824 utils.py:1231] [85100] examples_seen = 87142400.0
+I1204 02:07:46.161673 137274321021824 utils.py:1231] [85100] progress = 0.7557525110343419
+I1204 02:07:46.161739 137274321021824 utils.py:1231] [85100] epoch = 68.01798672616451
+I1204 02:07:46.161810 137274321021824 utils.py:1231] [85100] img/sec/core = 164.2156017767783
+I1204 02:07:46.161892 137274321021824 utils.py:1231] [85100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.25888347689082
+I1204 02:07:46.161952 137274321021824 utils.py:1231] [85100] core_hours = 148.25888347689082
+I1204 02:07:46.162016 137274321021824 train.py:125] NOTE: Steps:85100/112603 [75.6%]
+Walltime:6d4h17m (0s eval)
+ETA:1d23h54m
+Total train time:8d4h10m
+I1204 02:12:57.950079 137274321021824 utils.py:1231] [85150] l2_params = 253.05456433262734
+I1204 02:12:57.950310 137274321021824 utils.py:1231] [85150] train/loss = 1.7482470273971558
+I1204 02:12:57.950433 137274321021824 utils.py:1231] [85150] l2_grads = 2.351809501647949
+I1204 02:12:57.950510 137274321021824 utils.py:1231] [85150] lr = 0.00016649640751980534
+I1204 02:12:57.950578 137274321021824 utils.py:1231] [85150] uptime = 534167.312939099
+I1204 02:12:57.950648 137274321021824 utils.py:1231] [85150] examples_seen = 87193600.0
+I1204 02:12:57.950700 137274321021824 utils.py:1231] [85150] progress = 0.7561965489374173
+I1204 02:12:57.950753 137274321021824 utils.py:1231] [85150] epoch = 68.05795029063346
+I1204 02:12:57.950818 137274321021824 utils.py:1231] [85150] img/sec/core = 164.2135904858142
+I1204 02:12:57.950896 137274321021824 utils.py:1231] [85150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.34549154739307
+I1204 02:12:57.950967 137274321021824 utils.py:1231] [85150] core_hours = 148.34549154739307
+I1204 02:12:57.951050 137274321021824 train.py:125] NOTE: Steps:85150/112603 [75.6%]
+Walltime:6d4h22m (0s eval)
+ETA:1d23h49m
+Total train time:8d4h10m
+I1204 02:18:08.548477 137274321021824 utils.py:1231] [85200] l2_params = 252.98948209358744
+I1204 02:18:08.548688 137274321021824 utils.py:1231] [85200] train/loss = 1.5769270956516266
+I1204 02:18:08.548788 137274321021824 utils.py:1231] [85200] l2_grads = 2.347120523452759
+I1204 02:18:08.548849 137274321021824 utils.py:1231] [85200] lr = 0.0001659264820050684
+I1204 02:18:08.548912 137274321021824 utils.py:1231] [85200] uptime = 534477.911273231
+I1204 02:18:08.548967 137274321021824 utils.py:1231] [85200] examples_seen = 87244800.0
+I1204 02:18:08.549018 137274321021824 utils.py:1231] [85200] progress = 0.7566405868404927
+I1204 02:18:08.549068 137274321021824 utils.py:1231] [85200] epoch = 68.09791385510242
+I1204 02:18:08.549122 137274321021824 utils.py:1231] [85200] img/sec/core = 164.84312494168591
+I1204 02:18:08.549186 137274321021824 utils.py:1231] [85200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.43176886242972
+I1204 02:18:08.549247 137274321021824 utils.py:1231] [85200] core_hours = 148.43176886242972
+I1204 02:18:08.549323 137274321021824 train.py:125] NOTE: Steps:85200/112603 [75.7%]
+Walltime:6d4h27m (0s eval)
+ETA:1d23h44m
+Total train time:8d4h10m
+I1204 02:23:20.337393 137274321021824 utils.py:1231] [85250] l2_params = 252.91903938803193
+I1204 02:23:20.337595 137274321021824 utils.py:1231] [85250] train/loss = 1.7827416807413101
+I1204 02:23:20.337706 137274321021824 utils.py:1231] [85250] l2_grads = 2.52492356300354
+I1204 02:23:20.337783 137274321021824 utils.py:1231] [85250] lr = 0.00016535733949004142
+I1204 02:23:20.337839 137274321021824 utils.py:1231] [85250] uptime = 534789.700200599
+I1204 02:23:20.337897 137274321021824 utils.py:1231] [85250] examples_seen = 87296000.0
+I1204 02:23:20.337950 137274321021824 utils.py:1231] [85250] progress = 0.7570846247435681
+I1204 02:23:20.337998 137274321021824 utils.py:1231] [85250] epoch = 68.13787741957138
+I1204 02:23:20.338049 137274321021824 utils.py:1231] [85250] img/sec/core = 164.21365707951708
+I1204 02:23:20.338104 137274321021824 utils.py:1231] [85250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.5183768978097
+I1204 02:23:20.338155 137274321021824 utils.py:1231] [85250] core_hours = 148.5183768978097
+I1204 02:23:20.338217 137274321021824 train.py:125] NOTE: Steps:85250/112603 [75.7%]
+Walltime:6d4h33m (0s eval)
+ETA:1d23h39m
+Total train time:8d4h10m
+I1204 02:28:30.324763 137274321021824 utils.py:1231] [85300] l2_params = 252.8566587085713
+I1204 02:28:30.325015 137274321021824 utils.py:1231] [85300] train/loss = 2.444569617509842
+I1204 02:28:30.325139 137274321021824 utils.py:1231] [85300] l2_grads = 2.113132953643799
+I1204 02:28:30.325218 137274321021824 utils.py:1231] [85300] lr = 0.00016478898130867754
+I1204 02:28:30.325311 137274321021824 utils.py:1231] [85300] uptime = 535099.687668082
+I1204 02:28:30.325386 137274321021824 utils.py:1231] [85300] examples_seen = 87347200.0
+I1204 02:28:30.325450 137274321021824 utils.py:1231] [85300] progress = 0.7575286626466435
+I1204 02:28:30.325504 137274321021824 utils.py:1231] [85300] epoch = 68.17784098404033
+I1204 02:28:30.325562 137274321021824 utils.py:1231] [85300] img/sec/core = 165.16796764637348
+I1204 02:28:30.325626 137274321021824 utils.py:1231] [85300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.6044845276661
+I1204 02:28:30.325691 137274321021824 utils.py:1231] [85300] core_hours = 148.6044845276661
+I1204 02:28:30.325771 137274321021824 train.py:125] NOTE: Steps:85300/112603 [75.8%]
+Walltime:6d4h38m (0s eval)
+ETA:1d23h34m
+Total train time:8d4h10m
+I1204 02:33:39.893419 137274321021824 utils.py:1231] [85350] l2_params = 252.79124235742472
+I1204 02:33:39.893690 137274321021824 utils.py:1231] [85350] train/loss = 1.7290787547826767
+I1204 02:33:39.893816 137274321021824 utils.py:1231] [85350] l2_grads = 2.255868434906006
+I1204 02:33:39.893908 137274321021824 utils.py:1231] [85350] lr = 0.0001642214087930918
+I1204 02:33:39.893972 137274321021824 utils.py:1231] [85350] uptime = 535409.256330633
+I1204 02:33:39.894031 137274321021824 utils.py:1231] [85350] examples_seen = 87398400.0
+I1204 02:33:39.894088 137274321021824 utils.py:1231] [85350] progress = 0.7579727005497189
+I1204 02:33:39.894144 137274321021824 utils.py:1231] [85350] epoch = 68.21780454850929
+I1204 02:33:39.894201 137274321021824 utils.py:1231] [85350] img/sec/core = 165.3914177813952
+I1204 02:33:39.894260 137274321021824 utils.py:1231] [85350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.69047582281917
+I1204 02:33:39.894315 137274321021824 utils.py:1231] [85350] core_hours = 148.69047582281917
+I1204 02:33:39.894380 137274321021824 train.py:125] NOTE: Steps:85350/112603 [75.8%]
+Walltime:6d4h43m (0s eval)
+ETA:1d23h28m
+Total train time:8d4h10m
+I1204 02:38:49.849637 137274321021824 utils.py:1231] [85400] l2_params = 252.72499270563193
+I1204 02:38:49.849864 137274321021824 utils.py:1231] [85400] train/loss = 4.180846214294434
+I1204 02:38:49.850015 137274321021824 utils.py:1231] [85400] l2_grads = 2.3329532146453857
+I1204 02:38:49.850091 137274321021824 utils.py:1231] [85400] lr = 0.00016365462327355769
+I1204 02:38:49.850153 137274321021824 utils.py:1231] [85400] uptime = 535719.21251583
+I1204 02:38:49.850213 137274321021824 utils.py:1231] [85400] examples_seen = 87449600.0
+I1204 02:38:49.850263 137274321021824 utils.py:1231] [85400] progress = 0.7584167384527943
+I1204 02:38:49.850317 137274321021824 utils.py:1231] [85400] epoch = 68.25776811297824
+I1204 02:38:49.850373 137274321021824 utils.py:1231] [85400] img/sec/core = 165.18463720109327
+I1204 02:38:49.850431 137274321021824 utils.py:1231] [85400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.77657476315164
+I1204 02:38:49.850494 137274321021824 utils.py:1231] [85400] core_hours = 148.77657476315164
+I1204 02:38:49.850566 137274321021824 train.py:125] NOTE: Steps:85400/112603 [75.8%]
+Walltime:6d4h48m (0s eval)
+ETA:1d23h23m
+Total train time:8d4h10m
+I1204 02:44:01.638200 137274321021824 utils.py:1231] [85450] l2_params = 252.66298805917307
+I1204 02:44:01.638426 137274321021824 utils.py:1231] [85450] train/loss = 2.0946085900068283
+I1204 02:44:01.638562 137274321021824 utils.py:1231] [85450] l2_grads = 2.2014923095703125
+I1204 02:44:01.638680 137274321021824 utils.py:1231] [85450] lr = 0.0001630886260785036
+I1204 02:44:01.638738 137274321021824 utils.py:1231] [85450] uptime = 536031.001100612
+I1204 02:44:01.638789 137274321021824 utils.py:1231] [85450] examples_seen = 87500800.0
+I1204 02:44:01.638837 137274321021824 utils.py:1231] [85450] progress = 0.7588607763558697
+I1204 02:44:01.638890 137274321021824 utils.py:1231] [85450] epoch = 68.2977316774472
+I1204 02:44:01.638943 137274321021824 utils.py:1231] [85450] img/sec/core = 164.21383751361992
+I1204 02:44:01.638996 137274321021824 utils.py:1231] [85450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.86318270336886
+I1204 02:44:01.639045 137274321021824 utils.py:1231] [85450] core_hours = 148.86318270336886
+I1204 02:44:01.639102 137274321021824 train.py:125] NOTE: Steps:85450/112603 [75.9%]
+Walltime:6d4h53m (0s eval)
+ETA:1d23h18m
+Total train time:8d4h10m
+I1204 02:49:11.261644 137274321021824 utils.py:1231] [85500] l2_params = 252.59530311325503
+I1204 02:49:11.261856 137274321021824 utils.py:1231] [85500] train/loss = 4.1394259333610535
+I1204 02:49:11.261970 137274321021824 utils.py:1231] [85500] l2_grads = 2.356121063232422
+I1204 02:49:11.262040 137274321021824 utils.py:1231] [85500] lr = 0.0001625234185345109
+I1204 02:49:11.262099 137274321021824 utils.py:1231] [85500] uptime = 536340.624460402
+I1204 02:49:11.262158 137274321021824 utils.py:1231] [85500] examples_seen = 87552000.0
+I1204 02:49:11.262221 137274321021824 utils.py:1231] [85500] progress = 0.7593048142589451
+I1204 02:49:11.262283 137274321021824 utils.py:1231] [85500] epoch = 68.33769524191617
+I1204 02:49:11.262346 137274321021824 utils.py:1231] [85500] img/sec/core = 165.36220017357033
+I1204 02:49:11.262409 137274321021824 utils.py:1231] [85500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 148.9491891921994
+I1204 02:49:11.262466 137274321021824 utils.py:1231] [85500] core_hours = 148.9491891921994
+I1204 02:49:11.262538 137274321021824 train.py:125] NOTE: Steps:85500/112603 [75.9%]
+Walltime:6d4h59m (0s eval)
+ETA:1d23h13m
+Total train time:8d4h10m
+I1204 02:54:23.050189 137274321021824 utils.py:1231] [85550] l2_params = 252.52691053675264
+I1204 02:54:23.050444 137274321021824 utils.py:1231] [85550] train/loss = 1.7937484234571457
+I1204 02:54:23.050560 137274321021824 utils.py:1231] [85550] l2_grads = 2.182305097579956
+I1204 02:54:23.050645 137274321021824 utils.py:1231] [85550] lr = 0.00016195900196631013
+I1204 02:54:23.050713 137274321021824 utils.py:1231] [85550] uptime = 536652.413073265
+I1204 02:54:23.050768 137274321021824 utils.py:1231] [85550] examples_seen = 87603200.0
+I1204 02:54:23.050816 137274321021824 utils.py:1231] [85550] progress = 0.7597488521620206
+I1204 02:54:23.050863 137274321021824 utils.py:1231] [85550] epoch = 68.37765880638511
+I1204 02:54:23.050919 137274321021824 utils.py:1231] [85550] img/sec/core = 164.21382272382493
+I1204 02:54:23.050976 137274321021824 utils.py:1231] [85550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.03579714021694
+I1204 02:54:23.051025 137274321021824 utils.py:1231] [85550] core_hours = 149.03579714021694
+I1204 02:54:23.051085 137274321021824 train.py:125] NOTE: Steps:85550/112603 [76.0%]
+Walltime:6d5h4m (0s eval)
+ETA:1d23h7m
+Total train time:8d4h10m
+I1204 02:59:34.846405 137274321021824 utils.py:1231] [85600] l2_params = 252.46577178939856
+I1204 02:59:34.846611 137274321021824 utils.py:1231] [85600] train/loss = 3.4586923718452454
+I1204 02:59:34.846713 137274321021824 utils.py:1231] [85600] l2_grads = 2.1542582511901855
+I1204 02:59:34.846783 137274321021824 utils.py:1231] [85600] lr = 0.00016139537769677744
+I1204 02:59:34.846841 137274321021824 utils.py:1231] [85600] uptime = 536964.209202724
+I1204 02:59:34.846911 137274321021824 utils.py:1231] [85600] examples_seen = 87654400.0
+I1204 02:59:34.846967 137274321021824 utils.py:1231] [85600] progress = 0.7601928900650959
+I1204 02:59:34.847024 137274321021824 utils.py:1231] [85600] epoch = 68.41762237085408
+I1204 02:59:34.847087 137274321021824 utils.py:1231] [85600] img/sec/core = 164.2098639544673
+I1204 02:59:34.847157 137274321021824 utils.py:1231] [85600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.12240717617777
+I1204 02:59:34.847212 137274321021824 utils.py:1231] [85600] core_hours = 149.12240717617777
+I1204 02:59:34.847277 137274321021824 train.py:125] NOTE: Steps:85600/112603 [76.0%]
+Walltime:6d5h9m (0s eval)
+ETA:1d23h2m
+Total train time:8d4h10m
+I1204 03:04:46.730586 137274321021824 utils.py:1231] [85650] l2_params = 252.40017677681047
+I1204 03:04:46.730803 137274321021824 utils.py:1231] [85650] train/loss = 3.5522421300411224
+I1204 03:04:46.730918 137274321021824 utils.py:1231] [85650] l2_grads = 2.116147994995117
+I1204 03:04:46.730993 137274321021824 utils.py:1231] [85650] lr = 0.00016083254704693273
+I1204 03:04:46.731053 137274321021824 utils.py:1231] [85650] uptime = 537276.093414225
+I1204 03:04:46.731112 137274321021824 utils.py:1231] [85650] examples_seen = 87705600.0
+I1204 03:04:46.731170 137274321021824 utils.py:1231] [85650] progress = 0.7606369279681714
+I1204 03:04:46.731225 137274321021824 utils.py:1231] [85650] epoch = 68.45758593532302
+I1204 03:04:46.731282 137274321021824 utils.py:1231] [85650] img/sec/core = 164.1634879611001
+I1204 03:04:46.731344 137274321021824 utils.py:1231] [85650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.20904167937252
+I1204 03:04:46.731401 137274321021824 utils.py:1231] [85650] core_hours = 149.20904167937252
+I1204 03:04:46.731468 137274321021824 train.py:125] NOTE: Steps:85650/112603 [76.1%]
+Walltime:6d5h14m (0s eval)
+ETA:1d22h57m
+Total train time:8d4h10m
+I1204 03:09:58.520324 137274321021824 utils.py:1231] [85700] l2_params = 252.33689610683524
+I1204 03:09:58.520563 137274321021824 utils.py:1231] [85700] train/loss = 3.6064737737178802
+I1204 03:09:58.520678 137274321021824 utils.py:1231] [85700] l2_grads = 2.081667423248291
+I1204 03:09:58.520760 137274321021824 utils.py:1231] [85700] lr = 0.00016027051133593525
+I1204 03:09:58.520827 137274321021824 utils.py:1231] [85700] uptime = 537587.883189282
+I1204 03:09:58.520890 137274321021824 utils.py:1231] [85700] examples_seen = 87756800.0
+I1204 03:09:58.520947 137274321021824 utils.py:1231] [85700] progress = 0.7610809658712467
+I1204 03:09:58.521000 137274321021824 utils.py:1231] [85700] epoch = 68.49754949979199
+I1204 03:09:58.521058 137274321021824 utils.py:1231] [85700] img/sec/core = 164.21321061808484
+I1204 03:09:58.521124 137274321021824 utils.py:1231] [85700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.29564995022167
+I1204 03:09:58.521177 137274321021824 utils.py:1231] [85700] core_hours = 149.29564995022167
+I1204 03:09:58.521249 137274321021824 train.py:125] NOTE: Steps:85700/112603 [76.1%]
+Walltime:6d5h19m (0s eval)
+ETA:1d22h52m
+Total train time:8d4h10m
+I1204 03:15:10.300081 137274321021824 utils.py:1231] [85750] l2_params = 252.2689494202499
+I1204 03:15:10.300374 137274321021824 utils.py:1231] [85750] train/loss = 2.3772884011268616
+I1204 03:15:10.300555 137274321021824 utils.py:1231] [85750] l2_grads = 2.0495901107788086
+I1204 03:15:10.300632 137274321021824 utils.py:1231] [85750] lr = 0.00015970927188108127
+I1204 03:15:10.300698 137274321021824 utils.py:1231] [85750] uptime = 537899.663059895
+I1204 03:15:10.300763 137274321021824 utils.py:1231] [85750] examples_seen = 87808000.0
+I1204 03:15:10.300825 137274321021824 utils.py:1231] [85750] progress = 0.7615250037743222
+I1204 03:15:10.300892 137274321021824 utils.py:1231] [85750] epoch = 68.53751306426095
+I1204 03:15:10.300966 137274321021824 utils.py:1231] [85750] img/sec/core = 164.21842724911028
+I1204 03:15:10.301022 137274321021824 utils.py:1231] [85750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.3822554698364
+I1204 03:15:10.301073 137274321021824 utils.py:1231] [85750] core_hours = 149.3822554698364
+I1204 03:15:10.301134 137274321021824 train.py:125] NOTE: Steps:85750/112603 [76.2%]
+Walltime:6d5h24m (0s eval)
+ETA:1d22h46m
+Total train time:8d4h9m
+I1204 03:20:22.082575 137274321021824 utils.py:1231] [85800] l2_params = 252.20371971109884
+I1204 03:20:22.082785 137274321021824 utils.py:1231] [85800] train/loss = 1.8767012655735016
+I1204 03:20:22.082880 137274321021824 utils.py:1231] [85800] l2_grads = 2.3056294918060303
+I1204 03:20:22.082948 137274321021824 utils.py:1231] [85800] lr = 0.000159148829997801
+I1204 03:20:22.083000 137274321021824 utils.py:1231] [85800] uptime = 538211.445361677
+I1204 03:20:22.083052 137274321021824 utils.py:1231] [85800] examples_seen = 87859200.0
+I1204 03:20:22.083101 137274321021824 utils.py:1231] [85800] progress = 0.7619690416773975
+I1204 03:20:22.083151 137274321021824 utils.py:1231] [85800] epoch = 68.5774766287299
+I1204 03:20:22.083201 137274321021824 utils.py:1231] [85800] img/sec/core = 164.2171467314733
+I1204 03:20:22.083258 137274321021824 utils.py:1231] [85800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.46886166477583
+I1204 03:20:22.083309 137274321021824 utils.py:1231] [85800] core_hours = 149.46886166477583
+I1204 03:20:22.083387 137274321021824 train.py:125] NOTE: Steps:85800/112603 [76.2%]
+Walltime:6d5h30m (0s eval)
+ETA:1d22h41m
+Total train time:8d4h9m
+I1204 03:25:33.859205 137274321021824 utils.py:1231] [85850] l2_params = 252.14193582533872
+I1204 03:25:33.859432 137274321021824 utils.py:1231] [85850] train/loss = 1.8556331992149353
+I1204 03:25:33.859575 137274321021824 utils.py:1231] [85850] l2_grads = 2.2384192943573
+I1204 03:25:33.859689 137274321021824 utils.py:1231] [85850] lr = 0.00015858918699965466
+I1204 03:25:33.859787 137274321021824 utils.py:1231] [85850] uptime = 538523.222143831
+I1204 03:25:33.859872 137274321021824 utils.py:1231] [85850] examples_seen = 87910400.0
+I1204 03:25:33.859958 137274321021824 utils.py:1231] [85850] progress = 0.762413079580473
+I1204 03:25:33.860033 137274321021824 utils.py:1231] [85850] epoch = 68.61744019319886
+I1204 03:25:33.860106 137274321021824 utils.py:1231] [85850] img/sec/core = 164.22005399591265
+I1204 03:25:33.860183 137274321021824 utils.py:1231] [85850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.55546632648526
+I1204 03:25:33.860263 137274321021824 utils.py:1231] [85850] core_hours = 149.55546632648526
+I1204 03:25:33.860356 137274321021824 train.py:125] NOTE: Steps:85850/112603 [76.2%]
+Walltime:6d5h35m (0s eval)
+ETA:1d22h36m
+Total train time:8d4h9m
+I1204 03:30:44.193019 137274321021824 utils.py:1231] [85900] l2_params = 252.07717571246502
+I1204 03:30:44.193242 137274321021824 utils.py:1231] [85900] train/loss = 1.7712940275669098
+I1204 03:30:44.193359 137274321021824 utils.py:1231] [85900] l2_grads = 2.16333270072937
+I1204 03:30:44.193445 137274321021824 utils.py:1231] [85900] lr = 0.00015803034419833105
+I1204 03:30:44.193526 137274321021824 utils.py:1231] [85900] uptime = 538833.555887028
+I1204 03:30:44.193584 137274321021824 utils.py:1231] [85900] examples_seen = 87961600.0
+I1204 03:30:44.193639 137274321021824 utils.py:1231] [85900] progress = 0.7628571174835483
+I1204 03:30:44.193693 137274321021824 utils.py:1231] [85900] epoch = 68.65740375766781
+I1204 03:30:44.193748 137274321021824 utils.py:1231] [85900] img/sec/core = 164.98367039479012
+I1204 03:30:44.193807 137274321021824 utils.py:1231] [85900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.64167014403998
+I1204 03:30:44.193862 137274321021824 utils.py:1231] [85900] core_hours = 149.64167014403998
+I1204 03:30:44.193932 137274321021824 train.py:125] NOTE: Steps:85900/112603 [76.3%]
+Walltime:6d5h40m (0s eval)
+ETA:1d22h31m
+Total train time:8d4h9m
+I1204 03:35:55.975671 137274321021824 utils.py:1231] [85950] l2_params = 252.01217130874483
+I1204 03:35:55.975898 137274321021824 utils.py:1231] [85950] train/loss = 1.7335197031497955
+I1204 03:35:55.976052 137274321021824 utils.py:1231] [85950] l2_grads = 2.348757028579712
+I1204 03:35:55.976133 137274321021824 utils.py:1231] [85950] lr = 0.00015747230290364233
+I1204 03:35:55.976194 137274321021824 utils.py:1231] [85950] uptime = 539145.338555805
+I1204 03:35:55.976255 137274321021824 utils.py:1231] [85950] examples_seen = 88012800.0
+I1204 03:35:55.976313 137274321021824 utils.py:1231] [85950] progress = 0.7633011553866238
+I1204 03:35:55.976369 137274321021824 utils.py:1231] [85950] epoch = 68.69736732213677
+I1204 03:35:55.976435 137274321021824 utils.py:1231] [85950] img/sec/core = 164.21695343371803
+I1204 03:35:55.976498 137274321021824 utils.py:1231] [85950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.7282764409225
+I1204 03:35:55.976555 137274321021824 utils.py:1231] [85950] core_hours = 149.7282764409225
+I1204 03:35:55.976620 137274321021824 train.py:125] NOTE: Steps:85950/112603 [76.3%]
+Walltime:6d5h45m (0s eval)
+ETA:1d22h25m
+Total train time:8d4h9m
+I1204 03:41:01.346907 137274321021824 utils.py:1231] [86000] l2_params = 251.94500398095025
+I1204 03:41:01.347187 137274321021824 utils.py:1231] [86000] train/loss = 1.8144307434558868
+I1204 03:41:01.347300 137274321021824 utils.py:1231] [86000] l2_grads = 2.417649507522583
+I1204 03:41:01.347382 137274321021824 utils.py:1231] [86000] lr = 0.00015691506442352311
+I1204 03:41:01.347433 137274321021824 utils.py:1231] [86000] uptime = 539450.709795297
+I1204 03:41:01.347491 137274321021824 utils.py:1231] [86000] examples_seen = 88064000.0
+I1204 03:41:01.347540 137274321021824 utils.py:1231] [86000] progress = 0.7637451932896993
+I1204 03:41:01.347587 137274321021824 utils.py:1231] [86000] epoch = 68.73733088660573
+I1204 03:41:01.347636 137274321021824 utils.py:1231] [86000] img/sec/core = 167.66477447312502
+I1204 03:41:01.347709 137274321021824 utils.py:1231] [86000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.81310178522583
+I1204 03:41:01.347772 137274321021824 utils.py:1231] [86000] core_hours = 149.81310178522583
+I1204 03:41:01.347836 137274321021824 train.py:125] NOTE: Steps:86000/112603 [76.4%]
+Walltime:6d5h50m (0s eval)
+ETA:1d22h20m
+Total train time:8d4h9m
+I1204 03:46:11.313666 137274321021824 utils.py:1231] [86050] l2_params = 251.88024777596638
+I1204 03:46:11.313875 137274321021824 utils.py:1231] [86050] train/loss = 2.5946326553821564
+I1204 03:46:11.314000 137274321021824 utils.py:1231] [86050] l2_grads = 2.085084915161133
+I1204 03:46:11.314084 137274321021824 utils.py:1231] [86050] lr = 0.00015635863006402596
+I1204 03:46:11.314148 137274321021824 utils.py:1231] [86050] uptime = 539760.676508463
+I1204 03:46:11.314219 137274321021824 utils.py:1231] [86050] examples_seen = 88115200.0
+I1204 03:46:11.314282 137274321021824 utils.py:1231] [86050] progress = 0.7641892311927746
+I1204 03:46:11.314344 137274321021824 utils.py:1231] [86050] epoch = 68.77729445107468
+I1204 03:46:11.314408 137274321021824 utils.py:1231] [86050] img/sec/core = 165.17902673176218
+I1204 03:46:11.314490 137274321021824 utils.py:1231] [86050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.89920364999415
+I1204 03:46:11.314567 137274321021824 utils.py:1231] [86050] core_hours = 149.89920364999415
+I1204 03:46:11.314644 137274321021824 train.py:125] NOTE: Steps:86050/112603 [76.4%]
+Walltime:6d5h56m (0s eval)
+ETA:1d22h15m
+Total train time:8d4h9m
+I1204 03:51:18.935105 137274321021824 utils.py:1231] [86100] l2_params = 251.82237763759724
+I1204 03:51:18.935312 137274321021824 utils.py:1231] [86100] train/loss = 1.6714541614055634
+I1204 03:51:18.935409 137274321021824 utils.py:1231] [86100] l2_grads = 2.1894001960754395
+I1204 03:51:18.935471 137274321021824 utils.py:1231] [86100] lr = 0.00015580300112931839
+I1204 03:51:18.935522 137274321021824 utils.py:1231] [86100] uptime = 540068.297885058
+I1204 03:51:18.935583 137274321021824 utils.py:1231] [86100] examples_seen = 88166400.0
+I1204 03:51:18.935633 137274321021824 utils.py:1231] [86100] progress = 0.76463326909585
+I1204 03:51:18.935679 137274321021824 utils.py:1231] [86100] epoch = 68.81725801554364
+I1204 03:51:18.935728 137274321021824 utils.py:1231] [86100] img/sec/core = 166.43836838233938
+I1204 03:51:18.935781 137274321021824 utils.py:1231] [86100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 149.98465403238166
+I1204 03:51:18.935829 137274321021824 utils.py:1231] [86100] core_hours = 149.98465403238166
+I1204 03:51:18.935891 137274321021824 train.py:125] NOTE: Steps:86100/112603 [76.5%]
+Walltime:6d6h1m (0s eval)
+ETA:1d22h10m
+Total train time:8d4h9m
+I1204 03:56:27.806090 137274321021824 utils.py:1231] [86150] l2_params = 251.75759903832878
+I1204 03:56:27.806366 137274321021824 utils.py:1231] [86150] train/loss = 2.545050859451294
+I1204 03:56:27.806494 137274321021824 utils.py:1231] [86150] l2_grads = 2.112201690673828
+I1204 03:56:27.806575 137274321021824 utils.py:1231] [86150] lr = 0.00015524817892168094
+I1204 03:56:27.806650 137274321021824 utils.py:1231] [86150] uptime = 540377.169007719
+I1204 03:56:27.806723 137274321021824 utils.py:1231] [86150] examples_seen = 88217600.0
+I1204 03:56:27.806790 137274321021824 utils.py:1231] [86150] progress = 0.7650773069989254
+I1204 03:56:27.806864 137274321021824 utils.py:1231] [86150] epoch = 68.85722158001259
+I1204 03:56:27.806929 137274321021824 utils.py:1231] [86150] img/sec/core = 165.76492991281214
+I1204 03:56:27.807006 137274321021824 utils.py:1231] [86150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.07045156645418
+I1204 03:56:27.807060 137274321021824 utils.py:1231] [86150] core_hours = 150.07045156645418
+I1204 03:56:27.807126 137274321021824 train.py:125] NOTE: Steps:86150/112603 [76.5%]
+Walltime:6d6h6m (0s eval)
+ETA:1d22h4m
+Total train time:8d4h9m
+I1204 04:01:39.611678 137274321021824 utils.py:1231] [86200] l2_params = 251.69351076314268
+I1204 04:01:39.611885 137274321021824 utils.py:1231] [86200] train/loss = 1.7783105075359344
+I1204 04:01:39.611996 137274321021824 utils.py:1231] [86200] l2_grads = 2.351198673248291
+I1204 04:01:39.612094 137274321021824 utils.py:1231] [86200] lr = 0.0001546941647415024
+I1204 04:01:39.612187 137274321021824 utils.py:1231] [86200] uptime = 540688.974540807
+I1204 04:01:39.612270 137274321021824 utils.py:1231] [86200] examples_seen = 88268800.0
+I1204 04:01:39.612339 137274321021824 utils.py:1231] [86200] progress = 0.7655213449020009
+I1204 04:01:39.612403 137274321021824 utils.py:1231] [86200] epoch = 68.89718514448155
+I1204 04:01:39.612478 137274321021824 utils.py:1231] [86200] img/sec/core = 164.2049116093192
+I1204 04:01:39.612565 137274321021824 utils.py:1231] [86200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.15706421453416
+I1204 04:01:39.612655 137274321021824 utils.py:1231] [86200] core_hours = 150.15706421453416
+I1204 04:01:39.612751 137274321021824 train.py:125] NOTE: Steps:86200/112603 [76.6%]
+Walltime:6d6h11m (0s eval)
+ETA:1d21h59m
+Total train time:8d4h9m
+I1204 04:06:49.234718 137274321021824 utils.py:1231] [86250] l2_params = 251.63029322988186
+I1204 04:06:49.234949 137274321021824 utils.py:1231] [86250] train/loss = 1.892592191696167
+I1204 04:06:49.235107 137274321021824 utils.py:1231] [86250] l2_grads = 2.281047821044922
+I1204 04:06:49.235176 137274321021824 utils.py:1231] [86250] lr = 0.00015414095988727875
+I1204 04:06:49.235227 137274321021824 utils.py:1231] [86250] uptime = 540998.597589288
+I1204 04:06:49.235279 137274321021824 utils.py:1231] [86250] examples_seen = 88320000.0
+I1204 04:06:49.235328 137274321021824 utils.py:1231] [86250] progress = 0.7659653828050762
+I1204 04:06:49.235390 137274321021824 utils.py:1231] [86250] epoch = 68.93714870895052
+I1204 04:06:49.235446 137274321021824 utils.py:1231] [86250] img/sec/core = 165.36236643613762
+I1204 04:06:49.235510 137274321021824 utils.py:1231] [86250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.24307061688998
+I1204 04:06:49.235565 137274321021824 utils.py:1231] [86250] core_hours = 150.24307061688998
+I1204 04:06:49.235624 137274321021824 train.py:125] NOTE: Steps:86250/112603 [76.6%]
+Walltime:6d6h16m (0s eval)
+ETA:1d21h54m
+Total train time:8d4h9m
+I1204 04:11:58.656474 137274321021824 utils.py:1231] [86300] l2_params = 251.56664158225507
+I1204 04:11:58.656723 137274321021824 utils.py:1231] [86300] train/loss = 2.461746484041214
+I1204 04:11:58.656841 137274321021824 utils.py:1231] [86300] l2_grads = 2.221390724182129
+I1204 04:11:58.656924 137274321021824 utils.py:1231] [86300] lr = 0.00015358856565560807
+I1204 04:11:58.656977 137274321021824 utils.py:1231] [86300] uptime = 541308.019338764
+I1204 04:11:58.657030 137274321021824 utils.py:1231] [86300] examples_seen = 88371200.0
+I1204 04:11:58.657080 137274321021824 utils.py:1231] [86300] progress = 0.7664094207081517
+I1204 04:11:58.657129 137274321021824 utils.py:1231] [86300] epoch = 68.97711227341946
+I1204 04:11:58.657180 137274321021824 utils.py:1231] [86300] img/sec/core = 165.46994542790614
+I1204 04:11:58.657237 137274321021824 utils.py:1231] [86300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.32902110285556
+I1204 04:11:58.657290 137274321021824 utils.py:1231] [86300] core_hours = 150.32902110285556
+I1204 04:11:58.657351 137274321021824 train.py:125] NOTE: Steps:86300/112603 [76.6%]
+Walltime:6d6h21m (0s eval)
+ETA:1d21h49m
+Total train time:8d4h9m
+I1204 04:17:07.814854 137274321021824 utils.py:1231] [86350] l2_params = 251.50591939641976
+I1204 04:17:07.815118 137274321021824 utils.py:1231] [86350] train/loss = 2.380596935749054
+I1204 04:17:07.815260 137274321021824 utils.py:1231] [86350] l2_grads = 2.117142915725708
+I1204 04:17:07.815351 137274321021824 utils.py:1231] [86350] lr = 0.00015303698334118919
+I1204 04:17:07.815417 137274321021824 utils.py:1231] [86350] uptime = 541617.177778163
+I1204 04:17:07.815493 137274321021824 utils.py:1231] [86350] examples_seen = 88422400.0
+I1204 04:17:07.815559 137274321021824 utils.py:1231] [86350] progress = 0.766853458611227
+I1204 04:17:07.815623 137274321021824 utils.py:1231] [86350] epoch = 69.01707583788843
+I1204 04:17:07.815681 137274321021824 utils.py:1231] [86350] img/sec/core = 165.61087609167683
+I1204 04:17:07.815747 137274321021824 utils.py:1231] [86350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.41489844713306
+I1204 04:17:07.815815 137274321021824 utils.py:1231] [86350] core_hours = 150.41489844713306
+I1204 04:17:07.815895 137274321021824 train.py:125] NOTE: Steps:86350/112603 [76.7%]
+Walltime:6d6h26m (0s eval)
+ETA:1d21h43m
+Total train time:8d4h9m
+I1204 04:22:18.700763 137274321021824 utils.py:1231] [86400] l2_params = 251.44519495313125
+I1204 04:22:18.700963 137274321021824 utils.py:1231] [86400] train/loss = 1.6492525488138199
+I1204 04:22:18.701069 137274321021824 utils.py:1231] [86400] l2_grads = 2.4987986087799072
+I1204 04:22:18.701133 137274321021824 utils.py:1231] [86400] lr = 0.00015248621423681803
+I1204 04:22:18.701185 137274321021824 utils.py:1231] [86400] uptime = 541928.063547445
+I1204 04:22:18.701240 137274321021824 utils.py:1231] [86400] examples_seen = 88473600.0
+I1204 04:22:18.701292 137274321021824 utils.py:1231] [86400] progress = 0.7672974965143025
+I1204 04:22:18.701341 137274321021824 utils.py:1231] [86400] epoch = 69.05703940235738
+I1204 04:22:18.701392 137274321021824 utils.py:1231] [86400] img/sec/core = 164.6907162017985
+I1204 04:22:18.701448 137274321021824 utils.py:1231] [86400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.50125560526695
+I1204 04:22:18.701500 137274321021824 utils.py:1231] [86400] core_hours = 150.50125560526695
+I1204 04:22:18.701561 137274321021824 train.py:125] NOTE: Steps:86400/112603 [76.7%]
+Walltime:6d6h32m (0s eval)
+ETA:1d21h38m
+Total train time:8d4h8m
+I1204 04:27:29.190030 137274321021824 utils.py:1231] [86450] l2_params = 251.3843718259668
+I1204 04:27:29.190262 137274321021824 utils.py:1231] [86450] train/loss = 1.7539092600345612
+I1204 04:27:29.190386 137274321021824 utils.py:1231] [86450] l2_grads = 2.361265182495117
+I1204 04:27:29.190463 137274321021824 utils.py:1231] [86450] lr = 0.00015193625963338354
+I1204 04:27:29.190520 137274321021824 utils.py:1231] [86450] uptime = 542238.552879566
+I1204 04:27:29.190574 137274321021824 utils.py:1231] [86450] examples_seen = 88524800.0
+I1204 04:27:29.190623 137274321021824 utils.py:1231] [86450] progress = 0.7677415344173779
+I1204 04:27:29.190670 137274321021824 utils.py:1231] [86450] epoch = 69.09700296682634
+I1204 04:27:29.190719 137274321021824 utils.py:1231] [86450] img/sec/core = 164.90099563243592
+I1204 04:27:29.190774 137274321021824 utils.py:1231] [86450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.5875026419672
+I1204 04:27:29.190833 137274321021824 utils.py:1231] [86450] core_hours = 150.5875026419672
+I1204 04:27:29.190905 137274321021824 train.py:125] NOTE: Steps:86450/112603 [76.8%]
+Walltime:6d6h37m (0s eval)
+ETA:1d21h33m
+Total train time:8d4h8m
+I1204 04:32:38.910171 137274321021824 utils.py:1231] [86500] l2_params = 251.31989149849136
+I1204 04:32:38.910377 137274321021824 utils.py:1231] [86500] train/loss = 1.7813011854887009
+I1204 04:32:38.910476 137274321021824 utils.py:1231] [86500] l2_grads = 2.313161611557007
+I1204 04:32:38.910535 137274321021824 utils.py:1231] [86500] lr = 0.00015138712081986706
+I1204 04:32:38.910586 137274321021824 utils.py:1231] [86500] uptime = 542548.272948152
+I1204 04:32:38.910637 137274321021824 utils.py:1231] [86500] examples_seen = 88576000.0
+I1204 04:32:38.910686 137274321021824 utils.py:1231] [86500] progress = 0.7681855723204533
+I1204 04:32:38.910735 137274321021824 utils.py:1231] [86500] epoch = 69.1369665312953
+I1204 04:32:38.910784 137274321021824 utils.py:1231] [86500] img/sec/core = 165.3105665181446
+I1204 04:32:38.910841 137274321021824 utils.py:1231] [86500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.67353599435222
+I1204 04:32:38.910896 137274321021824 utils.py:1231] [86500] core_hours = 150.67353599435222
+I1204 04:32:38.910968 137274321021824 train.py:125] NOTE: Steps:86500/112603 [76.8%]
+Walltime:6d6h42m (0s eval)
+ETA:1d21h28m
+Total train time:8d4h8m
+I1204 04:37:50.680003 137274321021824 utils.py:1231] [86550] l2_params = 251.25757871010643
+I1204 04:37:50.680198 137274321021824 utils.py:1231] [86550] train/loss = 1.7436784207820892
+I1204 04:37:50.680282 137274321021824 utils.py:1231] [86550] l2_grads = 2.4693031311035156
+I1204 04:37:50.680339 137274321021824 utils.py:1231] [86550] lr = 0.0001508387990833367
+I1204 04:37:50.680388 137274321021824 utils.py:1231] [86550] uptime = 542860.042751119
+I1204 04:37:50.680442 137274321021824 utils.py:1231] [86550] examples_seen = 88627200.0
+I1204 04:37:50.680489 137274321021824 utils.py:1231] [86550] progress = 0.7686296102235287
+I1204 04:37:50.680534 137274321021824 utils.py:1231] [86550] epoch = 69.17693009576425
+I1204 04:37:50.680580 137274321021824 utils.py:1231] [86550] img/sec/core = 164.22373017769237
+I1204 04:37:50.680633 137274321021824 utils.py:1231] [86550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.7601387173986
+I1204 04:37:50.680683 137274321021824 utils.py:1231] [86550] core_hours = 150.7601387173986
+I1204 04:37:50.680740 137274321021824 train.py:125] NOTE: Steps:86550/112603 [76.9%]
+Walltime:6d6h47m (0s eval)
+ETA:1d21h22m
+Total train time:8d4h8m
+I1204 04:43:02.465537 137274321021824 utils.py:1231] [86600] l2_params = 251.19831087888383
+I1204 04:43:02.465795 137274321021824 utils.py:1231] [86600] train/loss = 1.696321189403534
+I1204 04:43:02.465918 137274321021824 utils.py:1231] [86600] l2_grads = 2.2487263679504395
+I1204 04:43:02.465981 137274321021824 utils.py:1231] [86600] lr = 0.00015029129570894617
+I1204 04:43:02.466031 137274321021824 utils.py:1231] [86600] uptime = 543171.828393282
+I1204 04:43:02.466100 137274321021824 utils.py:1231] [86600] examples_seen = 88678400.0
+I1204 04:43:02.466149 137274321021824 utils.py:1231] [86600] progress = 0.7690736481266041
+I1204 04:43:02.466195 137274321021824 utils.py:1231] [86600] epoch = 69.21689366023321
+I1204 04:43:02.466245 137274321021824 utils.py:1231] [86600] img/sec/core = 164.2153873565183
+I1204 04:43:02.466300 137274321021824 utils.py:1231] [86600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.84674584022167
+I1204 04:43:02.466349 137274321021824 utils.py:1231] [86600] core_hours = 150.84674584022167
+I1204 04:43:02.466406 137274321021824 train.py:125] NOTE: Steps:86600/112603 [76.9%]
+Walltime:6d6h52m (0s eval)
+ETA:1d21h17m
+Total train time:8d4h8m
+I1204 04:48:11.711418 137274321021824 utils.py:1231] [86650] l2_params = 251.1352366630835
+I1204 04:48:11.711633 137274321021824 utils.py:1231] [86650] train/loss = 1.7794529348611832
+I1204 04:48:11.711736 137274321021824 utils.py:1231] [86650] l2_grads = 2.282832384109497
+I1204 04:48:11.711804 137274321021824 utils.py:1231] [86650] lr = 0.00014974461197993105
+I1204 04:48:11.711867 137274321021824 utils.py:1231] [86650] uptime = 543481.074228033
+I1204 04:48:11.711935 137274321021824 utils.py:1231] [86650] examples_seen = 88729600.0
+I1204 04:48:11.712003 137274321021824 utils.py:1231] [86650] progress = 0.7695176860296795
+I1204 04:48:11.712063 137274321021824 utils.py:1231] [86650] epoch = 69.25685722470216
+I1204 04:48:11.712121 137274321021824 utils.py:1231] [86650] img/sec/core = 165.56407313047782
+I1204 04:48:11.712184 137274321021824 utils.py:1231] [86650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 150.9326474609858
+I1204 04:48:11.712237 137274321021824 utils.py:1231] [86650] core_hours = 150.9326474609858
+I1204 04:48:11.712302 137274321021824 train.py:125] NOTE: Steps:86650/112603 [77.0%]
+Walltime:6d6h58m (0s eval)
+ETA:1d21h12m
+Total train time:8d4h8m
+I1204 04:53:20.096917 137274321021824 utils.py:1231] [86700] l2_params = 251.07195210653717
+I1204 04:53:20.097172 137274321021824 utils.py:1231] [86700] train/loss = 1.72162227332592
+I1204 04:53:20.097296 137274321021824 utils.py:1231] [86700] l2_grads = 2.2238221168518066
+I1204 04:53:20.097370 137274321021824 utils.py:1231] [86700] lr = 0.0001491987491776056
+I1204 04:53:20.097438 137274321021824 utils.py:1231] [86700] uptime = 543789.4597950369
+I1204 04:53:20.097495 137274321021824 utils.py:1231] [86700] examples_seen = 88780800.0
+I1204 04:53:20.097546 137274321021824 utils.py:1231] [86700] progress = 0.7699617239327549
+I1204 04:53:20.097603 137274321021824 utils.py:1231] [86700] epoch = 69.29682078917112
+I1204 04:53:20.097657 137274321021824 utils.py:1231] [86700] img/sec/core = 166.02592818275323
+I1204 04:53:20.097728 137274321021824 utils.py:1231] [86700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.01831011848694
+I1204 04:53:20.097783 137274321021824 utils.py:1231] [86700] core_hours = 151.01831011848694
+I1204 04:53:20.097846 137274321021824 train.py:125] NOTE: Steps:86700/112603 [77.0%]
+Walltime:6d7h3m (0s eval)
+ETA:1d21h7m
+Total train time:8d4h8m
+I1204 04:58:28.046384 137274321021824 utils.py:1231] [86750] l2_params = 251.0103036872693
+I1204 04:58:28.046576 137274321021824 utils.py:1231] [86750] train/loss = 2.0287273079156876
+I1204 04:58:28.046675 137274321021824 utils.py:1231] [86750] l2_grads = 2.1449410915374756
+I1204 04:58:28.046734 137274321021824 utils.py:1231] [86750] lr = 0.00014865370858136028
+I1204 04:58:28.046793 137274321021824 utils.py:1231] [86750] uptime = 544097.4091540549
+I1204 04:58:28.046852 137274321021824 utils.py:1231] [86750] examples_seen = 88832000.0
+I1204 04:58:28.046916 137274321021824 utils.py:1231] [86750] progress = 0.7704057618358303
+I1204 04:58:28.046968 137274321021824 utils.py:1231] [86750] epoch = 69.33678435364008
+I1204 04:58:28.047021 137274321021824 utils.py:1231] [86750] img/sec/core = 166.26110268021776
+I1204 04:58:28.047077 137274321021824 utils.py:1231] [86750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.10385160710302
+I1204 04:58:28.047129 137274321021824 utils.py:1231] [86750] core_hours = 151.10385160710302
+I1204 04:58:28.047191 137274321021824 train.py:125] NOTE: Steps:86750/112603 [77.0%]
+Walltime:6d7h8m (0s eval)
+ETA:1d21h1m
+Total train time:8d4h8m
+I1204 05:03:34.490995 137274321021824 utils.py:1231] [86800] l2_params = 250.94775399686318
+I1204 05:03:34.491246 137274321021824 utils.py:1231] [86800] train/loss = 3.8732395470142365
+I1204 05:03:34.491383 137274321021824 utils.py:1231] [86800] l2_grads = 2.290539264678955
+I1204 05:03:34.491469 137274321021824 utils.py:1231] [86800] lr = 0.00014810949146865787
+I1204 05:03:34.491537 137274321021824 utils.py:1231] [86800] uptime = 544403.853894496
+I1204 05:03:34.491622 137274321021824 utils.py:1231] [86800] examples_seen = 88883200.0
+I1204 05:03:34.491683 137274321021824 utils.py:1231] [86800] progress = 0.7708497997389057
+I1204 05:03:34.491757 137274321021824 utils.py:1231] [86800] epoch = 69.37674791810903
+I1204 05:03:34.491821 137274321021824 utils.py:1231] [86800] img/sec/core = 167.0774310771698
+I1204 05:03:34.491900 137274321021824 utils.py:1231] [86800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.18897514611444
+I1204 05:03:34.491957 137274321021824 utils.py:1231] [86800] core_hours = 151.18897514611444
+I1204 05:03:34.492025 137274321021824 train.py:125] NOTE: Steps:86800/112603 [77.1%]
+Walltime:6d7h13m (0s eval)
+ETA:1d20h56m
+Total train time:8d4h8m
+I1204 05:08:40.235569 137274321021824 utils.py:1231] [86850] l2_params = 250.89016032646592
+I1204 05:08:40.235832 137274321021824 utils.py:1231] [86850] train/loss = 2.471048027276993
+I1204 05:08:40.236014 137274321021824 utils.py:1231] [86850] l2_grads = 2.1765334606170654
+I1204 05:08:40.236108 137274321021824 utils.py:1231] [86850] lr = 0.00014756609911503195
+I1204 05:08:40.236185 137274321021824 utils.py:1231] [86850] uptime = 544709.598546012
+I1204 05:08:40.236259 137274321021824 utils.py:1231] [86850] examples_seen = 88934400.0
+I1204 05:08:40.236320 137274321021824 utils.py:1231] [86850] progress = 0.7712938376419811
+I1204 05:08:40.236381 137274321021824 utils.py:1231] [86850] epoch = 69.416711482578
+I1204 05:08:40.236438 137274321021824 utils.py:1231] [86850] img/sec/core = 167.4600021492967
+I1204 05:08:40.236507 137274321021824 utils.py:1231] [86850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.27390421598
+I1204 05:08:40.236571 137274321021824 utils.py:1231] [86850] core_hours = 151.27390421598
+I1204 05:08:40.236640 137274321021824 train.py:125] NOTE: Steps:86850/112603 [77.1%]
+Walltime:6d7h18m (0s eval)
+ETA:1d20h51m
+Total train time:8d4h8m
+I1204 05:13:48.442180 137274321021824 utils.py:1231] [86900] l2_params = 250.8245489974954
+I1204 05:13:48.442368 137274321021824 utils.py:1231] [86900] train/loss = 1.7558027803897858
+I1204 05:13:48.442456 137274321021824 utils.py:1231] [86900] l2_grads = 2.270256519317627
+I1204 05:13:48.442517 137274321021824 utils.py:1231] [86900] lr = 0.0001470235327940826
+I1204 05:13:48.442568 137274321021824 utils.py:1231] [86900] uptime = 545017.804930708
+I1204 05:13:48.442620 137274321021824 utils.py:1231] [86900] examples_seen = 88985600.0
+I1204 05:13:48.442671 137274321021824 utils.py:1231] [86900] progress = 0.7717378755450566
+I1204 05:13:48.442719 137274321021824 utils.py:1231] [86900] epoch = 69.45667504704696
+I1204 05:13:48.442769 137274321021824 utils.py:1231] [86900] img/sec/core = 166.12245087164627
+I1204 05:13:48.442826 137274321021824 utils.py:1231] [86900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.35951710061778
+I1204 05:13:48.442876 137274321021824 utils.py:1231] [86900] core_hours = 151.35951710061778
+I1204 05:13:48.442949 137274321021824 train.py:125] NOTE: Steps:86900/112603 [77.2%]
+Walltime:6d7h23m (0s eval)
+ETA:1d20h46m
+Total train time:8d4h7m
+I1204 05:18:56.456706 137274321021824 utils.py:1231] [86950] l2_params = 250.7649746412864
+I1204 05:18:56.456924 137274321021824 utils.py:1231] [86950] train/loss = 1.9105672985315323
+I1204 05:18:56.457030 137274321021824 utils.py:1231] [86950] l2_grads = 2.2825610637664795
+I1204 05:18:56.457107 137274321021824 utils.py:1231] [86950] lr = 0.0001464817937774736
+I1204 05:18:56.457167 137274321021824 utils.py:1231] [86950] uptime = 545325.81952846
+I1204 05:18:56.457228 137274321021824 utils.py:1231] [86950] examples_seen = 89036800.0
+I1204 05:18:56.457292 137274321021824 utils.py:1231] [86950] progress = 0.7721819134481319
+I1204 05:18:56.457349 137274321021824 utils.py:1231] [86950] epoch = 69.4966386115159
+I1204 05:18:56.457406 137274321021824 utils.py:1231] [86950] img/sec/core = 166.2258879081488
+I1204 05:18:56.457469 137274321021824 utils.py:1231] [86950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.44507671110443
+I1204 05:18:56.457525 137274321021824 utils.py:1231] [86950] core_hours = 151.44507671110443
+I1204 05:18:56.457591 137274321021824 train.py:125] NOTE: Steps:86950/112603 [77.2%]
+Walltime:6d7h28m (0s eval)
+ETA:1d20h40m
+Total train time:8d4h7m
+I1204 05:24:04.138976 137274321021824 utils.py:1231] [87000] l2_params = 250.7012509936615
+I1204 05:24:04.139193 137274321021824 utils.py:1231] [87000] train/loss = 1.812287598848343
+I1204 05:24:04.139296 137274321021824 utils.py:1231] [87000] l2_grads = 2.2895920276641846
+I1204 05:24:04.139367 137274321021824 utils.py:1231] [87000] lr = 0.00014594088333492984
+I1204 05:24:04.139449 137274321021824 utils.py:1231] [87000] uptime = 545633.501802196
+I1204 05:24:04.139518 137274321021824 utils.py:1231] [87000] examples_seen = 89088000.0
+I1204 05:24:04.139585 137274321021824 utils.py:1231] [87000] progress = 0.7726259513512074
+I1204 05:24:04.139643 137274321021824 utils.py:1231] [87000] epoch = 69.53660217598487
+I1204 05:24:04.139714 137274321021824 utils.py:1231] [87000] img/sec/core = 166.40542654053763
+I1204 05:24:04.139782 137274321021824 utils.py:1231] [87000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.53054400936443
+I1204 05:24:04.139837 137274321021824 utils.py:1231] [87000] core_hours = 151.53054400936443
+I1204 05:24:04.139924 137274321021824 train.py:125] NOTE: Steps:87000/112603 [77.3%]
+Walltime:6d7h33m (0s eval)
+ETA:1d20h35m
+Total train time:8d4h7m
+I1204 05:29:15.320644 137274321021824 utils.py:1231] [87050] l2_params = 250.63944892257797
+I1204 05:29:15.320887 137274321021824 utils.py:1231] [87050] train/loss = 2.4321529269218445
+I1204 05:29:15.320996 137274321021824 utils.py:1231] [87050] l2_grads = 2.086087942123413
+I1204 05:29:15.321071 137274321021824 utils.py:1231] [87050] lr = 0.00014540080273423443
+I1204 05:29:15.321136 137274321021824 utils.py:1231] [87050] uptime = 545944.683497129
+I1204 05:29:15.321197 137274321021824 utils.py:1231] [87050] examples_seen = 89139200.0
+I1204 05:29:15.321256 137274321021824 utils.py:1231] [87050] progress = 0.7730699892542827
+I1204 05:29:15.321314 137274321021824 utils.py:1231] [87050] epoch = 69.57656574045382
+I1204 05:29:15.321383 137274321021824 utils.py:1231] [87050] img/sec/core = 164.53409963920595
+I1204 05:29:15.321469 137274321021824 utils.py:1231] [87050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.61698336906804
+I1204 05:29:15.321540 137274321021824 utils.py:1231] [87050] core_hours = 151.61698336906804
+I1204 05:29:15.321608 137274321021824 train.py:125] NOTE: Steps:87050/112603 [77.3%]
+Walltime:6d7h39m (0s eval)
+ETA:1d20h30m
+Total train time:8d4h7m
+I1204 05:34:24.082126 137274321021824 utils.py:1231] [87100] l2_params = 250.58096374761016
+I1204 05:34:24.082348 137274321021824 utils.py:1231] [87100] train/loss = 2.9035979509353638
+I1204 05:34:24.082494 137274321021824 utils.py:1231] [87100] l2_grads = 1.987367033958435
+I1204 05:34:24.082607 137274321021824 utils.py:1231] [87100] lr = 0.0001448615532412254
+I1204 05:34:24.082695 137274321021824 utils.py:1231] [87100] uptime = 546253.445051659
+I1204 05:34:24.082781 137274321021824 utils.py:1231] [87100] examples_seen = 89190400.0
+I1204 05:34:24.082859 137274321021824 utils.py:1231] [87100] progress = 0.7735140271573582
+I1204 05:34:24.082944 137274321021824 utils.py:1231] [87100] epoch = 69.61652930492278
+I1204 05:34:24.083021 137274321021824 utils.py:1231] [87100] img/sec/core = 165.82375379582484
+I1204 05:34:24.083095 137274321021824 utils.py:1231] [87100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.7027504675486
+I1204 05:34:24.083165 137274321021824 utils.py:1231] [87100] core_hours = 151.7027504675486
+I1204 05:34:24.083266 137274321021824 train.py:125] NOTE: Steps:87100/112603 [77.4%]
+Walltime:6d7h44m (0s eval)
+ETA:1d20h25m
+Total train time:8d4h7m
+I1204 05:39:33.011014 137274321021824 utils.py:1231] [87150] l2_params = 250.52094208159173
+I1204 05:39:33.011241 137274321021824 utils.py:1231] [87150] train/loss = 1.7795701622962952
+I1204 05:39:33.011335 137274321021824 utils.py:1231] [87150] l2_grads = 2.361048460006714
+I1204 05:39:33.011394 137274321021824 utils.py:1231] [87150] lr = 0.00014432313611979295
+I1204 05:39:33.011445 137274321021824 utils.py:1231] [87150] uptime = 546562.373806739
+I1204 05:39:33.011497 137274321021824 utils.py:1231] [87150] examples_seen = 89241600.0
+I1204 05:39:33.011546 137274321021824 utils.py:1231] [87150] progress = 0.7739580650604335
+I1204 05:39:33.011593 137274321021824 utils.py:1231] [87150] epoch = 69.65649286939174
+I1204 05:39:33.011643 137274321021824 utils.py:1231] [87150] img/sec/core = 165.7340055209328
+I1204 05:39:33.011698 137274321021824 utils.py:1231] [87150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.7885640106264
+I1204 05:39:33.011748 137274321021824 utils.py:1231] [87150] core_hours = 151.7885640106264
+I1204 05:39:33.011808 137274321021824 train.py:125] NOTE: Steps:87150/112603 [77.4%]
+Walltime:6d7h49m (0s eval)
+ETA:1d20h19m
+Total train time:8d4h7m
+I1204 05:44:42.622691 137274321021824 utils.py:1231] [87200] l2_params = 250.45734735437975
+I1204 05:44:42.622902 137274321021824 utils.py:1231] [87200] train/loss = 1.7242033928632736
+I1204 05:44:42.623013 137274321021824 utils.py:1231] [87200] l2_grads = 2.484956741333008
+I1204 05:44:42.623082 137274321021824 utils.py:1231] [87200] lr = 0.00014378555263187603
+I1204 05:44:42.623142 137274321021824 utils.py:1231] [87200] uptime = 546871.985503632
+I1204 05:44:42.623211 137274321021824 utils.py:1231] [87200] examples_seen = 89292800.0
+I1204 05:44:42.623284 137274321021824 utils.py:1231] [87200] progress = 0.774402102963509
+I1204 05:44:42.623351 137274321021824 utils.py:1231] [87200] epoch = 69.69645643386069
+I1204 05:44:42.623420 137274321021824 utils.py:1231] [87200] img/sec/core = 165.36842927382816
+I1204 05:44:42.623497 137274321021824 utils.py:1231] [87200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.87456725976332
+I1204 05:44:42.623571 137274321021824 utils.py:1231] [87200] core_hours = 151.87456725976332
+I1204 05:44:42.623644 137274321021824 train.py:125] NOTE: Steps:87200/112603 [77.4%]
+Walltime:6d7h54m (0s eval)
+ETA:1d20h14m
+Total train time:8d4h7m
+I1204 05:49:52.387417 137274321021824 utils.py:1231] [87250] l2_params = 250.39739928140847
+I1204 05:49:52.387667 137274321021824 utils.py:1231] [87250] train/loss = 1.6932913213968277
+I1204 05:49:52.387784 137274321021824 utils.py:1231] [87250] l2_grads = 2.2966301441192627
+I1204 05:49:52.387865 137274321021824 utils.py:1231] [87250] lr = 0.0001432488040374599
+I1204 05:49:52.387941 137274321021824 utils.py:1231] [87250] uptime = 547181.75029895
+I1204 05:49:52.388007 137274321021824 utils.py:1231] [87250] examples_seen = 89344000.0
+I1204 05:49:52.388057 137274321021824 utils.py:1231] [87250] progress = 0.7748461408665843
+I1204 05:49:52.388111 137274321021824 utils.py:1231] [87250] epoch = 69.73641999832965
+I1204 05:49:52.388178 137274321021824 utils.py:1231] [87250] img/sec/core = 165.28669743582924
+I1204 05:49:52.388234 137274321021824 utils.py:1231] [87250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 151.96061303624055
+I1204 05:49:52.388284 137274321021824 utils.py:1231] [87250] core_hours = 151.96061303624055
+I1204 05:49:52.388345 137274321021824 train.py:125] NOTE: Steps:87250/112603 [77.5%]
+Walltime:6d7h59m (0s eval)
+ETA:1d20h9m
+Total train time:8d4h7m
+I1204 05:55:04.159659 137274321021824 utils.py:1231] [87300] l2_params = 250.33380317926972
+I1204 05:55:04.159893 137274321021824 utils.py:1231] [87300] train/loss = 1.6109357327222824
+I1204 05:55:04.160040 137274321021824 utils.py:1231] [87300] l2_grads = 2.2107038497924805
+I1204 05:55:04.160127 137274321021824 utils.py:1231] [87300] lr = 0.00014271289159457313
+I1204 05:55:04.160204 137274321021824 utils.py:1231] [87300] uptime = 547493.522565391
+I1204 05:55:04.160267 137274321021824 utils.py:1231] [87300] examples_seen = 89395200.0
+I1204 05:55:04.160331 137274321021824 utils.py:1231] [87300] progress = 0.7752901787696598
+I1204 05:55:04.160404 137274321021824 utils.py:1231] [87300] epoch = 69.7763835627986
+I1204 05:55:04.160469 137274321021824 utils.py:1231] [87300] img/sec/core = 164.22243256099574
+I1204 05:55:04.160556 137274321021824 utils.py:1231] [87300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.04721644358526
+I1204 05:55:04.160622 137274321021824 utils.py:1231] [87300] core_hours = 152.04721644358526
+I1204 05:55:04.160692 137274321021824 train.py:125] NOTE: Steps:87300/112603 [77.5%]
+Walltime:6d8h4m (0s eval)
+ETA:1d20h4m
+Total train time:8d4h7m
+I1204 06:00:12.896425 137274321021824 utils.py:1231] [87350] l2_params = 250.27304863028027
+I1204 06:00:12.896666 137274321021824 utils.py:1231] [87350] train/loss = 2.1428206264972687
+I1204 06:00:12.896785 137274321021824 utils.py:1231] [87350] l2_grads = 2.3955869674682617
+I1204 06:00:12.896871 137274321021824 utils.py:1231] [87350] lr = 0.00014217781655928439
+I1204 06:00:12.896968 137274321021824 utils.py:1231] [87350] uptime = 547802.25931443
+I1204 06:00:12.897038 137274321021824 utils.py:1231] [87350] examples_seen = 89446400.0
+I1204 06:00:12.897100 137274321021824 utils.py:1231] [87350] progress = 0.7757342166727352
+I1204 06:00:12.897155 137274321021824 utils.py:1231] [87350] epoch = 69.81634712726756
+I1204 06:00:12.897232 137274321021824 utils.py:1231] [87350] img/sec/core = 165.83707692519113
+I1204 06:00:12.897308 137274321021824 utils.py:1231] [87350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.13297665165163
+I1204 06:00:12.897361 137274321021824 utils.py:1231] [87350] core_hours = 152.13297665165163
+I1204 06:00:12.897432 137274321021824 train.py:125] NOTE: Steps:87350/112603 [77.6%]
+Walltime:6d8h10m (0s eval)
+ETA:1d19h58m
+Total train time:8d4h7m
+I1204 06:05:24.668866 137274321021824 utils.py:1231] [87400] l2_params = 250.21062631915552
+I1204 06:05:24.669100 137274321021824 utils.py:1231] [87400] train/loss = 2.9363605976104736
+I1204 06:05:24.669254 137274321021824 utils.py:1231] [87400] l2_grads = 2.065793037414551
+I1204 06:05:24.669360 137274321021824 utils.py:1231] [87400] lr = 0.00014164358018569953
+I1204 06:05:24.669446 137274321021824 utils.py:1231] [87400] uptime = 548114.031802853
+I1204 06:05:24.669534 137274321021824 utils.py:1231] [87400] examples_seen = 89497600.0
+I1204 06:05:24.669612 137274321021824 utils.py:1231] [87400] progress = 0.7761782545758106
+I1204 06:05:24.669694 137274321021824 utils.py:1231] [87400] epoch = 69.85631069173652
+I1204 06:05:24.669768 137274321021824 utils.py:1231] [87400] img/sec/core = 164.22231563462142
+I1204 06:05:24.669843 137274321021824 utils.py:1231] [87400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.21958012065804
+I1204 06:05:24.669923 137274321021824 utils.py:1231] [87400] core_hours = 152.21958012065804
+I1204 06:05:24.670009 137274321021824 train.py:125] NOTE: Steps:87400/112603 [77.6%]
+Walltime:6d8h15m (0s eval)
+ETA:1d19h53m
+Total train time:8d4h7m
+I1204 06:10:31.503736 137274321021824 utils.py:1231] [87450] l2_params = 250.1526137654197
+I1204 06:10:31.503956 137274321021824 utils.py:1231] [87450] train/loss = 2.3810958862304688
+I1204 06:10:31.504050 137274321021824 utils.py:1231] [87450] l2_grads = 2.1761131286621094
+I1204 06:10:31.504113 137274321021824 utils.py:1231] [87450] lr = 0.00014111018372595923
+I1204 06:10:31.504164 137274321021824 utils.py:1231] [87450] uptime = 548420.866526429
+I1204 06:10:31.504217 137274321021824 utils.py:1231] [87450] examples_seen = 89548800.0
+I1204 06:10:31.504266 137274321021824 utils.py:1231] [87450] progress = 0.776622292478886
+I1204 06:10:31.504321 137274321021824 utils.py:1231] [87450] epoch = 69.89627425620547
+I1204 06:10:31.504374 137274321021824 utils.py:1231] [87450] img/sec/core = 166.8650777307573
+I1204 06:10:31.504430 137274321021824 utils.py:1231] [87450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.30481198831802
+I1204 06:10:31.504482 137274321021824 utils.py:1231] [87450] core_hours = 152.30481198831802
+I1204 06:10:31.504542 137274321021824 train.py:125] NOTE: Steps:87450/112603 [77.7%]
+Walltime:6d8h20m (0s eval)
+ETA:1d19h48m
+Total train time:8d4h6m
+I1204 06:15:38.511297 137274321021824 utils.py:1231] [87500] l2_params = 250.0920097961904
+I1204 06:15:38.511513 137274321021824 utils.py:1231] [87500] train/loss = 1.8616688549518585
+I1204 06:15:38.511635 137274321021824 utils.py:1231] [87500] l2_grads = 2.3644022941589355
+I1204 06:15:38.511702 137274321021824 utils.py:1231] [87500] lr = 0.0001405776284302347
+I1204 06:15:38.511759 137274321021824 utils.py:1231] [87500] uptime = 548727.8741209949
+I1204 06:15:38.511823 137274321021824 utils.py:1231] [87500] examples_seen = 89600000.0
+I1204 06:15:38.511871 137274321021824 utils.py:1231] [87500] progress = 0.7770663303819614
+I1204 06:15:38.511924 137274321021824 utils.py:1231] [87500] epoch = 69.93623782067444
+I1204 06:15:38.511985 137274321021824 utils.py:1231] [87500] img/sec/core = 166.77111871575178
+I1204 06:15:38.512040 137274321021824 utils.py:1231] [87500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.3900918756975
+I1204 06:15:38.512096 137274321021824 utils.py:1231] [87500] core_hours = 152.3900918756975
+I1204 06:15:38.512156 137274321021824 train.py:125] NOTE: Steps:87500/112603 [77.7%]
+Walltime:6d8h25m (0s eval)
+ETA:1d19h43m
+Total train time:8d4h6m
+I1204 06:15:38.512259 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:87500/112603 [77.7%]
+Walltime:6d8h25m (0s eval)
+ETA:1d19h43m
+Total train time:8d4h6m
+I1204 06:17:09.409044 137274321021824 utils.py:1231] [87500] val/acc@1 = 0.7374641262755102
+I1204 06:17:09.409266 137274321021824 utils.py:1231] [87500] val/loss = 1.0311980816174526
+I1204 06:17:09.409449 137274321021824 utils.py:1231] [87500] z/secs/eval/val = 90.89711880800314
+I1204 06:17:09.409517 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 90.89711880800314
+I1204 06:22:13.474556 137274321021824 utils.py:1231] [87550] l2_params = 250.0321238127864
+I1204 06:22:13.474792 137274321021824 utils.py:1231] [87550] train/loss = 1.6395836472511292
+I1204 06:22:13.474904 137274321021824 utils.py:1231] [87550] l2_grads = 2.354396343231201
+I1204 06:22:13.474976 137274321021824 utils.py:1231] [87550] lr = 0.00014004591554672659
+I1204 06:22:13.475037 137274321021824 utils.py:1231] [87550] uptime = 549122.837397698
+I1204 06:22:13.475129 137274321021824 utils.py:1231] [87550] examples_seen = 89651200.0
+I1204 06:22:13.475198 137274321021824 utils.py:1231] [87550] progress = 0.7775103682850368
+I1204 06:22:13.475254 137274321021824 utils.py:1231] [87550] epoch = 69.97620138514338
+I1204 06:22:13.475310 137274321021824 utils.py:1231] [87550] img/sec/core = 129.63230512819877
+I1204 06:22:13.475367 137274321021824 utils.py:1231] [87550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.4998038970039
+I1204 06:22:13.475423 137274321021824 utils.py:1231] [87550] core_hours = 152.4998038970039
+I1204 06:22:13.475490 137274321021824 train.py:125] NOTE: Steps:87550/112603 [77.8%]
+Walltime:6d8h32m (0s eval)
+ETA:1d19h38m
+Total train time:8d4h8m
+I1204 06:27:17.106079 137274321021824 utils.py:1231] [87600] l2_params = 249.97135734748198
+I1204 06:27:17.106287 137274321021824 utils.py:1231] [87600] train/loss = 1.8661184459924698
+I1204 06:27:17.106385 137274321021824 utils.py:1231] [87600] l2_grads = 2.2360377311706543
+I1204 06:27:17.106476 137274321021824 utils.py:1231] [87600] lr = 0.00013951504632166078
+I1204 06:27:17.106537 137274321021824 utils.py:1231] [87600] uptime = 549426.468898684
+I1204 06:27:17.106596 137274321021824 utils.py:1231] [87600] examples_seen = 89702400.0
+I1204 06:27:17.106651 137274321021824 utils.py:1231] [87600] progress = 0.7779544061881122
+I1204 06:27:17.106706 137274321021824 utils.py:1231] [87600] epoch = 70.01616494961235
+I1204 06:27:17.106796 137274321021824 utils.py:1231] [87600] img/sec/core = 168.62545497988827
+I1204 06:27:17.106857 137274321021824 utils.py:1231] [87600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.5841459806111
+I1204 06:27:17.106928 137274321021824 utils.py:1231] [87600] core_hours = 152.5841459806111
+I1204 06:27:17.107016 137274321021824 train.py:125] NOTE: Steps:87600/112603 [77.8%]
+Walltime:6d8h37m (0s eval)
+ETA:1d19h33m
+Total train time:8d4h8m
+I1204 06:32:28.873026 137274321021824 utils.py:1231] [87650] l2_params = 249.91661521323167
+I1204 06:32:28.873301 137274321021824 utils.py:1231] [87650] train/loss = 2.9321391880512238
+I1204 06:32:28.873416 137274321021824 utils.py:1231] [87650] l2_grads = 2.066987991333008
+I1204 06:32:28.873490 137274321021824 utils.py:1231] [87650] lr = 0.00013898502199928537
+I1204 06:32:28.873542 137274321021824 utils.py:1231] [87650] uptime = 549738.23590426
+I1204 06:32:28.873597 137274321021824 utils.py:1231] [87650] examples_seen = 89753600.0
+I1204 06:32:28.873646 137274321021824 utils.py:1231] [87650] progress = 0.7783984440911876
+I1204 06:32:28.873693 137274321021824 utils.py:1231] [87650] epoch = 70.05612851408131
+I1204 06:32:28.873743 137274321021824 utils.py:1231] [87650] img/sec/core = 164.22520370754626
+I1204 06:32:28.873800 137274321021824 utils.py:1231] [87650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.6707479266044
+I1204 06:32:28.873850 137274321021824 utils.py:1231] [87650] core_hours = 152.6707479266044
+I1204 06:32:28.873915 137274321021824 train.py:125] NOTE: Steps:87650/112603 [77.8%]
+Walltime:6d8h42m (0s eval)
+ETA:1d19h27m
+Total train time:8d4h8m
+I1204 06:37:40.631060 137274321021824 utils.py:1231] [87700] l2_params = 249.85938991205217
+I1204 06:37:40.631252 137274321021824 utils.py:1231] [87700] train/loss = 2.2797919511795044
+I1204 06:37:40.631365 137274321021824 utils.py:1231] [87700] l2_grads = 2.2032744884490967
+I1204 06:37:40.631438 137274321021824 utils.py:1231] [87700] lr = 0.00013845584382186858
+I1204 06:37:40.631499 137274321021824 utils.py:1231] [87700] uptime = 550049.993860753
+I1204 06:37:40.631567 137274321021824 utils.py:1231] [87700] examples_seen = 89804800.0
+I1204 06:37:40.631626 137274321021824 utils.py:1231] [87700] progress = 0.778842481994263
+I1204 06:37:40.631683 137274321021824 utils.py:1231] [87700] epoch = 70.09609207855026
+I1204 06:37:40.631741 137274321021824 utils.py:1231] [87700] img/sec/core = 164.22997050642755
+I1204 06:37:40.631812 137274321021824 utils.py:1231] [87700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.7573473589636
+I1204 06:37:40.631868 137274321021824 utils.py:1231] [87700] core_hours = 152.7573473589636
+I1204 06:37:40.631941 137274321021824 train.py:125] NOTE: Steps:87700/112603 [77.9%]
+Walltime:6d8h47m (0s eval)
+ETA:1d19h22m
+Total train time:8d4h8m
+I1204 06:42:50.103158 137274321021824 utils.py:1231] [87750] l2_params = 249.79756307850167
+I1204 06:42:50.103367 137274321021824 utils.py:1231] [87750] train/loss = 3.660575211048126
+I1204 06:42:50.103508 137274321021824 utils.py:1231] [87750] l2_grads = 2.24892258644104
+I1204 06:42:50.103597 137274321021824 utils.py:1231] [87750] lr = 0.0001379275130296953
+I1204 06:42:50.103678 137274321021824 utils.py:1231] [87750] uptime = 550359.466036554
+I1204 06:42:50.103754 137274321021824 utils.py:1231] [87750] examples_seen = 89856000.0
+I1204 06:42:50.103826 137274321021824 utils.py:1231] [87750] progress = 0.7792865198973384
+I1204 06:42:50.103905 137274321021824 utils.py:1231] [87750] epoch = 70.13605564301922
+I1204 06:42:50.103984 137274321021824 utils.py:1231] [87750] img/sec/core = 165.44298325847905
+I1204 06:42:50.104059 137274321021824 utils.py:1231] [87750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.84331185224164
+I1204 06:42:50.104123 137274321021824 utils.py:1231] [87750] core_hours = 152.84331185224164
+I1204 06:42:50.104212 137274321021824 train.py:125] NOTE: Steps:87750/112603 [77.9%]
+Walltime:6d8h52m (0s eval)
+ETA:1d19h17m
+Total train time:8d4h8m
+I1204 06:48:00.148258 137274321021824 utils.py:1231] [87800] l2_params = 249.7390754223705
+I1204 06:48:00.148480 137274321021824 utils.py:1231] [87800] train/loss = 2.826297700405121
+I1204 06:48:00.148577 137274321021824 utils.py:1231] [87800] l2_grads = 2.1610984802246094
+I1204 06:48:00.148643 137274321021824 utils.py:1231] [87800] lr = 0.00013740003086106455
+I1204 06:48:00.148712 137274321021824 utils.py:1231] [87800] uptime = 550669.511070918
+I1204 06:48:00.148769 137274321021824 utils.py:1231] [87800] examples_seen = 89907200.0
+I1204 06:48:00.148819 137274321021824 utils.py:1231] [87800] progress = 0.7797305578004139
+I1204 06:48:00.148867 137274321021824 utils.py:1231] [87800] epoch = 70.17601920748817
+I1204 06:48:00.148925 137274321021824 utils.py:1231] [87800] img/sec/core = 165.1373004732369
+I1204 06:48:00.148996 137274321021824 utils.py:1231] [87800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 152.9294354728983
+I1204 06:48:00.149055 137274321021824 utils.py:1231] [87800] core_hours = 152.9294354728983
+I1204 06:48:00.149143 137274321021824 train.py:125] NOTE: Steps:87800/112603 [78.0%]
+Walltime:6d8h57m (0s eval)
+ETA:1d19h12m
+Total train time:8d4h8m
+I1204 06:53:10.666194 137274321021824 utils.py:1231] [87850] l2_params = 249.68572176895674
+I1204 06:53:10.666497 137274321021824 utils.py:1231] [87850] train/loss = 2.226947396993637
+I1204 06:53:10.666666 137274321021824 utils.py:1231] [87850] l2_grads = 2.268829345703125
+I1204 06:53:10.666759 137274321021824 utils.py:1231] [87850] lr = 0.00013687339855228626
+I1204 06:53:10.666832 137274321021824 utils.py:1231] [87850] uptime = 550980.02919313
+I1204 06:53:10.666918 137274321021824 utils.py:1231] [87850] examples_seen = 89958400.0
+I1204 06:53:10.666977 137274321021824 utils.py:1231] [87850] progress = 0.7801745957034892
+I1204 06:53:10.667037 137274321021824 utils.py:1231] [87850] epoch = 70.21598277195713
+I1204 06:53:10.667096 137274321021824 utils.py:1231] [87850] img/sec/core = 164.88570662242793
+I1204 06:53:10.667160 137274321021824 utils.py:1231] [87850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.0156905068461
+I1204 06:53:10.667218 137274321021824 utils.py:1231] [87850] core_hours = 153.0156905068461
+I1204 06:53:10.667283 137274321021824 train.py:125] NOTE: Steps:87850/112603 [78.0%]
+Walltime:6d9h3m (0s eval)
+ETA:1d19h6m
+Total train time:8d4h8m
+I1204 06:58:19.738315 137274321021824 utils.py:1231] [87900] l2_params = 249.62547451687175
+I1204 06:58:19.738530 137274321021824 utils.py:1231] [87900] train/loss = 4.082070350646973
+I1204 06:58:19.738659 137274321021824 utils.py:1231] [87900] l2_grads = 2.372856855392456
+I1204 06:58:19.738732 137274321021824 utils.py:1231] [87900] lr = 0.00013634761733767803
+I1204 06:58:19.738790 137274321021824 utils.py:1231] [87900] uptime = 551289.10115178
+I1204 06:58:19.738849 137274321021824 utils.py:1231] [87900] examples_seen = 90009600.0
+I1204 06:58:19.738913 137274321021824 utils.py:1231] [87900] progress = 0.7806186336065647
+I1204 06:58:19.738964 137274321021824 utils.py:1231] [87900] epoch = 70.25594633642609
+I1204 06:58:19.739018 137274321021824 utils.py:1231] [87900] img/sec/core = 165.65721530880361
+I1204 06:58:19.739078 137274321021824 utils.py:1231] [87900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.1015438286933
+I1204 06:58:19.739129 137274321021824 utils.py:1231] [87900] core_hours = 153.1015438286933
+I1204 06:58:19.739205 137274321021824 train.py:125] NOTE: Steps:87900/112603 [78.1%]
+Walltime:6d9h8m (0s eval)
+ETA:1d19h1m
+Total train time:8d4h7m
+I1204 07:03:22.344640 137274321021824 utils.py:1231] [87950] l2_params = 249.57350877335247
+I1204 07:03:22.344856 137274321021824 utils.py:1231] [87950] train/loss = 3.736176609992981
+I1204 07:03:22.345005 137274321021824 utils.py:1231] [87950] l2_grads = 2.2197957038879395
+I1204 07:03:22.345076 137274321021824 utils.py:1231] [87950] lr = 0.0001358226884495633
+I1204 07:03:22.345126 137274321021824 utils.py:1231] [87950] uptime = 551591.707488732
+I1204 07:03:22.345187 137274321021824 utils.py:1231] [87950] examples_seen = 90060800.0
+I1204 07:03:22.345240 137274321021824 utils.py:1231] [87950] progress = 0.78106267150964
+I1204 07:03:22.345287 137274321021824 utils.py:1231] [87950] epoch = 70.29590990089504
+I1204 07:03:22.345337 137274321021824 utils.py:1231] [87950] img/sec/core = 169.19672111205384
+I1204 07:03:22.345391 137274321021824 utils.py:1231] [87950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.18560114451333
+I1204 07:03:22.345448 137274321021824 utils.py:1231] [87950] core_hours = 153.18560114451333
+I1204 07:03:22.345507 137274321021824 train.py:125] NOTE: Steps:87950/112603 [78.1%]
+Walltime:6d9h13m (0s eval)
+ETA:1d18h56m
+Total train time:8d4h7m
+I1204 07:08:33.190848 137274321021824 utils.py:1231] [88000] l2_params = 249.51589660097045
+I1204 07:08:33.191058 137274321021824 utils.py:1231] [88000] train/loss = 3.9945660531520844
+I1204 07:08:33.191151 137274321021824 utils.py:1231] [88000] l2_grads = 2.3675198554992676
+I1204 07:08:33.191210 137274321021824 utils.py:1231] [88000] lr = 0.00013529861311826788
+I1204 07:08:33.191262 137274321021824 utils.py:1231] [88000] uptime = 551902.55362378
+I1204 07:08:33.191317 137274321021824 utils.py:1231] [88000] examples_seen = 90112000.0
+I1204 07:08:33.191365 137274321021824 utils.py:1231] [88000] progress = 0.7815067094127155
+I1204 07:08:33.191415 137274321021824 utils.py:1231] [88000] epoch = 70.335873465364
+I1204 07:08:33.191466 137274321021824 utils.py:1231] [88000] img/sec/core = 164.71171498426634
+I1204 07:08:33.191521 137274321021824 utils.py:1231] [88000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.27194729313777
+I1204 07:08:33.191574 137274321021824 utils.py:1231] [88000] core_hours = 153.27194729313777
+I1204 07:08:33.191634 137274321021824 train.py:125] NOTE: Steps:88000/112603 [78.2%]
+Walltime:6d9h18m (0s eval)
+ETA:1d18h51m
+Total train time:8d4h7m
+I1204 07:13:38.801051 137274321021824 utils.py:1231] [88050] l2_params = 249.46008832359954
+I1204 07:13:38.801294 137274321021824 utils.py:1231] [88050] train/loss = 1.678487777709961
+I1204 07:13:38.801416 137274321021824 utils.py:1231] [88050] l2_grads = 2.411135196685791
+I1204 07:13:38.801482 137274321021824 utils.py:1231] [88050] lr = 0.00013477539257211632
+I1204 07:13:38.801535 137274321021824 utils.py:1231] [88050] uptime = 552208.163897097
+I1204 07:13:38.801590 137274321021824 utils.py:1231] [88050] examples_seen = 90163200.0
+I1204 07:13:38.801641 137274321021824 utils.py:1231] [88050] progress = 0.7819507473157908
+I1204 07:13:38.801694 137274321021824 utils.py:1231] [88050] epoch = 70.37583702983295
+I1204 07:13:38.801748 137274321021824 utils.py:1231] [88050] img/sec/core = 167.53363505845616
+I1204 07:13:38.801814 137274321021824 utils.py:1231] [88050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.35683903572584
+I1204 07:13:38.801864 137274321021824 utils.py:1231] [88050] core_hours = 153.35683903572584
+I1204 07:13:38.801942 137274321021824 train.py:125] NOTE: Steps:88050/112603 [78.2%]
+Walltime:6d9h23m (0s eval)
+ETA:1d18h45m
+Total train time:8d4h7m
+I1204 07:18:48.164635 137274321021824 utils.py:1231] [88100] l2_params = 249.39858818904048
+I1204 07:18:48.164877 137274321021824 utils.py:1231] [88100] train/loss = 2.7666701674461365
+I1204 07:18:48.164989 137274321021824 utils.py:1231] [88100] l2_grads = 2.1138558387756348
+I1204 07:18:48.165056 137274321021824 utils.py:1231] [88100] lr = 0.00013425302803743
+I1204 07:18:48.165112 137274321021824 utils.py:1231] [88100] uptime = 552517.52747468
+I1204 07:18:48.165164 137274321021824 utils.py:1231] [88100] examples_seen = 90214400.0
+I1204 07:18:48.165215 137274321021824 utils.py:1231] [88100] progress = 0.7823947852188663
+I1204 07:18:48.165271 137274321021824 utils.py:1231] [88100] epoch = 70.41580059430191
+I1204 07:18:48.165320 137274321021824 utils.py:1231] [88100] img/sec/core = 165.50105995029455
+I1204 07:18:48.165379 137274321021824 utils.py:1231] [88100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.4427733628322
+I1204 07:18:48.165428 137274321021824 utils.py:1231] [88100] core_hours = 153.4427733628322
+I1204 07:18:48.165487 137274321021824 train.py:125] NOTE: Steps:88100/112603 [78.2%]
+Walltime:6d9h28m (0s eval)
+ETA:1d18h40m
+Total train time:8d4h7m
+I1204 07:23:59.945969 137274321021824 utils.py:1231] [88150] l2_params = 249.34140483223308
+I1204 07:23:59.946169 137274321021824 utils.py:1231] [88150] train/loss = 1.6334338188171387
+I1204 07:23:59.946272 137274321021824 utils.py:1231] [88150] l2_grads = 2.3187220096588135
+I1204 07:23:59.946333 137274321021824 utils.py:1231] [88150] lr = 0.00013373152073852432
+I1204 07:23:59.946384 137274321021824 utils.py:1231] [88150] uptime = 552829.308746006
+I1204 07:23:59.946446 137274321021824 utils.py:1231] [88150] examples_seen = 90265600.0
+I1204 07:23:59.946496 137274321021824 utils.py:1231] [88150] progress = 0.7828388231219416
+I1204 07:23:59.946544 137274321021824 utils.py:1231] [88150] epoch = 70.45576415877088
+I1204 07:23:59.946595 137274321021824 utils.py:1231] [88150] img/sec/core = 164.21768947905784
+I1204 07:23:59.946651 137274321021824 utils.py:1231] [88150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.52937927153388
+I1204 07:23:59.946706 137274321021824 utils.py:1231] [88150] core_hours = 153.52937927153388
+I1204 07:23:59.946766 137274321021824 train.py:125] NOTE: Steps:88150/112603 [78.3%]
+Walltime:6d9h33m (0s eval)
+ETA:1d18h35m
+Total train time:8d4h7m
+I1204 07:29:09.068557 137274321021824 utils.py:1231] [88200] l2_params = 249.28493202912273
+I1204 07:29:09.068762 137274321021824 utils.py:1231] [88200] train/loss = 3.2446931302547455
+I1204 07:29:09.068875 137274321021824 utils.py:1231] [88200] l2_grads = 2.1385421752929688
+I1204 07:29:09.068943 137274321021824 utils.py:1231] [88200] lr = 0.00013321087189770525
+I1204 07:29:09.068993 137274321021824 utils.py:1231] [88200] uptime = 553138.43135513
+I1204 07:29:09.069045 137274321021824 utils.py:1231] [88200] examples_seen = 90316800.0
+I1204 07:29:09.069094 137274321021824 utils.py:1231] [88200] progress = 0.7832828610250171
+I1204 07:29:09.069150 137274321021824 utils.py:1231] [88200] epoch = 70.49572772323982
+I1204 07:29:09.069211 137274321021824 utils.py:1231] [88200] img/sec/core = 165.6300719804453
+I1204 07:29:09.069269 137274321021824 utils.py:1231] [88200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.61524666295725
+I1204 07:29:09.069323 137274321021824 utils.py:1231] [88200] core_hours = 153.61524666295725
+I1204 07:29:09.069394 137274321021824 train.py:125] NOTE: Steps:88200/112603 [78.3%]
+Walltime:6d9h38m (0s eval)
+ETA:1d18h30m
+Total train time:8d4h7m
+I1204 07:34:17.131178 137274321021824 utils.py:1231] [88250] l2_params = 249.22726867191648
+I1204 07:34:17.131499 137274321021824 utils.py:1231] [88250] train/loss = 3.5040185749530792
+I1204 07:34:17.131696 137274321021824 utils.py:1231] [88250] l2_grads = 2.3052260875701904
+I1204 07:34:17.131788 137274321021824 utils.py:1231] [88250] lr = 0.00013269108273526685
+I1204 07:34:17.131864 137274321021824 utils.py:1231] [88250] uptime = 553446.494225679
+I1204 07:34:17.131935 137274321021824 utils.py:1231] [88250] examples_seen = 90368000.0
+I1204 07:34:17.131994 137274321021824 utils.py:1231] [88250] progress = 0.7837268989280926
+I1204 07:34:17.132050 137274321021824 utils.py:1231] [88250] epoch = 70.53569128770879
+I1204 07:34:17.132110 137274321021824 utils.py:1231] [88250] img/sec/core = 166.19984066489192
+I1204 07:34:17.132180 137274321021824 utils.py:1231] [88250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.70081968255414
+I1204 07:34:17.132252 137274321021824 utils.py:1231] [88250] core_hours = 153.70081968255414
+I1204 07:34:17.132318 137274321021824 train.py:125] NOTE: Steps:88250/112603 [78.4%]
+Walltime:6d9h44m (0s eval)
+ETA:1d18h24m
+Total train time:8d4h7m
+I1204 07:39:28.112000 137274321021824 utils.py:1231] [88300] l2_params = 249.16809795976178
+I1204 07:39:28.112197 137274321021824 utils.py:1231] [88300] train/loss = 4.039557129144669
+I1204 07:39:28.112297 137274321021824 utils.py:1231] [88300] l2_grads = 2.365553140640259
+I1204 07:39:28.112359 137274321021824 utils.py:1231] [88300] lr = 0.0001321721544694881
+I1204 07:39:28.112410 137274321021824 utils.py:1231] [88300] uptime = 553757.47477212
+I1204 07:39:28.112464 137274321021824 utils.py:1231] [88300] examples_seen = 90419200.0
+I1204 07:39:28.112514 137274321021824 utils.py:1231] [88300] progress = 0.7841709368311679
+I1204 07:39:28.112566 137274321021824 utils.py:1231] [88300] epoch = 70.57565485217773
+I1204 07:39:28.112616 137274321021824 utils.py:1231] [88300] img/sec/core = 164.6405236145737
+I1204 07:39:28.112671 137274321021824 utils.py:1231] [88300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.78720316767667
+I1204 07:39:28.112720 137274321021824 utils.py:1231] [88300] core_hours = 153.78720316767667
+I1204 07:39:28.112778 137274321021824 train.py:125] NOTE: Steps:88300/112603 [78.4%]
+Walltime:6d9h49m (0s eval)
+ETA:1d18h19m
+Total train time:8d4h7m
+I1204 07:44:39.896631 137274321021824 utils.py:1231] [88350] l2_params = 249.1137165623688
+I1204 07:44:39.896832 137274321021824 utils.py:1231] [88350] train/loss = 1.8907674551010132
+I1204 07:44:39.896944 137274321021824 utils.py:1231] [88350] l2_grads = 2.2497735023498535
+I1204 07:44:39.897028 137274321021824 utils.py:1231] [88350] lr = 0.00013165408831662993
+I1204 07:44:39.897096 137274321021824 utils.py:1231] [88350] uptime = 554069.259457647
+I1204 07:44:39.897168 137274321021824 utils.py:1231] [88350] examples_seen = 90470400.0
+I1204 07:44:39.897226 137274321021824 utils.py:1231] [88350] progress = 0.7846149747342434
+I1204 07:44:39.897290 137274321021824 utils.py:1231] [88350] epoch = 70.6156184166467
+I1204 07:44:39.897362 137274321021824 utils.py:1231] [88350] img/sec/core = 164.21589121179065
+I1204 07:44:39.897420 137274321021824 utils.py:1231] [88350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.8738100247675
+I1204 07:44:39.897499 137274321021824 utils.py:1231] [88350] core_hours = 153.8738100247675
+I1204 07:44:39.897569 137274321021824 train.py:125] NOTE: Steps:88350/112603 [78.5%]
+Walltime:6d9h54m (0s eval)
+ETA:1d18h14m
+Total train time:8d4h7m
+I1204 07:49:49.562814 137274321021824 utils.py:1231] [88400] l2_params = 249.05719184747178
+I1204 07:49:49.563081 137274321021824 utils.py:1231] [88400] train/loss = 1.6482049971818924
+I1204 07:49:49.563300 137274321021824 utils.py:1231] [88400] l2_grads = 2.3061232566833496
+I1204 07:49:49.563376 137274321021824 utils.py:1231] [88400] lr = 0.00013113688549093322
+I1204 07:49:49.563438 137274321021824 utils.py:1231] [88400] uptime = 554378.925800166
+I1204 07:49:49.563491 137274321021824 utils.py:1231] [88400] examples_seen = 90521600.0
+I1204 07:49:49.563546 137274321021824 utils.py:1231] [88400] progress = 0.7850590126373187
+I1204 07:49:49.563594 137274321021824 utils.py:1231] [88400] epoch = 70.65558198111566
+I1204 07:49:49.563644 137274321021824 utils.py:1231] [88400] img/sec/core = 165.33924734445452
+I1204 07:49:49.563700 137274321021824 utils.py:1231] [88400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 153.959828453245
+I1204 07:49:49.563750 137274321021824 utils.py:1231] [88400] core_hours = 153.959828453245
+I1204 07:49:49.563820 137274321021824 train.py:125] NOTE: Steps:88400/112603 [78.5%]
+Walltime:6d9h59m (0s eval)
+ETA:1d18h9m
+Total train time:8d4h7m
+I1204 07:54:58.943402 137274321021824 utils.py:1231] [88450] l2_params = 249.0009430423175
+I1204 07:54:58.943600 137274321021824 utils.py:1231] [88450] train/loss = 3.850642293691635
+I1204 07:54:58.943692 137274321021824 utils.py:1231] [88450] l2_grads = 2.444694995880127
+I1204 07:54:58.943751 137274321021824 utils.py:1231] [88450] lr = 0.00013062054720461527
+I1204 07:54:58.943803 137274321021824 utils.py:1231] [88450] uptime = 554688.306165189
+I1204 07:54:58.943857 137274321021824 utils.py:1231] [88450] examples_seen = 90572800.0
+I1204 07:54:58.943909 137274321021824 utils.py:1231] [88450] progress = 0.7855030505403942
+I1204 07:54:58.943957 137274321021824 utils.py:1231] [88450] epoch = 70.69554554558461
+I1204 07:54:58.944007 137274321021824 utils.py:1231] [88450] img/sec/core = 165.49207961593498
+I1204 07:54:58.944063 137274321021824 utils.py:1231] [88450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.04576744352917
+I1204 07:54:58.944112 137274321021824 utils.py:1231] [88450] core_hours = 154.04576744352917
+I1204 07:54:58.944171 137274321021824 train.py:125] NOTE: Steps:88450/112603 [78.6%]
+Walltime:6d10h4m (0s eval)
+ETA:1d18h3m
+Total train time:8d4h6m
+I1204 08:00:10.715173 137274321021824 utils.py:1231] [88500] l2_params = 248.94306576521706
+I1204 08:00:10.715388 137274321021824 utils.py:1231] [88500] train/loss = 1.612803503870964
+I1204 08:00:10.715494 137274321021824 utils.py:1231] [88500] l2_grads = 2.432213068008423
+I1204 08:00:10.715562 137274321021824 utils.py:1231] [88500] lr = 0.00013010507466786712
+I1204 08:00:10.715621 137274321021824 utils.py:1231] [88500] uptime = 555000.07798269
+I1204 08:00:10.715681 137274321021824 utils.py:1231] [88500] examples_seen = 90624000.0
+I1204 08:00:10.715738 137274321021824 utils.py:1231] [88500] progress = 0.7859470884434695
+I1204 08:00:10.715795 137274321021824 utils.py:1231] [88500] epoch = 70.73550911005357
+I1204 08:00:10.715852 137274321021824 utils.py:1231] [88500] img/sec/core = 164.22266903533162
+I1204 08:00:10.715921 137274321021824 utils.py:1231] [88500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.13237072616832
+I1204 08:00:10.715978 137274321021824 utils.py:1231] [88500] core_hours = 154.13237072616832
+I1204 08:00:10.716043 137274321021824 train.py:125] NOTE: Steps:88500/112603 [78.6%]
+Walltime:6d10h10m (0s eval)
+ETA:1d17h58m
+Total train time:8d4h6m
+I1204 08:05:19.801336 137274321021824 utils.py:1231] [88550] l2_params = 248.88304303372288
+I1204 08:05:19.801566 137274321021824 utils.py:1231] [88550] train/loss = 1.8647817373275757
+I1204 08:05:19.801682 137274321021824 utils.py:1231] [88550] l2_grads = 2.5001182556152344
+I1204 08:05:19.801759 137274321021824 utils.py:1231] [88550] lr = 0.00012959046908885024
+I1204 08:05:19.801821 137274321021824 utils.py:1231] [88550] uptime = 555309.164182717
+I1204 08:05:19.801873 137274321021824 utils.py:1231] [88550] examples_seen = 90675200.0
+I1204 08:05:19.801936 137274321021824 utils.py:1231] [88550] progress = 0.786391126346545
+I1204 08:05:19.801989 137274321021824 utils.py:1231] [88550] epoch = 70.77547267452252
+I1204 08:05:19.802041 137274321021824 utils.py:1231] [88550] img/sec/core = 165.64958252915469
+I1204 08:05:19.802106 137274321021824 utils.py:1231] [88550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.2182280039536
+I1204 08:05:19.802163 137274321021824 utils.py:1231] [88550] core_hours = 154.2182280039536
+I1204 08:05:19.802236 137274321021824 train.py:125] NOTE: Steps:88550/112603 [78.6%]
+Walltime:6d10h15m (0s eval)
+ETA:1d17h53m
+Total train time:8d4h6m
+I1204 08:10:25.425715 137274321021824 utils.py:1231] [88600] l2_params = 248.82772063709024
+I1204 08:10:25.425968 137274321021824 utils.py:1231] [88600] train/loss = 3.6045978367328644
+I1204 08:10:25.426110 137274321021824 utils.py:1231] [88600] l2_grads = 2.4017701148986816
+I1204 08:10:25.426220 137274321021824 utils.py:1231] [88600] lr = 0.00012907673167369423
+I1204 08:10:25.426291 137274321021824 utils.py:1231] [88600] uptime = 555614.788652884
+I1204 08:10:25.426362 137274321021824 utils.py:1231] [88600] examples_seen = 90726400.0
+I1204 08:10:25.426429 137274321021824 utils.py:1231] [88600] progress = 0.7868351642496203
+I1204 08:10:25.426503 137274321021824 utils.py:1231] [88600] epoch = 70.81543623899148
+I1204 08:10:25.426569 137274321021824 utils.py:1231] [88600] img/sec/core = 167.5258527958406
+I1204 08:10:25.426646 137274321021824 utils.py:1231] [88600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.3031236901111
+I1204 08:10:25.426747 137274321021824 utils.py:1231] [88600] core_hours = 154.3031236901111
+I1204 08:10:25.426839 137274321021824 train.py:125] NOTE: Steps:88600/112603 [78.7%]
+Walltime:6d10h20m (0s eval)
+ETA:1d17h48m
+Total train time:8d4h6m
+I1204 08:15:31.671195 137274321021824 utils.py:1231] [88650] l2_params = 248.77043017805778
+I1204 08:15:31.671401 137274321021824 utils.py:1231] [88650] train/loss = 1.6772546172142029
+I1204 08:15:31.671501 137274321021824 utils.py:1231] [88650] l2_grads = 2.3509724140167236
+I1204 08:15:31.671565 137274321021824 utils.py:1231] [88650] lr = 0.0001285638636264944
+I1204 08:15:31.671620 137274321021824 utils.py:1231] [88650] uptime = 555921.033981574
+I1204 08:15:31.671676 137274321021824 utils.py:1231] [88650] examples_seen = 90777600.0
+I1204 08:15:31.671731 137274321021824 utils.py:1231] [88650] progress = 0.7872792021526958
+I1204 08:15:31.671783 137274321021824 utils.py:1231] [88650] epoch = 70.85539980346044
+I1204 08:15:31.671839 137274321021824 utils.py:1231] [88650] img/sec/core = 167.18622360382432
+I1204 08:15:31.671905 137274321021824 utils.py:1231] [88650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.38819183696944
+I1204 08:15:31.671964 137274321021824 utils.py:1231] [88650] core_hours = 154.38819183696944
+I1204 08:15:31.672028 137274321021824 train.py:125] NOTE: Steps:88650/112603 [78.7%]
+Walltime:6d10h25m (0s eval)
+ETA:1d17h42m
+Total train time:8d4h6m
+I1204 08:20:38.370480 137274321021824 utils.py:1231] [88700] l2_params = 248.71135764058812
+I1204 08:20:38.370682 137274321021824 utils.py:1231] [88700] train/loss = 2.0698339641094208
+I1204 08:20:38.370771 137274321021824 utils.py:1231] [88700] l2_grads = 2.151101589202881
+I1204 08:20:38.370830 137274321021824 utils.py:1231] [88700] lr = 0.00012805186614930813
+I1204 08:20:38.370889 137274321021824 utils.py:1231] [88700] uptime = 556227.733243201
+I1204 08:20:38.370944 137274321021824 utils.py:1231] [88700] examples_seen = 90828800.0
+I1204 08:20:38.370993 137274321021824 utils.py:1231] [88700] progress = 0.7877232400557712
+I1204 08:20:38.371042 137274321021824 utils.py:1231] [88700] epoch = 70.89536336792939
+I1204 08:20:38.371092 137274321021824 utils.py:1231] [88700] img/sec/core = 166.93877816458811
+I1204 08:20:38.371160 137274321021824 utils.py:1231] [88700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.4733860763103
+I1204 08:20:38.371224 137274321021824 utils.py:1231] [88700] core_hours = 154.4733860763103
+I1204 08:20:38.371287 137274321021824 train.py:125] NOTE: Steps:88700/112603 [78.8%]
+Walltime:6d10h30m (0s eval)
+ETA:1d17h37m
+Total train time:8d4h6m
+I1204 08:25:44.761870 137274321021824 utils.py:1231] [88750] l2_params = 248.65392846997386
+I1204 08:25:44.762108 137274321021824 utils.py:1231] [88750] train/loss = 1.9599097967147827
+I1204 08:25:44.762207 137274321021824 utils.py:1231] [88750] l2_grads = 2.269345760345459
+I1204 08:25:44.762280 137274321021824 utils.py:1231] [88750] lr = 0.00012754074044215232
+I1204 08:25:44.762340 137274321021824 utils.py:1231] [88750] uptime = 556534.124701783
+I1204 08:25:44.762402 137274321021824 utils.py:1231] [88750] examples_seen = 90880000.0
+I1204 08:25:44.762458 137274321021824 utils.py:1231] [88750] progress = 0.7881672779588466
+I1204 08:25:44.762517 137274321021824 utils.py:1231] [88750] epoch = 70.93532693239835
+I1204 08:25:44.762582 137274321021824 utils.py:1231] [88750] img/sec/core = 167.10648605210363
+I1204 08:25:44.762644 137274321021824 utils.py:1231] [88750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.55849481480527
+I1204 08:25:44.762700 137274321021824 utils.py:1231] [88750] core_hours = 154.55849481480527
+I1204 08:25:44.762770 137274321021824 train.py:125] NOTE: Steps:88750/112603 [78.8%]
+Walltime:6d10h35m (0s eval)
+ETA:1d17h32m
+Total train time:8d4h6m
+I1204 08:30:50.647416 137274321021824 utils.py:1231] [88800] l2_params = 248.59838525321715
+I1204 08:30:50.647626 137274321021824 utils.py:1231] [88800] train/loss = 2.666897028684616
+I1204 08:30:50.891316 137274321021824 utils.py:1231] [88800] l2_grads = 2.1876940727233887
+I1204 08:30:50.891597 137274321021824 utils.py:1231] [88800] lr = 0.00012703048770300042
+I1204 08:30:50.891678 137274321021824 utils.py:1231] [88800] uptime = 556840.254038875
+I1204 08:30:50.891754 137274321021824 utils.py:1231] [88800] examples_seen = 90931200.0
+I1204 08:30:50.891818 137274321021824 utils.py:1231] [88800] progress = 0.788611315861922
+I1204 08:30:50.891891 137274321021824 utils.py:1231] [88800] epoch = 70.9752904968673
+I1204 08:30:50.891955 137274321021824 utils.py:1231] [88800] img/sec/core = 167.24957002276864
+I1204 08:30:50.892020 137274321021824 utils.py:1231] [88800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.64353074177527
+I1204 08:30:50.892076 137274321021824 utils.py:1231] [88800] core_hours = 154.64353074177527
+I1204 08:30:50.892144 137274321021824 train.py:125] NOTE: Steps:88800/112603 [78.9%]
+Walltime:6d10h40m (0s eval)
+ETA:1d17h27m
+Total train time:8d4h6m
+I1204 08:35:57.005616 137274321021824 utils.py:1231] [88850] l2_params = 248.54410228068772
+I1204 08:35:57.005833 137274321021824 utils.py:1231] [88850] train/loss = 1.7445371448993683
+I1204 08:35:57.005959 137274321021824 utils.py:1231] [88850] l2_grads = 2.247697114944458
+I1204 08:35:57.006036 137274321021824 utils.py:1231] [88850] lr = 0.00012652110912778023
+I1204 08:35:57.006100 137274321021824 utils.py:1231] [88850] uptime = 557146.368461294
+I1204 08:35:57.006166 137274321021824 utils.py:1231] [88850] examples_seen = 90982400.0
+I1204 08:35:57.006225 137274321021824 utils.py:1231] [88850] progress = 0.7890553537649974
+I1204 08:35:57.006285 137274321021824 utils.py:1231] [88850] epoch = 71.01525406133626
+I1204 08:35:57.006345 137274321021824 utils.py:1231] [88850] img/sec/core = 167.25771884710596
+I1204 08:35:57.006409 137274321021824 utils.py:1231] [88850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.72856252578055
+I1204 08:35:57.006471 137274321021824 utils.py:1231] [88850] core_hours = 154.72856252578055
+I1204 08:35:57.006540 137274321021824 train.py:125] NOTE: Steps:88850/112603 [78.9%]
+Walltime:6d10h45m (0s eval)
+ETA:1d17h21m
+Total train time:8d4h5m
+I1204 08:40:59.067843 137274321021824 utils.py:1231] [88900] l2_params = 248.48622843834258
+I1204 08:40:59.068126 137274321021824 utils.py:1231] [88900] train/loss = 1.6219765543937683
+I1204 08:40:59.068294 137274321021824 utils.py:1231] [88900] l2_grads = 2.4724462032318115
+I1204 08:40:59.068362 137274321021824 utils.py:1231] [88900] lr = 0.0001260126059103704
+I1204 08:40:59.068425 137274321021824 utils.py:1231] [88900] uptime = 557448.430782766
+I1204 08:40:59.068478 137274321021824 utils.py:1231] [88900] examples_seen = 91033600.0
+I1204 08:40:59.068527 137274321021824 utils.py:1231] [88900] progress = 0.7894993916680728
+I1204 08:40:59.068574 137274321021824 utils.py:1231] [88900] epoch = 71.05521762580523
+I1204 08:40:59.068633 137274321021824 utils.py:1231] [88900] img/sec/core = 169.50144510078906
+I1204 08:40:59.068693 137274321021824 utils.py:1231] [88900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.81246872618942
+I1204 08:40:59.068744 137274321021824 utils.py:1231] [88900] core_hours = 154.81246872618942
+I1204 08:40:59.068804 137274321021824 train.py:125] NOTE: Steps:88900/112603 [78.9%]
+Walltime:6d10h50m (0s eval)
+ETA:1d17h16m
+Total train time:8d4h5m
+I1204 08:46:01.579761 137274321021824 utils.py:1231] [88950] l2_params = 248.4291852545717
+I1204 08:46:01.580041 137274321021824 utils.py:1231] [88950] train/loss = 3.9037104845046997
+I1204 08:46:01.580225 137274321021824 utils.py:1231] [88950] l2_grads = 2.518913745880127
+I1204 08:46:01.580300 137274321021824 utils.py:1231] [88950] lr = 0.00012550497924259822
+I1204 08:46:01.580354 137274321021824 utils.py:1231] [88950] uptime = 557750.942715954
+I1204 08:46:01.580410 137274321021824 utils.py:1231] [88950] examples_seen = 91084800.0
+I1204 08:46:01.580462 137274321021824 utils.py:1231] [88950] progress = 0.7899434295711482
+I1204 08:46:01.580511 137274321021824 utils.py:1231] [88950] epoch = 71.09518119027418
+I1204 08:46:01.580566 137274321021824 utils.py:1231] [88950] img/sec/core = 169.2495216979396
+I1204 08:46:01.580623 137274321021824 utils.py:1231] [88950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.89649981874166
+I1204 08:46:01.580676 137274321021824 utils.py:1231] [88950] core_hours = 154.89649981874166
+I1204 08:46:01.580737 137274321021824 train.py:125] NOTE: Steps:88950/112603 [79.0%]
+Walltime:6d10h55m (0s eval)
+ETA:1d17h11m
+Total train time:8d4h5m
+I1204 08:51:11.727049 137274321021824 utils.py:1231] [89000] l2_params = 248.37045696190052
+I1204 08:51:11.727305 137274321021824 utils.py:1231] [89000] train/loss = 2.1128815710544586
+I1204 08:51:11.727446 137274321021824 utils.py:1231] [89000] l2_grads = 2.283241033554077
+I1204 08:51:11.727560 137274321021824 utils.py:1231] [89000] lr = 0.0001249982303142363
+I1204 08:51:11.727620 137274321021824 utils.py:1231] [89000] uptime = 558061.089981263
+I1204 08:51:11.727692 137274321021824 utils.py:1231] [89000] examples_seen = 91136000.0
+I1204 08:51:11.727748 137274321021824 utils.py:1231] [89000] progress = 0.7903874674742236
+I1204 08:51:11.727803 137274321021824 utils.py:1231] [89000] epoch = 71.13514475474314
+I1204 08:51:11.727863 137274321021824 utils.py:1231] [89000] img/sec/core = 165.08286780793964
+I1204 08:51:11.727945 137274321021824 utils.py:1231] [89000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 154.98265183688306
+I1204 08:51:11.728004 137274321021824 utils.py:1231] [89000] core_hours = 154.98265183688306
+I1204 08:51:11.728070 137274321021824 train.py:125] NOTE: Steps:89000/112603 [79.0%]
+Walltime:6d11h1m (0s eval)
+ETA:1d17h6m
+Total train time:8d4h5m
+I1204 08:56:18.332353 137274321021824 utils.py:1231] [89050] l2_params = 248.31956586306967
+I1204 08:56:18.332562 137274321021824 utils.py:1231] [89050] train/loss = 3.665409564971924
+I1204 08:56:18.332656 137274321021824 utils.py:1231] [89050] l2_grads = 2.2453010082244873
+I1204 08:56:18.332720 137274321021824 utils.py:1231] [89050] lr = 0.00012449236031299984
+I1204 08:56:18.332782 137274321021824 utils.py:1231] [89050] uptime = 558367.69514492
+I1204 08:56:18.332833 137274321021824 utils.py:1231] [89050] examples_seen = 91187200.0
+I1204 08:56:18.332885 137274321021824 utils.py:1231] [89050] progress = 0.790831505377299
+I1204 08:56:18.332934 137274321021824 utils.py:1231] [89050] epoch = 71.1751083192121
+I1204 08:56:18.332983 137274321021824 utils.py:1231] [89050] img/sec/core = 166.99001213586618
+I1204 08:56:18.333037 137274321021824 utils.py:1231] [89050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.0678199378989
+I1204 08:56:18.333085 137274321021824 utils.py:1231] [89050] core_hours = 155.0678199378989
+I1204 08:56:18.333143 137274321021824 train.py:125] NOTE: Steps:89050/112603 [79.1%]
+Walltime:6d11h6m (0s eval)
+ETA:1d17h0m
+Total train time:8d4h5m
+I1204 09:01:30.120102 137274321021824 utils.py:1231] [89100] l2_params = 248.26095893160777
+I1204 09:01:30.120317 137274321021824 utils.py:1231] [89100] train/loss = 2.231596142053604
+I1204 09:01:30.120410 137274321021824 utils.py:1231] [89100] l2_grads = 2.0176970958709717
+I1204 09:01:30.120478 137274321021824 utils.py:1231] [89100] lr = 0.00012398737042454433
+I1204 09:01:30.120539 137274321021824 utils.py:1231] [89100] uptime = 558679.482901432
+I1204 09:01:30.120594 137274321021824 utils.py:1231] [89100] examples_seen = 91238400.0
+I1204 09:01:30.120643 137274321021824 utils.py:1231] [89100] progress = 0.7912755432803744
+I1204 09:01:30.120692 137274321021824 utils.py:1231] [89100] epoch = 71.21507188368105
+I1204 09:01:30.120742 137274321021824 utils.py:1231] [89100] img/sec/core = 164.21427375080083
+I1204 09:01:30.120803 137274321021824 utils.py:1231] [89100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.15442764804112
+I1204 09:01:30.120857 137274321021824 utils.py:1231] [89100] core_hours = 155.15442764804112
+I1204 09:01:30.120931 137274321021824 train.py:125] NOTE: Steps:89100/112603 [79.1%]
+Walltime:6d11h11m (0s eval)
+ETA:1d16h55m
+Total train time:8d4h5m
+I1204 09:06:34.194615 137274321021824 utils.py:1231] [89150] l2_params = 248.208249618127
+I1204 09:06:34.194812 137274321021824 utils.py:1231] [89150] train/loss = 1.698310375213623
+I1204 09:06:34.194928 137274321021824 utils.py:1231] [89150] l2_grads = 2.3526861667633057
+I1204 09:06:34.195020 137274321021824 utils.py:1231] [89150] lr = 0.00012348326183246237
+I1204 09:06:34.195082 137274321021824 utils.py:1231] [89150] uptime = 558983.557444333
+I1204 09:06:34.195142 137274321021824 utils.py:1231] [89150] examples_seen = 91289600.0
+I1204 09:06:34.195197 137274321021824 utils.py:1231] [89150] progress = 0.7917195811834499
+I1204 09:06:34.195250 137274321021824 utils.py:1231] [89150] epoch = 71.25503544815001
+I1204 09:06:34.195301 137274321021824 utils.py:1231] [89150] img/sec/core = 168.37976474955155
+I1204 09:06:34.195361 137274321021824 utils.py:1231] [89150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.23889279884693
+I1204 09:06:34.195415 137274321021824 utils.py:1231] [89150] core_hours = 155.23889279884693
+I1204 09:06:34.195476 137274321021824 train.py:125] NOTE: Steps:89150/112603 [79.2%]
+Walltime:6d11h16m (0s eval)
+ETA:1d16h50m
+Total train time:8d4h4m
+I1204 09:11:45.980715 137274321021824 utils.py:1231] [89200] l2_params = 248.1555559407784
+I1204 09:11:45.980970 137274321021824 utils.py:1231] [89200] train/loss = 3.9460058212280273
+I1204 09:11:45.981101 137274321021824 utils.py:1231] [89200] l2_grads = 2.434345245361328
+I1204 09:11:45.981188 137274321021824 utils.py:1231] [89200] lr = 0.000122980035718281
+I1204 09:11:45.981261 137274321021824 utils.py:1231] [89200] uptime = 559295.343622984
+I1204 09:11:45.981333 137274321021824 utils.py:1231] [89200] examples_seen = 91340800.0
+I1204 09:11:45.981404 137274321021824 utils.py:1231] [89200] progress = 0.7921636190865252
+I1204 09:11:45.981470 137274321021824 utils.py:1231] [89200] epoch = 71.29499901261896
+I1204 09:11:45.981542 137274321021824 utils.py:1231] [89200] img/sec/core = 164.21510479239575
+I1204 09:11:46.206407 137274321021824 utils.py:1231] [89200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.32550007069446
+I1204 09:11:46.206645 137274321021824 utils.py:1231] [89200] core_hours = 155.32550007069446
+I1204 09:11:46.206739 137274321021824 train.py:125] NOTE: Steps:89200/112603 [79.2%]
+Walltime:6d11h21m (0s eval)
+ETA:1d16h45m
+Total train time:8d4h4m
+I1204 09:16:54.985180 137274321021824 utils.py:1231] [89250] l2_params = 248.1025936859522
+I1204 09:16:54.985424 137274321021824 utils.py:1231] [89250] train/loss = 2.1786050647497177
+I1204 09:16:54.985529 137274321021824 utils.py:1231] [89250] l2_grads = 2.245553731918335
+I1204 09:16:54.985609 137274321021824 utils.py:1231] [89250] lr = 0.00012247769326145886
+I1204 09:16:54.985671 137274321021824 utils.py:1231] [89250] uptime = 559604.348032204
+I1204 09:16:54.985765 137274321021824 utils.py:1231] [89250] examples_seen = 91392000.0
+I1204 09:16:54.985822 137274321021824 utils.py:1231] [89250] progress = 0.7926076569896007
+I1204 09:16:54.985885 137274321021824 utils.py:1231] [89250] epoch = 71.33496257708792
+I1204 09:16:54.985961 137274321021824 utils.py:1231] [89250] img/sec/core = 165.69342854766577
+I1204 09:16:54.986022 137274321021824 utils.py:1231] [89250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.4113346288111
+I1204 09:16:54.986079 137274321021824 utils.py:1231] [89250] core_hours = 155.4113346288111
+I1204 09:16:54.986145 137274321021824 train.py:125] NOTE: Steps:89250/112603 [79.3%]
+Walltime:6d11h26m (0s eval)
+ETA:1d16h39m
+Total train time:8d4h4m
+I1204 09:22:02.807311 137274321021824 utils.py:1231] [89300] l2_params = 248.04654268197453
+I1204 09:22:02.807531 137274321021824 utils.py:1231] [89300] train/loss = 3.8739963471889496
+I1204 09:22:02.807632 137274321021824 utils.py:1231] [89300] l2_grads = 2.3601810932159424
+I1204 09:22:02.807698 137274321021824 utils.py:1231] [89300] lr = 0.00012197623563938353
+I1204 09:22:02.807752 137274321021824 utils.py:1231] [89300] uptime = 559912.170114692
+I1204 09:22:02.807807 137274321021824 utils.py:1231] [89300] examples_seen = 91443200.0
+I1204 09:22:02.807861 137274321021824 utils.py:1231] [89300] progress = 0.793051694892676
+I1204 09:22:02.807920 137274321021824 utils.py:1231] [89300] epoch = 71.37492614155688
+I1204 09:22:02.807975 137274321021824 utils.py:1231] [89300] img/sec/core = 166.3298473786358
+I1204 09:22:02.808035 137274321021824 utils.py:1231] [89300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.49684076283555
+I1204 09:22:02.808088 137274321021824 utils.py:1231] [89300] core_hours = 155.49684076283555
+I1204 09:22:02.808163 137274321021824 train.py:125] NOTE: Steps:89300/112603 [79.3%]
+Walltime:6d11h31m (0s eval)
+ETA:1d16h34m
+Total train time:8d4h4m
+I1204 09:27:09.533816 137274321021824 utils.py:1231] [89350] l2_params = 247.9919456341592
+I1204 09:27:09.534024 137274321021824 utils.py:1231] [89350] train/loss = 3.7317066192626953
+I1204 09:27:09.534111 137274321021824 utils.py:1231] [89350] l2_grads = 2.3280768394470215
+I1204 09:27:09.534170 137274321021824 utils.py:1231] [89350] lr = 0.00012147566402736844
+I1204 09:27:09.534237 137274321021824 utils.py:1231] [89350] uptime = 560218.896597994
+I1204 09:27:09.534292 137274321021824 utils.py:1231] [89350] examples_seen = 91494400.0
+I1204 09:27:09.534340 137274321021824 utils.py:1231] [89350] progress = 0.7934957327957515
+I1204 09:27:09.534386 137274321021824 utils.py:1231] [89350] epoch = 71.41488970602583
+I1204 09:27:09.534446 137274321021824 utils.py:1231] [89350] img/sec/core = 166.92396251150095
+I1204 09:27:09.534507 137274321021824 utils.py:1231] [89350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.58204256375276
+I1204 09:27:09.534559 137274321021824 utils.py:1231] [89350] core_hours = 155.58204256375276
+I1204 09:27:09.534617 137274321021824 train.py:125] NOTE: Steps:89350/112603 [79.3%]
+Walltime:6d11h36m (0s eval)
+ETA:1d16h29m
+Total train time:8d4h4m
+I1204 09:32:21.313937 137274321021824 utils.py:1231] [89400] l2_params = 247.9385276326711
+I1204 09:32:21.314143 137274321021824 utils.py:1231] [89400] train/loss = 3.813335657119751
+I1204 09:32:21.314249 137274321021824 utils.py:1231] [89400] l2_grads = 2.371943950653076
+I1204 09:32:21.314306 137274321021824 utils.py:1231] [89400] lr = 0.00012097597959865098
+I1204 09:32:21.314356 137274321021824 utils.py:1231] [89400] uptime = 560530.676718511
+I1204 09:32:21.314408 137274321021824 utils.py:1231] [89400] examples_seen = 91545600.0
+I1204 09:32:21.314456 137274321021824 utils.py:1231] [89400] progress = 0.7939397706988268
+I1204 09:32:21.314504 137274321021824 utils.py:1231] [89400] epoch = 71.4548532704948
+I1204 09:32:21.314555 137274321021824 utils.py:1231] [89400] img/sec/core = 164.21829562159823
+I1204 09:32:21.314610 137274321021824 utils.py:1231] [89400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.6686481527853
+I1204 09:32:21.314660 137274321021824 utils.py:1231] [89400] core_hours = 155.6686481527853
+I1204 09:32:21.314719 137274321021824 train.py:125] NOTE: Steps:89400/112603 [79.4%]
+Walltime:6d11h42m (0s eval)
+ETA:1d16h24m
+Total train time:8d4h4m
+I1204 09:37:27.752513 137274321021824 utils.py:1231] [89450] l2_params = 247.88606393442936
+I1204 09:37:27.752727 137274321021824 utils.py:1231] [89450] train/loss = 1.6889640241861343
+I1204 09:37:27.752834 137274321021824 utils.py:1231] [89450] l2_grads = 2.324403762817383
+I1204 09:37:27.752914 137274321021824 utils.py:1231] [89450] lr = 0.0001204771835243885
+I1204 09:37:27.752976 137274321021824 utils.py:1231] [89450] uptime = 560837.115337257
+I1204 09:37:27.753036 137274321021824 utils.py:1231] [89450] examples_seen = 91596800.0
+I1204 09:37:27.753093 137274321021824 utils.py:1231] [89450] progress = 0.7943838086019023
+I1204 09:37:27.753149 137274321021824 utils.py:1231] [89450] epoch = 71.49481683496374
+I1204 09:37:27.753210 137274321021824 utils.py:1231] [89450] img/sec/core = 167.08076876706534
+I1204 09:37:27.753272 137274321021824 utils.py:1231] [89450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.75376999132584
+I1204 09:37:27.753330 137274321021824 utils.py:1231] [89450] core_hours = 155.75376999132584
+I1204 09:37:27.753396 137274321021824 train.py:125] NOTE: Steps:89450/112603 [79.4%]
+Walltime:6d11h47m (0s eval)
+ETA:1d16h18m
+Total train time:8d4h4m
+I1204 09:42:33.555919 137274321021824 utils.py:1231] [89500] l2_params = 247.83024298489988
+I1204 09:42:33.556172 137274321021824 utils.py:1231] [89500] train/loss = 3.3814982175827026
+I1204 09:42:33.556293 137274321021824 utils.py:1231] [89500] l2_grads = 2.2326507568359375
+I1204 09:42:33.556367 137274321021824 utils.py:1231] [89500] lr = 0.00011997927697365667
+I1204 09:42:33.556443 137274321021824 utils.py:1231] [89500] uptime = 561142.9188050419
+I1204 09:42:33.556510 137274321021824 utils.py:1231] [89500] examples_seen = 91648000.0
+I1204 09:42:33.556558 137274321021824 utils.py:1231] [89500] progress = 0.7948278465049776
+I1204 09:42:33.556604 137274321021824 utils.py:1231] [89500] epoch = 71.5347803994327
+I1204 09:42:33.556655 137274321021824 utils.py:1231] [89500] img/sec/core = 167.42779397130673
+I1204 09:42:33.556713 137274321021824 utils.py:1231] [89500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.83871539904388
+I1204 09:42:33.556768 137274321021824 utils.py:1231] [89500] core_hours = 155.83871539904388
+I1204 09:42:33.556828 137274321021824 train.py:125] NOTE: Steps:89500/112603 [79.5%]
+Walltime:6d11h52m (0s eval)
+ETA:1d16h13m
+Total train time:8d4h4m
+I1204 09:47:38.745342 137274321021824 utils.py:1231] [89550] l2_params = 247.77565690185793
+I1204 09:47:38.745592 137274321021824 utils.py:1231] [89550] train/loss = 3.466719537973404
+I1204 09:47:38.745715 137274321021824 utils.py:1231] [89550] l2_grads = 2.2807607650756836
+I1204 09:47:38.745793 137274321021824 utils.py:1231] [89550] lr = 0.0001194822611134464
+I1204 09:47:38.745857 137274321021824 utils.py:1231] [89550] uptime = 561448.108216048
+I1204 09:47:38.745931 137274321021824 utils.py:1231] [89550] examples_seen = 91699200.0
+I1204 09:47:38.745988 137274321021824 utils.py:1231] [89550] progress = 0.7952718844080531
+I1204 09:47:38.746038 137274321021824 utils.py:1231] [89550] epoch = 71.57474396390167
+I1204 09:47:38.746089 137274321021824 utils.py:1231] [89550] img/sec/core = 167.76466729700698
+I1204 09:47:38.746149 137274321021824 utils.py:1231] [89550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 155.92349023543446
+I1204 09:47:38.746200 137274321021824 utils.py:1231] [89550] core_hours = 155.92349023543446
+I1204 09:47:38.746265 137274321021824 train.py:125] NOTE: Steps:89550/112603 [79.5%]
+Walltime:6d11h57m (0s eval)
+ETA:1d16h8m
+Total train time:8d4h4m
+I1204 09:52:50.540046 137274321021824 utils.py:1231] [89600] l2_params = 247.72430709535905
+I1204 09:52:50.540274 137274321021824 utils.py:1231] [89600] train/loss = 1.6254772245883942
+I1204 09:52:50.540398 137274321021824 utils.py:1231] [89600] l2_grads = 2.4584829807281494
+I1204 09:52:50.540486 137274321021824 utils.py:1231] [89600] lr = 0.00011898613710866043
+I1204 09:52:50.540537 137274321021824 utils.py:1231] [89600] uptime = 561759.9028991649
+I1204 09:52:50.540592 137274321021824 utils.py:1231] [89600] examples_seen = 91750400.0
+I1204 09:52:50.540640 137274321021824 utils.py:1231] [89600] progress = 0.7957159223111285
+I1204 09:52:50.540689 137274321021824 utils.py:1231] [89600] epoch = 71.61470752837062
+I1204 09:52:50.540738 137274321021824 utils.py:1231] [89600] img/sec/core = 164.21062568534865
+I1204 09:52:50.540791 137274321021824 utils.py:1231] [89600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.0100998696336
+I1204 09:52:50.540840 137274321021824 utils.py:1231] [89600] core_hours = 156.0100998696336
+I1204 09:52:50.540905 137274321021824 train.py:125] NOTE: Steps:89600/112603 [79.6%]
+Walltime:6d12h2m (0s eval)
+ETA:1d16h3m
+Total train time:8d4h4m
+I1204 09:57:58.280288 137274321021824 utils.py:1231] [89650] l2_params = 247.6686922428034
+I1204 09:57:58.280522 137274321021824 utils.py:1231] [89650] train/loss = 1.8936234563589096
+I1204 09:57:58.280644 137274321021824 utils.py:1231] [89650] l2_grads = 2.353776454925537
+I1204 09:57:58.280723 137274321021824 utils.py:1231] [89650] lr = 0.00011849090612211183
+I1204 09:57:58.280783 137274321021824 utils.py:1231] [89650] uptime = 562067.643140119
+I1204 09:57:58.280844 137274321021824 utils.py:1231] [89650] examples_seen = 91801600.0
+I1204 09:57:58.280910 137274321021824 utils.py:1231] [89650] progress = 0.7961599602142039
+I1204 09:57:58.280960 137274321021824 utils.py:1231] [89650] epoch = 71.65467109283958
+I1204 09:57:58.281012 137274321021824 utils.py:1231] [89650] img/sec/core = 166.37408172965368
+I1204 09:57:58.281069 137274321021824 utils.py:1231] [89650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.0955832698986
+I1204 09:57:58.281120 137274321021824 utils.py:1231] [89650] core_hours = 156.0955832698986
+I1204 09:57:58.281182 137274321021824 train.py:125] NOTE: Steps:89650/112603 [79.6%]
+Walltime:6d12h7m (0s eval)
+ETA:1d15h57m
+Total train time:8d4h3m
+I1204 10:03:10.068424 137274321021824 utils.py:1231] [89700] l2_params = 247.61623322446974
+I1204 10:03:10.068629 137274321021824 utils.py:1231] [89700] train/loss = 1.8674969375133514
+I1204 10:03:10.068758 137274321021824 utils.py:1231] [89700] l2_grads = 2.259648084640503
+I1204 10:03:10.068833 137274321021824 utils.py:1231] [89700] lr = 0.0001179965693145201
+I1204 10:03:10.068897 137274321021824 utils.py:1231] [89700] uptime = 562379.431257901
+I1204 10:03:10.068959 137274321021824 utils.py:1231] [89700] examples_seen = 91852800.0
+I1204 10:03:10.069018 137274321021824 utils.py:1231] [89700] progress = 0.7966039981172793
+I1204 10:03:10.069072 137274321021824 utils.py:1231] [89700] epoch = 71.69463465730853
+I1204 10:03:10.069130 137274321021824 utils.py:1231] [89700] img/sec/core = 164.21408347511806
+I1204 10:03:10.069189 137274321021824 utils.py:1231] [89700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.1821910803936
+I1204 10:03:10.069245 137274321021824 utils.py:1231] [89700] core_hours = 156.1821910803936
+I1204 10:03:10.069312 137274321021824 train.py:125] NOTE: Steps:89700/112603 [79.7%]
+Walltime:6d12h12m (0s eval)
+ETA:1d15h52m
+Total train time:8d4h3m
+I1204 10:08:16.217305 137274321021824 utils.py:1231] [89750] l2_params = 247.5627406637259
+I1204 10:08:16.217513 137274321021824 utils.py:1231] [89750] train/loss = 1.7325201630592346
+I1204 10:08:16.217620 137274321021824 utils.py:1231] [89750] l2_grads = 2.402848243713379
+I1204 10:08:16.217719 137274321021824 utils.py:1231] [89750] lr = 0.00011750312784450915
+I1204 10:08:16.217786 137274321021824 utils.py:1231] [89750] uptime = 562685.580148173
+I1204 10:08:16.217849 137274321021824 utils.py:1231] [89750] examples_seen = 91904000.0
+I1204 10:08:16.217922 137274321021824 utils.py:1231] [89750] progress = 0.7970480360203547
+I1204 10:08:16.217979 137274321021824 utils.py:1231] [89750] epoch = 71.73459822177749
+I1204 10:08:16.218035 137274321021824 utils.py:1231] [89750] img/sec/core = 167.2388880930073
+I1204 10:08:16.218094 137274321021824 utils.py:1231] [89750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.26723243880252
+I1204 10:08:16.218152 137274321021824 utils.py:1231] [89750] core_hours = 156.26723243880252
+I1204 10:08:16.218234 137274321021824 train.py:125] NOTE: Steps:89750/112603 [79.7%]
+Walltime:6d12h18m (0s eval)
+ETA:1d15h47m
+Total train time:8d4h3m
+I1204 10:13:27.995939 137274321021824 utils.py:1231] [89800] l2_params = 247.50760270516776
+I1204 10:13:27.996152 137274321021824 utils.py:1231] [89800] train/loss = 1.8519396036863327
+I1204 10:13:27.996251 137274321021824 utils.py:1231] [89800] l2_grads = 2.444427251815796
+I1204 10:13:27.996320 137274321021824 utils.py:1231] [89800] lr = 0.00011701058286860469
+I1204 10:13:27.996378 137274321021824 utils.py:1231] [89800] uptime = 562997.358740112
+I1204 10:13:27.996440 137274321021824 utils.py:1231] [89800] examples_seen = 91955200.0
+I1204 10:13:27.996494 137274321021824 utils.py:1231] [89800] progress = 0.7974920739234301
+I1204 10:13:27.996548 137274321021824 utils.py:1231] [89800] epoch = 71.77456178624645
+I1204 10:13:27.996602 137274321021824 utils.py:1231] [89800] img/sec/core = 164.21910074574603
+I1204 10:13:27.996661 137274321021824 utils.py:1231] [89800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.35383760323003
+I1204 10:13:27.996718 137274321021824 utils.py:1231] [89800] core_hours = 156.35383760323003
+I1204 10:13:27.996782 137274321021824 train.py:125] NOTE: Steps:89800/112603 [79.7%]
+Walltime:6d12h23m (0s eval)
+ETA:1d15h42m
+Total train time:8d4h3m
+I1204 10:18:35.704411 137274321021824 utils.py:1231] [89850] l2_params = 247.45493833717265
+I1204 10:18:35.704623 137274321021824 utils.py:1231] [89850] train/loss = 1.9129635393619537
+I1204 10:18:35.704722 137274321021824 utils.py:1231] [89850] l2_grads = 2.5499045848846436
+I1204 10:18:35.704789 137274321021824 utils.py:1231] [89850] lr = 0.00011651893554123069
+I1204 10:18:35.704847 137274321021824 utils.py:1231] [89850] uptime = 563305.06720921
+I1204 10:18:35.704911 137274321021824 utils.py:1231] [89850] examples_seen = 92006400.0
+I1204 10:18:35.704967 137274321021824 utils.py:1231] [89850] progress = 0.7979361118265055
+I1204 10:18:35.705020 137274321021824 utils.py:1231] [89850] epoch = 71.8145253507154
+I1204 10:18:35.705075 137274321021824 utils.py:1231] [89850] img/sec/core = 166.39126037086405
+I1204 10:18:35.705136 137274321021824 utils.py:1231] [89850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.43931217797945
+I1204 10:18:35.705186 137274321021824 utils.py:1231] [89850] core_hours = 156.43931217797945
+I1204 10:18:35.705247 137274321021824 train.py:125] NOTE: Steps:89850/112603 [79.8%]
+Walltime:6d12h28m (0s eval)
+ETA:1d15h36m
+Total train time:8d4h3m
+I1204 10:23:46.507183 137274321021824 utils.py:1231] [89900] l2_params = 247.40226880291726
+I1204 10:23:46.507425 137274321021824 utils.py:1231] [89900] train/loss = 2.848644882440567
+I1204 10:23:46.507550 137274321021824 utils.py:1231] [89900] l2_grads = 2.233883857727051
+I1204 10:23:46.507631 137274321021824 utils.py:1231] [89900] lr = 0.00011602818701470787
+I1204 10:23:46.507702 137274321021824 utils.py:1231] [89900] uptime = 563615.87006023
+I1204 10:23:46.507768 137274321021824 utils.py:1231] [89900] examples_seen = 92057600.0
+I1204 10:23:46.507829 137274321021824 utils.py:1231] [89900] progress = 0.7983801497295809
+I1204 10:23:46.507890 137274321021824 utils.py:1231] [89900] epoch = 71.85448891518436
+I1204 10:23:46.507946 137274321021824 utils.py:1231] [89900] img/sec/core = 164.73465359783262
+I1204 10:23:46.508005 137274321021824 utils.py:1231] [89900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.5256463032628
+I1204 10:23:46.508054 137274321021824 utils.py:1231] [89900] core_hours = 156.5256463032628
+I1204 10:23:46.508115 137274321021824 train.py:125] NOTE: Steps:89900/112603 [79.8%]
+Walltime:6d12h33m (0s eval)
+ETA:1d15h31m
+Total train time:8d4h3m
+I1204 10:28:56.279403 137274321021824 utils.py:1231] [89950] l2_params = 247.35085941298027
+I1204 10:28:56.279655 137274321021824 utils.py:1231] [89950] train/loss = 1.8581188768148422
+I1204 10:28:56.279761 137274321021824 utils.py:1231] [89950] l2_grads = 2.3953564167022705
+I1204 10:28:56.279831 137274321021824 utils.py:1231] [89950] lr = 0.00011553833843924952
+I1204 10:28:56.279897 137274321021824 utils.py:1231] [89950] uptime = 563925.6422513049
+I1204 10:28:56.279958 137274321021824 utils.py:1231] [89950] examples_seen = 92108800.0
+I1204 10:28:56.280016 137274321021824 utils.py:1231] [89950] progress = 0.7988241876326563
+I1204 10:28:56.280076 137274321021824 utils.py:1231] [89950] epoch = 71.89445247965331
+I1204 10:28:56.280135 137274321021824 utils.py:1231] [89950] img/sec/core = 165.28275124482923
+I1204 10:28:56.280193 137274321021824 utils.py:1231] [89950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.61169413411693
+I1204 10:28:56.280249 137274321021824 utils.py:1231] [89950] core_hours = 156.61169413411693
+I1204 10:28:56.280316 137274321021824 train.py:125] NOTE: Steps:89950/112603 [79.9%]
+Walltime:6d12h38m (0s eval)
+ETA:1d15h26m
+Total train time:8d4h3m
+I1204 10:34:05.495860 137274321021824 utils.py:1231] [90000] l2_params = 247.29811495680198
+I1204 10:34:05.496091 137274321021824 utils.py:1231] [90000] train/loss = 1.9639144092798233
+I1204 10:34:05.496199 137274321021824 utils.py:1231] [90000] l2_grads = 2.4027934074401855
+I1204 10:34:05.496292 137274321021824 utils.py:1231] [90000] lr = 0.00011504939096296039
+I1204 10:34:05.496353 137274321021824 utils.py:1231] [90000] uptime = 564234.858710621
+I1204 10:34:05.496411 137274321021824 utils.py:1231] [90000] examples_seen = 92160000.0
+I1204 10:34:05.496464 137274321021824 utils.py:1231] [90000] progress = 0.7992682255357317
+I1204 10:34:05.496518 137274321021824 utils.py:1231] [90000] epoch = 71.93441604412227
+I1204 10:34:05.496573 137274321021824 utils.py:1231] [90000] img/sec/core = 165.5798016484714
+I1204 10:34:05.496628 137274321021824 utils.py:1231] [90000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.69758759503804
+I1204 10:34:05.496691 137274321021824 utils.py:1231] [90000] core_hours = 156.69758759503804
+I1204 10:34:05.496773 137274321021824 train.py:125] NOTE: Steps:90000/112603 [79.9%]
+Walltime:6d12h43m (0s eval)
+ETA:1d15h21m
+Total train time:8d4h3m
+I1204 10:34:05.823884 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:90000/112603 [79.9%]
+Walltime:6d12h43m (0s eval)
+ETA:1d15h21m
+Total train time:8d4h3m
+I1204 10:35:40.563125 137274321021824 utils.py:1231] [90000] val/acc@1 = 0.7438815369897959
+I1204 10:35:40.563360 137274321021824 utils.py:1231] [90000] val/loss = 1.0086827202110875
+I1204 10:35:40.563553 137274321021824 utils.py:1231] [90000] z/secs/eval/val = 94.73942872905172
+I1204 10:35:40.563648 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 94.73942872905172
+I1204 10:40:51.469060 137274321021824 utils.py:1231] [90050] l2_params = 247.2470282980164
+I1204 10:40:51.469308 137274321021824 utils.py:1231] [90050] train/loss = 2.816767603158951
+I1204 10:40:51.469422 137274321021824 utils.py:1231] [90050] l2_grads = 2.1723923683166504
+I1204 10:40:51.469491 137274321021824 utils.py:1231] [90050] lr = 0.00011456134573183291
+I1204 10:40:51.469552 137274321021824 utils.py:1231] [90050] uptime = 564640.831914146
+I1204 10:40:51.469606 137274321021824 utils.py:1231] [90050] examples_seen = 92211200.0
+I1204 10:40:51.469653 137274321021824 utils.py:1231] [90050] progress = 0.7997122634388072
+I1204 10:40:51.469699 137274321021824 utils.py:1231] [90050] epoch = 71.97437960859124
+I1204 10:40:51.469748 137274321021824 utils.py:1231] [90050] img/sec/core = 126.11669823388662
+I1204 10:40:51.469801 137274321021824 utils.py:1231] [90050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.81035792935054
+I1204 10:40:51.469858 137274321021824 utils.py:1231] [90050] core_hours = 156.81035792935054
+I1204 10:40:51.469928 137274321021824 train.py:125] NOTE: Steps:90050/112603 [80.0%]
+Walltime:6d12h50m (0s eval)
+ETA:1d15h16m
+Total train time:8d4h5m
+I1204 10:46:01.464160 137274321021824 utils.py:1231] [90100] l2_params = 247.19389051213622
+I1204 10:46:01.464404 137274321021824 utils.py:1231] [90100] train/loss = 1.705420270562172
+I1204 10:46:01.464508 137274321021824 utils.py:1231] [90100] l2_grads = 2.5202698707580566
+I1204 10:46:01.464571 137274321021824 utils.py:1231] [90100] lr = 0.00011407420388974492
+I1204 10:46:01.464641 137274321021824 utils.py:1231] [90100] uptime = 564950.82699939
+I1204 10:46:01.464718 137274321021824 utils.py:1231] [90100] examples_seen = 92262400.0
+I1204 10:46:01.464774 137274321021824 utils.py:1231] [90100] progress = 0.8001563013418825
+I1204 10:46:01.464826 137274321021824 utils.py:1231] [90100] epoch = 72.01434317306018
+I1204 10:46:01.464878 137274321021824 utils.py:1231] [90100] img/sec/core = 165.16390883968026
+I1204 10:46:01.464946 137274321021824 utils.py:1231] [90100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.89646767525167
+I1204 10:46:01.464998 137274321021824 utils.py:1231] [90100] core_hours = 156.89646767525167
+I1204 10:46:01.465063 137274321021824 train.py:125] NOTE: Steps:90100/112603 [80.0%]
+Walltime:6d12h55m (0s eval)
+ETA:1d15h11m
+Total train time:8d4h5m
+I1204 10:51:13.205582 137274321021824 utils.py:1231] [90150] l2_params = 247.14314288658133
+I1204 10:51:13.205813 137274321021824 utils.py:1231] [90150] train/loss = 1.6211534142494202
+I1204 10:51:13.205921 137274321021824 utils.py:1231] [90150] l2_grads = 2.4478085041046143
+I1204 10:51:13.205993 137274321021824 utils.py:1231] [90150] lr = 0.00011358796657845665
+I1204 10:51:13.206062 137274321021824 utils.py:1231] [90150] uptime = 565262.568423157
+I1204 10:51:13.206123 137274321021824 utils.py:1231] [90150] examples_seen = 92313600.0
+I1204 10:51:13.206181 137274321021824 utils.py:1231] [90150] progress = 0.800600339244958
+I1204 10:51:13.206237 137274321021824 utils.py:1231] [90150] epoch = 72.05430673752915
+I1204 10:51:13.206295 137274321021824 utils.py:1231] [90150] img/sec/core = 164.23868018988992
+I1204 10:51:13.206362 137274321021824 utils.py:1231] [90150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 156.98306251518696
+I1204 10:51:13.206454 137274321021824 utils.py:1231] [90150] core_hours = 156.98306251518696
+I1204 10:51:13.206522 137274321021824 train.py:125] NOTE: Steps:90150/112603 [80.1%]
+Walltime:6d13h1m (0s eval)
+ETA:1d15h5m
+Total train time:8d4h5m
+I1204 10:56:21.747322 137274321021824 utils.py:1231] [90200] l2_params = 247.09062230378976
+I1204 10:56:21.747578 137274321021824 utils.py:1231] [90200] train/loss = 1.6836562901735306
+I1204 10:56:21.747734 137274321021824 utils.py:1231] [90200] l2_grads = 2.404118537902832
+I1204 10:56:21.747853 137274321021824 utils.py:1231] [90200] lr = 0.00011310263493760858
+I1204 10:56:21.747947 137274321021824 utils.py:1231] [90200] uptime = 565571.110303573
+I1204 10:56:21.748039 137274321021824 utils.py:1231] [90200] examples_seen = 92364800.0
+I1204 10:56:21.748133 137274321021824 utils.py:1231] [90200] progress = 0.8010443771480333
+I1204 10:56:21.748218 137274321021824 utils.py:1231] [90200] epoch = 72.0942703019981
+I1204 10:56:21.748300 137274321021824 utils.py:1231] [90200] img/sec/core = 165.941816167616
+I1204 10:56:21.748383 137274321021824 utils.py:1231] [90200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.06876859308025
+I1204 10:56:21.748450 137274321021824 utils.py:1231] [90200] core_hours = 157.06876859308025
+I1204 10:56:21.748522 137274321021824 train.py:125] NOTE: Steps:90200/112603 [80.1%]
+Walltime:6d13h6m (0s eval)
+ETA:1d15h0m
+Total train time:8d4h5m
+I1204 11:01:29.146594 137274321021824 utils.py:1231] [90250] l2_params = 247.0367160153786
+I1204 11:01:29.146798 137274321021824 utils.py:1231] [90250] train/loss = 1.945885807275772
+I1204 11:01:29.146909 137274321021824 utils.py:1231] [90250] l2_grads = 2.370171308517456
+I1204 11:01:29.147002 137274321021824 utils.py:1231] [90250] lr = 0.0001126182101047184
+I1204 11:01:29.147091 137274321021824 utils.py:1231] [90250] uptime = 565878.509443566
+I1204 11:01:29.147164 137274321021824 utils.py:1231] [90250] examples_seen = 92416000.0
+I1204 11:01:29.147216 137274321021824 utils.py:1231] [90250] progress = 0.8014884150511088
+I1204 11:01:29.147263 137274321021824 utils.py:1231] [90250] epoch = 72.13423386646706
+I1204 11:01:29.147312 137274321021824 utils.py:1231] [90250] img/sec/core = 166.55869629682238
+I1204 11:01:29.147367 137274321021824 utils.py:1231] [90250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.15415724307832
+I1204 11:01:29.147417 137274321021824 utils.py:1231] [90250] core_hours = 157.15415724307832
+I1204 11:01:29.147477 137274321021824 train.py:125] NOTE: Steps:90250/112603 [80.1%]
+Walltime:6d13h11m (0s eval)
+ETA:1d14h55m
+Total train time:8d4h4m
+I1204 11:06:29.955861 137274321021824 utils.py:1231] [90300] l2_params = 246.98713881922168
+I1204 11:06:29.956129 137274321021824 utils.py:1231] [90300] train/loss = 3.2970772087574005
+I1204 11:06:29.956249 137274321021824 utils.py:1231] [90300] l2_grads = 2.3176486492156982
+I1204 11:06:29.956335 137274321021824 utils.py:1231] [90300] lr = 0.0001121346932151785
+I1204 11:06:29.956403 137274321021824 utils.py:1231] [90300] uptime = 566179.318764703
+I1204 11:06:29.956469 137274321021824 utils.py:1231] [90300] examples_seen = 92467200.0
+I1204 11:06:29.956524 137274321021824 utils.py:1231] [90300] progress = 0.8019324529541841
+I1204 11:06:29.956575 137274321021824 utils.py:1231] [90300] epoch = 72.17419743093602
+I1204 11:06:29.956628 137274321021824 utils.py:1231] [90300] img/sec/core = 170.20749159789727
+I1204 11:06:29.956691 137274321021824 utils.py:1231] [90300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.23771538783862
+I1204 11:06:29.956744 137274321021824 utils.py:1231] [90300] core_hours = 157.23771538783862
+I1204 11:06:29.956806 137274321021824 train.py:125] NOTE: Steps:90300/112603 [80.2%]
+Walltime:6d13h16m (0s eval)
+ETA:1d14h50m
+Total train time:8d4h4m
+I1204 11:11:31.542338 137274321021824 utils.py:1231] [90350] l2_params = 246.9338984702313
+I1204 11:11:31.542645 137274321021824 utils.py:1231] [90350] train/loss = 1.6256300956010818
+I1204 11:11:31.542852 137274321021824 utils.py:1231] [90350] l2_grads = 2.492628335952759
+I1204 11:11:31.543026 137274321021824 utils.py:1231] [90350] lr = 0.00011165208540225316
+I1204 11:11:31.543127 137274321021824 utils.py:1231] [90350] uptime = 566480.905482146
+I1204 11:11:31.543229 137274321021824 utils.py:1231] [90350] examples_seen = 92518400.0
+I1204 11:11:31.543320 137274321021824 utils.py:1231] [90350] progress = 0.8023764908572596
+I1204 11:11:31.543388 137274321021824 utils.py:1231] [90350] epoch = 72.21416099540497
+I1204 11:11:31.543459 137274321021824 utils.py:1231] [90350] img/sec/core = 169.7687498776661
+I1204 11:11:31.543552 137274321021824 utils.py:1231] [90350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.3214894760172
+I1204 11:11:31.543623 137274321021824 utils.py:1231] [90350] core_hours = 157.3214894760172
+I1204 11:11:31.543702 137274321021824 train.py:125] NOTE: Steps:90350/112603 [80.2%]
+Walltime:6d13h21m (0s eval)
+ETA:1d14h44m
+Total train time:8d4h4m
+I1204 11:16:43.308269 137274321021824 utils.py:1231] [90400] l2_params = 246.88376287118848
+I1204 11:16:43.308561 137274321021824 utils.py:1231] [90400] train/loss = 1.5604118406772614
+I1204 11:16:43.308712 137274321021824 utils.py:1231] [90400] l2_grads = 2.405320644378662
+I1204 11:16:43.308795 137274321021824 utils.py:1231] [90400] lr = 0.00011117038779707575
+I1204 11:16:43.308864 137274321021824 utils.py:1231] [90400] uptime = 566792.671224171
+I1204 11:16:43.308940 137274321021824 utils.py:1231] [90400] examples_seen = 92569600.0
+I1204 11:16:43.309005 137274321021824 utils.py:1231] [90400] progress = 0.8028205287603349
+I1204 11:16:43.309066 137274321021824 utils.py:1231] [90400] epoch = 72.25412455987393
+I1204 11:16:43.309131 137274321021824 utils.py:1231] [90400] img/sec/core = 164.22586929354213
+I1204 11:16:43.309202 137274321021824 utils.py:1231] [90400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.40809107102413
+I1204 11:16:43.309265 137274321021824 utils.py:1231] [90400] core_hours = 157.40809107102413
+I1204 11:16:43.309338 137274321021824 train.py:125] NOTE: Steps:90400/112603 [80.3%]
+Walltime:6d13h26m (0s eval)
+ETA:1d14h39m
+Total train time:8d4h4m
+I1204 11:21:55.076311 137274321021824 utils.py:1231] [90450] l2_params = 246.83551361342188
+I1204 11:21:55.076551 137274321021824 utils.py:1231] [90450] train/loss = 2.150150030851364
+I1204 11:21:55.076662 137274321021824 utils.py:1231] [90450] l2_grads = 2.240137815475464
+I1204 11:21:55.076733 137274321021824 utils.py:1231] [90450] lr = 0.00011068960152864657
+I1204 11:21:55.076809 137274321021824 utils.py:1231] [90450] uptime = 567104.439165981
+I1204 11:21:55.076893 137274321021824 utils.py:1231] [90450] examples_seen = 92620800.0
+I1204 11:21:55.076959 137274321021824 utils.py:1231] [90450] progress = 0.8032645666634104
+I1204 11:21:55.077022 137274321021824 utils.py:1231] [90450] epoch = 72.29408812434288
+I1204 11:21:55.077079 137274321021824 utils.py:1231] [90450] img/sec/core = 164.2247105418892
+I1204 11:21:55.077146 137274321021824 utils.py:1231] [90450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.49469327708252
+I1204 11:21:55.077204 137274321021824 utils.py:1231] [90450] core_hours = 157.49469327708252
+I1204 11:21:55.077271 137274321021824 train.py:125] NOTE: Steps:90450/112603 [80.3%]
+Walltime:6d13h31m (0s eval)
+ETA:1d14h34m
+Total train time:8d4h4m
+I1204 11:27:03.282905 137274321021824 utils.py:1231] [90500] l2_params = 246.7838486949688
+I1204 11:27:03.283142 137274321021824 utils.py:1231] [90500] train/loss = 1.7955121397972107
+I1204 11:27:03.283284 137274321021824 utils.py:1231] [90500] l2_grads = 2.562234878540039
+I1204 11:27:03.283376 137274321021824 utils.py:1231] [90500] lr = 0.00011020972772383015
+I1204 11:27:03.283454 137274321021824 utils.py:1231] [90500] uptime = 567412.645810917
+I1204 11:27:03.283538 137274321021824 utils.py:1231] [90500] examples_seen = 92672000.0
+I1204 11:27:03.283609 137274321021824 utils.py:1231] [90500] progress = 0.8037086045664857
+I1204 11:27:03.283686 137274321021824 utils.py:1231] [90500] epoch = 72.33405168881184
+I1204 11:27:03.283748 137274321021824 utils.py:1231] [90500] img/sec/core = 166.12231060310845
+I1204 11:27:03.283828 137274321021824 utils.py:1231] [90500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.58030623400916
+I1204 11:27:03.283900 137274321021824 utils.py:1231] [90500] core_hours = 157.58030623400916
+I1204 11:27:03.283980 137274321021824 train.py:125] NOTE: Steps:90500/112603 [80.4%]
+Walltime:6d13h36m (0s eval)
+ETA:1d14h29m
+Total train time:8d4h4m
+I1204 11:32:10.310231 137274321021824 utils.py:1231] [90550] l2_params = 246.7324359965694
+I1204 11:32:10.310473 137274321021824 utils.py:1231] [90550] train/loss = 1.619974061846733
+I1204 11:32:10.310603 137274321021824 utils.py:1231] [90550] l2_grads = 2.4697699546813965
+I1204 11:32:10.310676 137274321021824 utils.py:1231] [90550] lr = 0.00010973076750735175
+I1204 11:32:10.310756 137274321021824 utils.py:1231] [90550] uptime = 567719.673117243
+I1204 11:32:10.310841 137274321021824 utils.py:1231] [90550] examples_seen = 92723200.0
+I1204 11:32:10.310906 137274321021824 utils.py:1231] [90550] progress = 0.8041526424695612
+I1204 11:32:10.310964 137274321021824 utils.py:1231] [90550] epoch = 72.3740152532808
+I1204 11:32:10.311021 137274321021824 utils.py:1231] [90550] img/sec/core = 166.76041168026174
+I1204 11:32:10.311083 137274321021824 utils.py:1231] [90550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.6655915968775
+I1204 11:32:10.311139 137274321021824 utils.py:1231] [90550] core_hours = 157.6655915968775
+I1204 11:32:10.311203 137274321021824 train.py:125] NOTE: Steps:90550/112603 [80.4%]
+Walltime:6d13h41m (0s eval)
+ETA:1d14h23m
+Total train time:8d4h4m
+I1204 11:37:21.391388 137274321021824 utils.py:1231] [90600] l2_params = 246.68342538087205
+I1204 11:37:21.391590 137274321021824 utils.py:1231] [90600] train/loss = 2.814419686794281
+I1204 11:37:21.391684 137274321021824 utils.py:1231] [90600] l2_grads = 2.3113210201263428
+I1204 11:37:21.391743 137274321021824 utils.py:1231] [90600] lr = 0.00010925272200179608
+I1204 11:37:21.391793 137274321021824 utils.py:1231] [90600] uptime = 568030.754155537
+I1204 11:37:21.391860 137274321021824 utils.py:1231] [90600] examples_seen = 92774400.0
+I1204 11:37:21.391916 137274321021824 utils.py:1231] [90600] progress = 0.8045966803726367
+I1204 11:37:21.391963 137274321021824 utils.py:1231] [90600] epoch = 72.41397881774975
+I1204 11:37:21.392012 137274321021824 utils.py:1231] [90600] img/sec/core = 164.58733801579234
+I1204 11:37:21.392065 137274321021824 utils.py:1231] [90600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.7520029964036
+I1204 11:37:21.392116 137274321021824 utils.py:1231] [90600] core_hours = 157.7520029964036
+I1204 11:37:21.392174 137274321021824 train.py:125] NOTE: Steps:90600/112603 [80.5%]
+Walltime:6d13h47m (0s eval)
+ETA:1d14h18m
+Total train time:8d4h4m
+I1204 11:42:32.383518 137274321021824 utils.py:1231] [90650] l2_params = 246.63225286277017
+I1204 11:42:32.383751 137274321021824 utils.py:1231] [90650] train/loss = 2.0625618398189545
+I1204 11:42:32.383934 137274321021824 utils.py:1231] [90650] l2_grads = 2.3294384479522705
+I1204 11:42:32.384055 137274321021824 utils.py:1231] [90650] lr = 0.00010877559232760339
+I1204 11:42:32.384131 137274321021824 utils.py:1231] [90650] uptime = 568341.746493037
+I1204 11:42:32.384207 137274321021824 utils.py:1231] [90650] examples_seen = 92825600.0
+I1204 11:42:32.384255 137274321021824 utils.py:1231] [90650] progress = 0.805040718275712
+I1204 11:42:32.384303 137274321021824 utils.py:1231] [90650] epoch = 72.45394238221871
+I1204 11:42:32.384354 137274321021824 utils.py:1231] [90650] img/sec/core = 164.63428138323061
+I1204 11:42:32.384415 137274321021824 utils.py:1231] [90650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.83838975682025
+I1204 11:42:32.384465 137274321021824 utils.py:1231] [90650] core_hours = 157.83838975682025
+I1204 11:42:32.384526 137274321021824 train.py:125] NOTE: Steps:90650/112603 [80.5%]
+Walltime:6d13h52m (0s eval)
+ETA:1d14h13m
+Total train time:8d4h4m
+I1204 11:47:43.333395 137274321021824 utils.py:1231] [90700] l2_params = 246.58150208913094
+I1204 11:47:43.333626 137274321021824 utils.py:1231] [90700] train/loss = 1.614901825785637
+I1204 11:47:43.333739 137274321021824 utils.py:1231] [90700] l2_grads = 2.3199379444122314
+I1204 11:47:43.333810 137274321021824 utils.py:1231] [90700] lr = 0.00010829937960306777
+I1204 11:47:43.333868 137274321021824 utils.py:1231] [90700] uptime = 568652.696229983
+I1204 11:47:43.333949 137274321021824 utils.py:1231] [90700] examples_seen = 92876800.0
+I1204 11:47:43.334009 137274321021824 utils.py:1231] [90700] progress = 0.8054847561787875
+I1204 11:47:43.334065 137274321021824 utils.py:1231] [90700] epoch = 72.49390594668766
+I1204 11:47:43.334120 137274321021824 utils.py:1231] [90700] img/sec/core = 164.6568365127339
+I1204 11:47:43.334177 137274321021824 utils.py:1231] [90700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 157.92476468374971
+I1204 11:47:43.334232 137274321021824 utils.py:1231] [90700] core_hours = 157.92476468374971
+I1204 11:47:43.334297 137274321021824 train.py:125] NOTE: Steps:90700/112603 [80.5%]
+Walltime:6d13h57m (0s eval)
+ETA:1d14h8m
+Total train time:8d4h3m
+I1204 11:52:54.570789 137274321021824 utils.py:1231] [90750] l2_params = 246.53122615439347
+I1204 11:52:54.571003 137274321021824 utils.py:1231] [90750] train/loss = 3.8674599528312683
+I1204 11:52:54.571108 137274321021824 utils.py:1231] [90750] l2_grads = 2.499567747116089
+I1204 11:52:54.571187 137274321021824 utils.py:1231] [90750] lr = 0.00010782408494433412
+I1204 11:52:54.571249 137274321021824 utils.py:1231] [90750] uptime = 568963.933610397
+I1204 11:52:54.571310 137274321021824 utils.py:1231] [90750] examples_seen = 92928000.0
+I1204 11:52:54.571367 137274321021824 utils.py:1231] [90750] progress = 0.8059287940818628
+I1204 11:52:54.571426 137274321021824 utils.py:1231] [90750] epoch = 72.53386951115662
+I1204 11:52:54.571485 137274321021824 utils.py:1231] [90750] img/sec/core = 164.50466178544806
+I1204 11:52:54.571549 137274321021824 utils.py:1231] [90750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.0112195116425
+I1204 11:52:54.571607 137274321021824 utils.py:1231] [90750] core_hours = 158.0112195116425
+I1204 11:52:54.571678 137274321021824 train.py:125] NOTE: Steps:90750/112603 [80.6%]
+Walltime:6d14h2m (0s eval)
+ETA:1d14h3m
+Total train time:8d4h3m
+I1204 11:58:05.882764 137274321021824 utils.py:1231] [90800] l2_params = 246.48271631471684
+I1204 11:58:05.882986 137274321021824 utils.py:1231] [90800] train/loss = 1.6378155648708344
+I1204 11:58:05.883092 137274321021824 utils.py:1231] [90800] l2_grads = 2.4482433795928955
+I1204 11:58:05.883162 137274321021824 utils.py:1231] [90800] lr = 0.00010734970946539514
+I1204 11:58:05.883223 137274321021824 utils.py:1231] [90800] uptime = 569275.245584076
+I1204 11:58:05.883283 137274321021824 utils.py:1231] [90800] examples_seen = 92979200.0
+I1204 11:58:05.883340 137274321021824 utils.py:1231] [90800] progress = 0.8063728319849383
+I1204 11:58:05.883403 137274321021824 utils.py:1231] [90800] epoch = 72.57383307562559
+I1204 11:58:05.883479 137274321021824 utils.py:1231] [90800] img/sec/core = 164.4652449275659
+I1204 11:58:05.883541 137274321021824 utils.py:1231] [90800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.09769505988666
+I1204 11:58:05.883600 137274321021824 utils.py:1231] [90800] core_hours = 158.09769505988666
+I1204 11:58:05.883680 137274321021824 train.py:125] NOTE: Steps:90800/112603 [80.6%]
+Walltime:6d14h7m (0s eval)
+ETA:1d13h57m
+Total train time:8d4h3m
+I1204 12:03:17.021429 137274321021824 utils.py:1231] [90850] l2_params = 246.43362601290266
+I1204 12:03:17.021646 137274321021824 utils.py:1231] [90850] train/loss = 3.3808528780937195
+I1204 12:03:17.021747 137274321021824 utils.py:1231] [90850] l2_grads = 2.346708059310913
+I1204 12:03:17.021814 137274321021824 utils.py:1231] [90850] lr = 0.00010687625427808997
+I1204 12:03:17.021876 137274321021824 utils.py:1231] [90850] uptime = 569586.384238049
+I1204 12:03:17.021941 137274321021824 utils.py:1231] [90850] examples_seen = 93030400.0
+I1204 12:03:17.021996 137274321021824 utils.py:1231] [90850] progress = 0.8068168698880136
+I1204 12:03:17.022051 137274321021824 utils.py:1231] [90850] epoch = 72.61379664009453
+I1204 12:03:17.022120 137274321021824 utils.py:1231] [90850] img/sec/core = 164.5568602493307
+I1204 12:03:17.022178 137274321021824 utils.py:1231] [90850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.1841224637681
+I1204 12:03:17.022231 137274321021824 utils.py:1231] [90850] core_hours = 158.1841224637681
+I1204 12:03:17.022293 137274321021824 train.py:125] NOTE: Steps:90850/112603 [80.7%]
+Walltime:6d14h13m (0s eval)
+ETA:1d13h52m
+Total train time:8d4h3m
+I1204 12:08:28.153872 137274321021824 utils.py:1231] [90900] l2_params = 246.38109247352548
+I1204 12:08:28.154114 137274321021824 utils.py:1231] [90900] train/loss = 1.6718323677778244
+I1204 12:08:28.154216 137274321021824 utils.py:1231] [90900] l2_grads = 2.2869606018066406
+I1204 12:08:28.154284 137274321021824 utils.py:1231] [90900] lr = 0.00010640372049209992
+I1204 12:08:28.154345 137274321021824 utils.py:1231] [90900] uptime = 569897.516706264
+I1204 12:08:28.154405 137274321021824 utils.py:1231] [90900] examples_seen = 93081600.0
+I1204 12:08:28.154464 137274321021824 utils.py:1231] [90900] progress = 0.807260907791089
+I1204 12:08:28.154520 137274321021824 utils.py:1231] [90900] epoch = 72.6537602045635
+I1204 12:08:28.154586 137274321021824 utils.py:1231] [90900] img/sec/core = 164.5601318748471
+I1204 12:08:28.154650 137274321021824 utils.py:1231] [90900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.2705481493833
+I1204 12:08:28.154706 137274321021824 utils.py:1231] [90900] core_hours = 158.2705481493833
+I1204 12:08:28.154775 137274321021824 train.py:125] NOTE: Steps:90900/112603 [80.7%]
+Walltime:6d14h18m (0s eval)
+ETA:1d13h47m
+Total train time:8d4h3m
+I1204 12:13:39.466824 137274321021824 utils.py:1231] [90950] l2_params = 246.33161135165426
+I1204 12:13:39.467036 137274321021824 utils.py:1231] [90950] train/loss = 1.734895944595337
+I1204 12:13:39.467129 137274321021824 utils.py:1231] [90950] l2_grads = 2.568667411804199
+I1204 12:13:39.467186 137274321021824 utils.py:1231] [90950] lr = 0.0001059321092149473
+I1204 12:13:39.467235 137274321021824 utils.py:1231] [90950] uptime = 570208.829597948
+I1204 12:13:39.467296 137274321021824 utils.py:1231] [90950] examples_seen = 93132800.0
+I1204 12:13:39.467343 137274321021824 utils.py:1231] [90950] progress = 0.8077049456941644
+I1204 12:13:39.467388 137274321021824 utils.py:1231] [90950] epoch = 72.69372376903246
+I1204 12:13:39.467435 137274321021824 utils.py:1231] [90950] img/sec/core = 164.4647599495164
+I1204 12:13:39.467488 137274321021824 utils.py:1231] [90950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.35702395262888
+I1204 12:13:39.467540 137274321021824 utils.py:1231] [90950] core_hours = 158.35702395262888
+I1204 12:13:39.467598 137274321021824 train.py:125] NOTE: Steps:90950/112603 [80.8%]
+Walltime:6d14h23m (0s eval)
+ETA:1d13h42m
+Total train time:8d4h3m
+I1204 12:18:51.278664 137274321021824 utils.py:1231] [91000] l2_params = 246.28361099562613
+I1204 12:18:51.278921 137274321021824 utils.py:1231] [91000] train/loss = 1.5906340330839157
+I1204 12:18:51.279062 137274321021824 utils.py:1231] [91000] l2_grads = 2.4461169242858887
+I1204 12:18:51.279131 137274321021824 utils.py:1231] [91000] lr = 0.00010546142155199233
+I1204 12:18:51.279183 137274321021824 utils.py:1231] [91000] uptime = 570520.641545056
+I1204 12:18:51.279237 137274321021824 utils.py:1231] [91000] examples_seen = 93184000.0
+I1204 12:18:51.279287 137274321021824 utils.py:1231] [91000] progress = 0.8081489835972399
+I1204 12:18:51.279335 137274321021824 utils.py:1231] [91000] epoch = 72.73368733350141
+I1204 12:18:51.279386 137274321021824 utils.py:1231] [91000] img/sec/core = 164.20153388883463
+I1204 12:18:51.279442 137274321021824 utils.py:1231] [91000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.44363838238112
+I1204 12:18:51.279499 137274321021824 utils.py:1231] [91000] core_hours = 158.44363838238112
+I1204 12:18:51.279559 137274321021824 train.py:125] NOTE: Steps:91000/112603 [80.8%]
+Walltime:6d14h28m (0s eval)
+ETA:1d13h36m
+Total train time:8d4h3m
+I1204 12:24:02.656495 137274321021824 utils.py:1231] [91050] l2_params = 246.23916885898853
+I1204 12:24:02.656702 137274321021824 utils.py:1231] [91050] train/loss = 3.869361788034439
+I1204 12:24:02.656802 137274321021824 utils.py:1231] [91050] l2_grads = 2.4108142852783203
+I1204 12:24:02.656872 137274321021824 utils.py:1231] [91050] lr = 0.00010499165860642971
+I1204 12:24:02.656936 137274321021824 utils.py:1231] [91050] uptime = 570832.019297341
+I1204 12:24:02.656995 137274321021824 utils.py:1231] [91050] examples_seen = 93235200.0
+I1204 12:24:02.657047 137274321021824 utils.py:1231] [91050] progress = 0.8085930215003153
+I1204 12:24:02.657095 137274321021824 utils.py:1231] [91050] epoch = 72.77365089797037
+I1204 12:24:02.657157 137274321021824 utils.py:1231] [91050] img/sec/core = 164.43050161509217
+I1204 12:24:02.657221 137274321021824 utils.py:1231] [91050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.53013220246027
+I1204 12:24:02.657272 137274321021824 utils.py:1231] [91050] core_hours = 158.53013220246027
+I1204 12:24:02.657333 137274321021824 train.py:125] NOTE: Steps:91050/112603 [80.9%]
+Walltime:6d14h33m (0s eval)
+ETA:1d13h31m
+Total train time:8d4h3m
+I1204 12:29:14.444523 137274321021824 utils.py:1231] [91100] l2_params = 246.18434789324272
+I1204 12:29:14.444747 137274321021824 utils.py:1231] [91100] train/loss = 2.775000810623169
+I1204 12:29:14.444856 137274321021824 utils.py:1231] [91100] l2_grads = 2.2007038593292236
+I1204 12:29:14.444943 137274321021824 utils.py:1231] [91100] lr = 0.00010452282147928775
+I1204 12:29:14.445032 137274321021824 utils.py:1231] [91100] uptime = 571143.807386832
+I1204 12:29:14.445124 137274321021824 utils.py:1231] [91100] examples_seen = 93286400.0
+I1204 12:29:14.445185 137274321021824 utils.py:1231] [91100] progress = 0.8090370594033907
+I1204 12:29:14.445246 137274321021824 utils.py:1231] [91100] epoch = 72.81361446243932
+I1204 12:29:14.445312 137274321021824 utils.py:1231] [91100] img/sec/core = 164.21409837557084
+I1204 12:29:14.445394 137274321021824 utils.py:1231] [91100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.61674000509666
+I1204 12:29:14.445479 137274321021824 utils.py:1231] [91100] core_hours = 158.61674000509666
+I1204 12:29:14.445553 137274321021824 train.py:125] NOTE: Steps:91100/112603 [80.9%]
+Walltime:6d14h39m (0s eval)
+ETA:1d13h26m
+Total train time:8d4h3m
+I1204 12:34:26.203557 137274321021824 utils.py:1231] [91150] l2_params = 246.13347089467806
+I1204 12:34:26.203836 137274321021824 utils.py:1231] [91150] train/loss = 2.352405548095703
+I1204 12:34:26.204018 137274321021824 utils.py:1231] [91150] l2_grads = 2.1831471920013428
+I1204 12:34:26.204104 137274321021824 utils.py:1231] [91150] lr = 0.00010405491126942405
+I1204 12:34:26.204184 137274321021824 utils.py:1231] [91150] uptime = 571455.56653733
+I1204 12:34:26.204261 137274321021824 utils.py:1231] [91150] examples_seen = 93337600.0
+I1204 12:34:26.204326 137274321021824 utils.py:1231] [91150] progress = 0.8094810973064661
+I1204 12:34:26.204409 137274321021824 utils.py:1231] [91150] epoch = 72.85357802690828
+I1204 12:34:26.204483 137274321021824 utils.py:1231] [91150] img/sec/core = 164.22934152277642
+I1204 12:34:26.204577 137274321021824 utils.py:1231] [91150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.7033397691239
+I1204 12:34:26.204649 137274321021824 utils.py:1231] [91150] core_hours = 158.7033397691239
+I1204 12:34:26.204752 137274321021824 train.py:125] NOTE: Steps:91150/112603 [80.9%]
+Walltime:6d14h44m (0s eval)
+ETA:1d13h21m
+Total train time:8d4h3m
+I1204 12:39:37.996840 137274321021824 utils.py:1231] [91200] l2_params = 246.08540016674434
+I1204 12:39:37.997065 137274321021824 utils.py:1231] [91200] train/loss = 1.7333573251962662
+I1204 12:39:37.997170 137274321021824 utils.py:1231] [91200] l2_grads = 2.346733331680298
+I1204 12:39:37.997265 137274321021824 utils.py:1231] [91200] lr = 0.00010358792907352435
+I1204 12:39:37.997357 137274321021824 utils.py:1231] [91200] uptime = 571767.359715191
+I1204 12:39:37.997445 137274321021824 utils.py:1231] [91200] examples_seen = 93388800.0
+I1204 12:39:37.997522 137274321021824 utils.py:1231] [91200] progress = 0.8099251352095415
+I1204 12:39:37.997585 137274321021824 utils.py:1231] [91200] epoch = 72.89354159137724
+I1204 12:39:37.997661 137274321021824 utils.py:1231] [91200] img/sec/core = 164.21141845137137
+I1204 12:39:37.997733 137274321021824 utils.py:1231] [91200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.7899489851964
+I1204 12:39:37.997795 137274321021824 utils.py:1231] [91200] core_hours = 158.7899489851964
+I1204 12:39:37.997909 137274321021824 train.py:125] NOTE: Steps:91200/112603 [81.0%]
+Walltime:6d14h49m (0s eval)
+ETA:1d13h15m
+Total train time:8d4h3m
+I1204 12:44:49.782448 137274321021824 utils.py:1231] [91250] l2_params = 246.03929184116913
+I1204 12:44:49.782685 137274321021824 utils.py:1231] [91250] train/loss = 2.6840329468250275
+I1204 12:44:49.782791 137274321021824 utils.py:1231] [91250] l2_grads = 2.2407236099243164
+I1204 12:44:49.782879 137274321021824 utils.py:1231] [91250] lr = 0.00010312187598609862
+I1204 12:44:49.782948 137274321021824 utils.py:1231] [91250] uptime = 572079.145307474
+I1204 12:44:49.783013 137274321021824 utils.py:1231] [91250] examples_seen = 93440000.0
+I1204 12:44:49.783069 137274321021824 utils.py:1231] [91250] progress = 0.8103691731126169
+I1204 12:44:49.783124 137274321021824 utils.py:1231] [91250] epoch = 72.93350515584619
+I1204 12:44:49.783191 137274321021824 utils.py:1231] [91250] img/sec/core = 164.2154136279847
+I1204 12:44:49.783259 137274321021824 utils.py:1231] [91250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.87655609416387
+I1204 12:44:49.783323 137274321021824 utils.py:1231] [91250] core_hours = 158.87655609416387
+I1204 12:44:49.783412 137274321021824 train.py:125] NOTE: Steps:91250/112603 [81.0%]
+Walltime:6d14h54m (0s eval)
+ETA:1d13h10m
+Total train time:8d4h3m
+I1204 12:50:01.528095 137274321021824 utils.py:1231] [91300] l2_params = 245.9874897067674
+I1204 12:50:01.528326 137274321021824 utils.py:1231] [91300] train/loss = 1.7264331877231598
+I1204 12:50:01.528455 137274321021824 utils.py:1231] [91300] l2_grads = 2.4631950855255127
+I1204 12:50:01.528543 137274321021824 utils.py:1231] [91300] lr = 0.00010265675309947983
+I1204 12:50:01.528625 137274321021824 utils.py:1231] [91300] uptime = 572390.8909830659
+I1204 12:50:01.528701 137274321021824 utils.py:1231] [91300] examples_seen = 93491200.0
+I1204 12:50:01.528771 137274321021824 utils.py:1231] [91300] progress = 0.8108132110156923
+I1204 12:50:01.528838 137274321021824 utils.py:1231] [91300] epoch = 72.97346872031515
+I1204 12:50:01.528951 137274321021824 utils.py:1231] [91300] img/sec/core = 164.2364401776369
+I1204 12:50:01.529028 137274321021824 utils.py:1231] [91300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 158.96315211516165
+I1204 12:50:01.529093 137274321021824 utils.py:1231] [91300] core_hours = 158.96315211516165
+I1204 12:50:01.529168 137274321021824 train.py:125] NOTE: Steps:91300/112603 [81.1%]
+Walltime:6d14h59m (0s eval)
+ETA:1d13h5m
+Total train time:8d4h3m
+I1204 12:55:13.315558 137274321021824 utils.py:1231] [91350] l2_params = 245.93833975254734
+I1204 12:55:13.315825 137274321021824 utils.py:1231] [91350] train/loss = 3.5387825667858124
+I1204 12:55:13.315958 137274321021824 utils.py:1231] [91350] l2_grads = 2.3917431831359863
+I1204 12:55:13.316033 137274321021824 utils.py:1231] [91350] lr = 0.00010219256150382053
+I1204 12:55:13.316095 137274321021824 utils.py:1231] [91350] uptime = 572702.678456114
+I1204 12:55:13.316164 137274321021824 utils.py:1231] [91350] examples_seen = 93542400.0
+I1204 12:55:13.316238 137274321021824 utils.py:1231] [91350] progress = 0.8112572489187677
+I1204 12:55:13.316302 137274321021824 utils.py:1231] [91350] epoch = 73.0134322847841
+I1204 12:55:13.316366 137274321021824 utils.py:1231] [91350] img/sec/core = 164.21442304740702
+I1204 12:55:13.316443 137274321021824 utils.py:1231] [91350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.0497597465639
+I1204 12:55:13.316509 137274321021824 utils.py:1231] [91350] core_hours = 159.0497597465639
+I1204 12:55:13.316583 137274321021824 train.py:125] NOTE: Steps:91350/112603 [81.1%]
+Walltime:6d15h5m (0s eval)
+ETA:1d13h0m
+Total train time:8d4h3m
+I1204 13:00:25.108610 137274321021824 utils.py:1231] [91400] l2_params = 245.89059462895267
+I1204 13:00:25.108870 137274321021824 utils.py:1231] [91400] train/loss = 3.0415656566619873
+I1204 13:00:25.109013 137274321021824 utils.py:1231] [91400] l2_grads = 2.333103656768799
+I1204 13:00:25.109086 137274321021824 utils.py:1231] [91400] lr = 0.00010172930228709034
+I1204 13:00:25.109149 137274321021824 utils.py:1231] [91400] uptime = 573014.471509879
+I1204 13:00:25.109211 137274321021824 utils.py:1231] [91400] examples_seen = 93593600.0
+I1204 13:00:25.109270 137274321021824 utils.py:1231] [91400] progress = 0.8117012868218431
+I1204 13:00:25.109326 137274321021824 utils.py:1231] [91400] epoch = 73.05339584925306
+I1204 13:00:25.109382 137274321021824 utils.py:1231] [91400] img/sec/core = 164.2114838087292
+I1204 13:00:25.109456 137274321021824 utils.py:1231] [91400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.13636892816527
+I1204 13:00:25.109513 137274321021824 utils.py:1231] [91400] core_hours = 159.13636892816527
+I1204 13:00:25.109591 137274321021824 train.py:125] NOTE: Steps:91400/112603 [81.2%]
+Walltime:6d15h10m (0s eval)
+ETA:1d12h55m
+Total train time:8d4h3m
+I1204 13:05:36.862885 137274321021824 utils.py:1231] [91450] l2_params = 245.84384334065902
+I1204 13:05:36.863130 137274321021824 utils.py:1231] [91450] train/loss = 3.463860422372818
+I1204 13:05:36.863263 137274321021824 utils.py:1231] [91450] l2_grads = 2.4110476970672607
+I1204 13:05:36.863343 137274321021824 utils.py:1231] [91450] lr = 0.00010126697653507394
+I1204 13:05:36.863414 137274321021824 utils.py:1231] [91450] uptime = 573326.225775501
+I1204 13:05:36.863505 137274321021824 utils.py:1231] [91450] examples_seen = 93644800.0
+I1204 13:05:36.863567 137274321021824 utils.py:1231] [91450] progress = 0.8121453247249185
+I1204 13:05:36.863634 137274321021824 utils.py:1231] [91450] epoch = 73.09335941372203
+I1204 13:05:36.863691 137274321021824 utils.py:1231] [91450] img/sec/core = 164.2319148315333
+I1204 13:05:36.863753 137274321021824 utils.py:1231] [91450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.2229673352825
+I1204 13:05:36.863807 137274321021824 utils.py:1231] [91450] core_hours = 159.2229673352825
+I1204 13:05:36.863870 137274321021824 train.py:125] NOTE: Steps:91450/112603 [81.2%]
+Walltime:6d15h15m (0s eval)
+ETA:1d12h49m
+Total train time:8d4h3m
+I1204 13:10:48.665124 137274321021824 utils.py:1231] [91500] l2_params = 245.79751009935134
+I1204 13:10:48.665397 137274321021824 utils.py:1231] [91500] train/loss = 3.5562371611595154
+I1204 13:10:48.665619 137274321021824 utils.py:1231] [91500] l2_grads = 2.364051580429077
+I1204 13:10:48.665708 137274321021824 utils.py:1231] [91500] lr = 0.0001008055853313676
+I1204 13:10:48.665760 137274321021824 utils.py:1231] [91500] uptime = 573638.028122404
+I1204 13:10:48.665819 137274321021824 utils.py:1231] [91500] examples_seen = 93696000.0
+I1204 13:10:48.665871 137274321021824 utils.py:1231] [91500] progress = 0.812589362627994
+I1204 13:10:48.665936 137274321021824 utils.py:1231] [91500] epoch = 73.13332297819098
+I1204 13:10:48.665998 137274321021824 utils.py:1231] [91500] img/sec/core = 164.20658955439305
+I1204 13:10:48.666061 137274321021824 utils.py:1231] [91500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.3095790983111
+I1204 13:10:48.666124 137274321021824 utils.py:1231] [91500] core_hours = 159.3095790983111
+I1204 13:10:48.666197 137274321021824 train.py:125] NOTE: Steps:91500/112603 [81.3%]
+Walltime:6d15h20m (0s eval)
+ETA:1d12h44m
+Total train time:8d4h3m
+I1204 13:16:00.432610 137274321021824 utils.py:1231] [91550] l2_params = 245.75272498118917
+I1204 13:16:00.432813 137274321021824 utils.py:1231] [91550] train/loss = 2.808947265148163
+I1204 13:16:00.432913 137274321021824 utils.py:1231] [91550] l2_grads = 2.314518928527832
+I1204 13:16:00.432977 137274321021824 utils.py:1231] [91550] lr = 0.0001003451297573777
+I1204 13:16:00.433030 137274321021824 utils.py:1231] [91550] uptime = 573949.795392476
+I1204 13:16:00.433085 137274321021824 utils.py:1231] [91550] examples_seen = 93747200.0
+I1204 13:16:00.433134 137274321021824 utils.py:1231] [91550] progress = 0.8130334005310693
+I1204 13:16:00.433182 137274321021824 utils.py:1231] [91550] epoch = 73.17328654265994
+I1204 13:16:00.433233 137274321021824 utils.py:1231] [91550] img/sec/core = 164.22506438271563
+I1204 13:16:00.433290 137274321021824 utils.py:1231] [91550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.39618111777557
+I1204 13:16:00.433340 137274321021824 utils.py:1231] [91550] core_hours = 159.39618111777557
+I1204 13:16:00.433402 137274321021824 train.py:125] NOTE: Steps:91550/112603 [81.3%]
+Walltime:6d15h25m (0s eval)
+ETA:1d12h39m
+Total train time:8d4h3m
+I1204 13:21:12.183576 137274321021824 utils.py:1231] [91600] l2_params = 245.7016390327677
+I1204 13:21:12.183788 137274321021824 utils.py:1231] [91600] train/loss = 3.9192977249622345
+I1204 13:21:12.183900 137274321021824 utils.py:1231] [91600] l2_grads = 2.6912319660186768
+I1204 13:21:12.192494 137274321021824 utils.py:1231] [91600] lr = 9.988561089231769e-05
+I1204 13:21:12.192797 137274321021824 utils.py:1231] [91600] uptime = 574261.555130187
+I1204 13:21:12.192897 137274321021824 utils.py:1231] [91600] examples_seen = 93798400.0
+I1204 13:21:12.192973 137274321021824 utils.py:1231] [91600] progress = 0.8134774384341448
+I1204 13:21:12.193050 137274321021824 utils.py:1231] [91600] epoch = 73.21325010712889
+I1204 13:21:12.193104 137274321021824 utils.py:1231] [91600] img/sec/core = 164.22903218975955
+I1204 13:21:12.193194 137274321021824 utils.py:1231] [91600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.4827810449175
+I1204 13:21:12.193246 137274321021824 utils.py:1231] [91600] core_hours = 159.4827810449175
+I1204 13:21:12.193326 137274321021824 train.py:125] NOTE: Steps:91600/112603 [81.3%]
+Walltime:6d15h31m (0s eval)
+ETA:1d12h34m
+Total train time:8d4h3m
+I1204 13:26:23.971486 137274321021824 utils.py:1231] [91650] l2_params = 245.6535850738725
+I1204 13:26:23.971757 137274321021824 utils.py:1231] [91650] train/loss = 1.6577794700860977
+I1204 13:26:23.971869 137274321021824 utils.py:1231] [91650] l2_grads = 2.3754749298095703
+I1204 13:26:23.971976 137274321021824 utils.py:1231] [91650] lr = 9.94270298132051e-05
+I1204 13:26:23.972047 137274321021824 utils.py:1231] [91650] uptime = 574573.334409205
+I1204 13:26:23.972098 137274321021824 utils.py:1231] [91650] examples_seen = 93849600.0
+I1204 13:26:23.972145 137274321021824 utils.py:1231] [91650] progress = 0.8139214763372201
+I1204 13:26:23.972191 137274321021824 utils.py:1231] [91650] epoch = 73.25321367159785
+I1204 13:26:23.972240 137274321021824 utils.py:1231] [91650] img/sec/core = 164.2187388502981
+I1204 13:26:23.972294 137274321021824 utils.py:1231] [91650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.56938640020027
+I1204 13:26:23.972343 137274321021824 utils.py:1231] [91650] core_hours = 159.56938640020027
+I1204 13:26:23.972401 137274321021824 train.py:125] NOTE: Steps:91650/112603 [81.4%]
+Walltime:6d15h36m (0s eval)
+ETA:1d12h28m
+Total train time:8d4h3m
+I1204 13:31:35.756194 137274321021824 utils.py:1231] [91700] l2_params = 245.6053144828054
+I1204 13:31:35.756479 137274321021824 utils.py:1231] [91700] train/loss = 1.5331576466560364
+I1204 13:31:35.756675 137274321021824 utils.py:1231] [91700] l2_grads = 2.5501389503479004
+I1204 13:31:35.756801 137274321021824 utils.py:1231] [91700] lr = 9.896938759486002e-05
+I1204 13:31:35.756908 137274321021824 utils.py:1231] [91700] uptime = 574885.119260372
+I1204 13:31:35.757021 137274321021824 utils.py:1231] [91700] examples_seen = 93900800.0
+I1204 13:31:35.757103 137274321021824 utils.py:1231] [91700] progress = 0.8143655142402956
+I1204 13:31:35.757198 137274321021824 utils.py:1231] [91700] epoch = 73.29317723606681
+I1204 13:31:35.757307 137274321021824 utils.py:1231] [91700] img/sec/core = 164.21580396982043
+I1204 13:31:35.757402 137274321021824 utils.py:1231] [91700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.6559933033022
+I1204 13:31:35.757473 137274321021824 utils.py:1231] [91700] core_hours = 159.6559933033022
+I1204 13:31:35.757562 137274321021824 train.py:125] NOTE: Steps:91700/112603 [81.4%]
+Walltime:6d15h41m (0s eval)
+ETA:1d12h23m
+Total train time:8d4h3m
+I1204 13:36:47.464861 137274321021824 utils.py:1231] [91750] l2_params = 245.55725111464676
+I1204 13:36:47.465121 137274321021824 utils.py:1231] [91750] train/loss = 1.5949702262878418
+I1204 13:36:47.465255 137274321021824 utils.py:1231] [91750] l2_grads = 2.4087979793548584
+I1204 13:36:47.465360 137274321021824 utils.py:1231] [91750] lr = 9.851268530990162e-05
+I1204 13:36:47.465432 137274321021824 utils.py:1231] [91750] uptime = 575196.827787918
+I1204 13:36:47.465503 137274321021824 utils.py:1231] [91750] examples_seen = 93952000.0
+I1204 13:36:47.465561 137274321021824 utils.py:1231] [91750] progress = 0.8148095521433709
+I1204 13:36:47.465617 137274321021824 utils.py:1231] [91750] epoch = 73.33314080053576
+I1204 13:36:47.465676 137274321021824 utils.py:1231] [91750] img/sec/core = 164.25601315138815
+I1204 13:36:47.465740 137274321021824 utils.py:1231] [91750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.74257900539834
+I1204 13:36:47.465803 137274321021824 utils.py:1231] [91750] core_hours = 159.74257900539834
+I1204 13:36:47.465904 137274321021824 train.py:125] NOTE: Steps:91750/112603 [81.5%]
+Walltime:6d15h46m (0s eval)
+ETA:1d12h18m
+Total train time:8d4h3m
+I1204 13:41:59.115255 137274321021824 utils.py:1231] [91800] l2_params = 245.5092023134307
+I1204 13:41:59.115595 137274321021824 utils.py:1231] [91800] train/loss = 1.5193701535463333
+I1204 13:41:59.115806 137274321021824 utils.py:1231] [91800] l2_grads = 2.3707058429718018
+I1204 13:41:59.115934 137274321021824 utils.py:1231] [91800] lr = 9.80569240287464e-05
+I1204 13:41:59.116032 137274321021824 utils.py:1231] [91800] uptime = 575508.478386103
+I1204 13:41:59.116132 137274321021824 utils.py:1231] [91800] examples_seen = 94003200.0
+I1204 13:41:59.116225 137274321021824 utils.py:1231] [91800] progress = 0.8152535900464464
+I1204 13:41:59.116296 137274321021824 utils.py:1231] [91800] epoch = 73.37310436500472
+I1204 13:41:59.116373 137274321021824 utils.py:1231] [91800] img/sec/core = 164.28654492625984
+I1204 13:41:59.116445 137274321021824 utils.py:1231] [91800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.82914861600526
+I1204 13:41:59.116510 137274321021824 utils.py:1231] [91800] core_hours = 159.82914861600526
+I1204 13:41:59.116590 137274321021824 train.py:125] NOTE: Steps:91800/112603 [81.5%]
+Walltime:6d15h51m (0s eval)
+ETA:1d12h13m
+Total train time:8d4h3m
+I1204 13:47:10.903412 137274321021824 utils.py:1231] [91850] l2_params = 245.4614013520096
+I1204 13:47:10.903657 137274321021824 utils.py:1231] [91850] train/loss = 1.907950535416603
+I1204 13:47:10.903762 137274321021824 utils.py:1231] [91850] l2_grads = 2.3217806816101074
+I1204 13:47:10.903841 137274321021824 utils.py:1231] [91850] lr = 9.760210481960528e-05
+I1204 13:47:10.903909 137274321021824 utils.py:1231] [91850] uptime = 575820.266269257
+I1204 13:47:10.903970 137274321021824 utils.py:1231] [91850] examples_seen = 94054400.0
+I1204 13:47:10.904028 137274321021824 utils.py:1231] [91850] progress = 0.8156976279495217
+I1204 13:47:10.904086 137274321021824 utils.py:1231] [91850] epoch = 73.41306792947367
+I1204 13:47:10.904144 137274321021824 utils.py:1231] [91850] img/sec/core = 164.21420705021748
+I1204 13:47:10.904207 137274321021824 utils.py:1231] [91850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 159.9157563613258
+I1204 13:47:10.904265 137274321021824 utils.py:1231] [91850] core_hours = 159.9157563613258
+I1204 13:47:10.904334 137274321021824 train.py:125] NOTE: Steps:91850/112603 [81.6%]
+Walltime:6d15h57m (0s eval)
+ETA:1d12h7m
+Total train time:8d4h3m
+I1204 13:52:22.667823 137274321021824 utils.py:1231] [91900] l2_params = 245.41514229848866
+I1204 13:52:22.668084 137274321021824 utils.py:1231] [91900] train/loss = 1.5525540858507156
+I1204 13:52:22.668204 137274321021824 utils.py:1231] [91900] l2_grads = 2.5111806392669678
+I1204 13:52:22.668276 137274321021824 utils.py:1231] [91900] lr = 9.714822874848103e-05
+I1204 13:52:22.668335 137274321021824 utils.py:1231] [91900] uptime = 576132.030696994
+I1204 13:52:22.668405 137274321021824 utils.py:1231] [91900] examples_seen = 94105600.0
+I1204 13:52:22.668453 137274321021824 utils.py:1231] [91900] progress = 0.8161416658525972
+I1204 13:52:22.668500 137274321021824 utils.py:1231] [91900] epoch = 73.45303149394263
+I1204 13:52:22.668550 137274321021824 utils.py:1231] [91900] img/sec/core = 164.22656161136882
+I1204 13:52:22.668605 137274321021824 utils.py:1231] [91900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.00235759125277
+I1204 13:52:22.668655 137274321021824 utils.py:1231] [91900] core_hours = 160.00235759125277
+I1204 13:52:22.668714 137274321021824 train.py:125] NOTE: Steps:91900/112603 [81.6%]
+Walltime:6d16h2m (0s eval)
+ETA:1d12h2m
+Total train time:8d4h3m
+I1204 13:57:34.449481 137274321021824 utils.py:1231] [91950] l2_params = 245.36754893401007
+I1204 13:57:34.449722 137274321021824 utils.py:1231] [91950] train/loss = 1.7127672731876373
+I1204 13:57:34.449823 137274321021824 utils.py:1231] [91950] l2_grads = 2.5274124145507812
+I1204 13:57:34.449899 137274321021824 utils.py:1231] [91950] lr = 9.669529687916576e-05
+I1204 13:57:34.449969 137274321021824 utils.py:1231] [91950] uptime = 576443.8123308109
+I1204 13:57:34.450026 137274321021824 utils.py:1231] [91950] examples_seen = 94156800.0
+I1204 13:57:34.450082 137274321021824 utils.py:1231] [91950] progress = 0.8165857037556726
+I1204 13:57:34.450136 137274321021824 utils.py:1231] [91950] epoch = 73.4929950584116
+I1204 13:57:34.450192 137274321021824 utils.py:1231] [91950] img/sec/core = 164.21749855243075
+I1204 13:57:34.450249 137274321021824 utils.py:1231] [91950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.08896360064637
+I1204 13:57:34.450301 137274321021824 utils.py:1231] [91950] core_hours = 160.08896360064637
+I1204 13:57:34.450366 137274321021824 train.py:125] NOTE: Steps:91950/112603 [81.7%]
+Walltime:6d16h7m (0s eval)
+ETA:1d11h57m
+Total train time:8d4h3m
+I1204 14:02:46.177386 137274321021824 utils.py:1231] [92000] l2_params = 245.32047422010183
+I1204 14:02:46.177591 137274321021824 utils.py:1231] [92000] train/loss = 4.021910399198532
+I1204 14:02:46.177691 137274321021824 utils.py:1231] [92000] l2_grads = 2.64506459236145
+I1204 14:02:46.177774 137274321021824 utils.py:1231] [92000] lr = 9.624331027323879e-05
+I1204 14:02:46.177834 137274321021824 utils.py:1231] [92000] uptime = 576755.540196293
+I1204 14:02:46.177911 137274321021824 utils.py:1231] [92000] examples_seen = 94208000.0
+I1204 14:02:46.177973 137274321021824 utils.py:1231] [92000] progress = 0.817029741658748
+I1204 14:02:46.178031 137274321021824 utils.py:1231] [92000] epoch = 73.53295862288054
+I1204 14:02:46.178086 137274321021824 utils.py:1231] [92000] img/sec/core = 164.2458235833917
+I1204 14:02:46.178147 137274321021824 utils.py:1231] [92000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.1755546743914
+I1204 14:02:46.178198 137274321021824 utils.py:1231] [92000] core_hours = 160.1755546743914
+I1204 14:02:46.178277 137274321021824 train.py:125] NOTE: Steps:92000/112603 [81.7%]
+Walltime:6d16h12m (0s eval)
+ETA:1d11h52m
+Total train time:8d4h3m
+I1204 14:07:58.312451 137274321021824 utils.py:1231] [92050] l2_params = 245.2732254587549
+I1204 14:07:58.312670 137274321021824 utils.py:1231] [92050] train/loss = 2.470121204853058
+I1204 14:07:58.312778 137274321021824 utils.py:1231] [92050] l2_grads = 2.434964656829834
+I1204 14:07:58.312861 137274321021824 utils.py:1231] [92050] lr = 9.579226999006394e-05
+I1204 14:07:58.312929 137274321021824 utils.py:1231] [92050] uptime = 577067.675289747
+I1204 14:07:58.312992 137274321021824 utils.py:1231] [92050] examples_seen = 94259200.0
+I1204 14:07:58.313053 137274321021824 utils.py:1231] [92050] progress = 0.8174737795618234
+I1204 14:07:58.313111 137274321021824 utils.py:1231] [92050] epoch = 73.5729221873495
+I1204 14:07:58.313174 137274321021824 utils.py:1231] [92050] img/sec/core = 164.03153978441773
+I1204 14:07:58.313237 137274321021824 utils.py:1231] [92050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.2622588670175
+I1204 14:07:58.313296 137274321021824 utils.py:1231] [92050] core_hours = 160.2622588670175
+I1204 14:07:58.313361 137274321021824 train.py:125] NOTE: Steps:92050/112603 [81.7%]
+Walltime:6d16h17m (0s eval)
+ETA:1d11h47m
+Total train time:8d4h3m
+I1204 14:13:10.105478 137274321021824 utils.py:1231] [92100] l2_params = 245.2248115339711
+I1204 14:13:10.105705 137274321021824 utils.py:1231] [92100] train/loss = 2.10051429271698
+I1204 14:13:10.105855 137274321021824 utils.py:1231] [92100] l2_grads = 2.3175790309906006
+I1204 14:13:10.105953 137274321021824 utils.py:1231] [92100] lr = 9.534217708678702e-05
+I1204 14:13:10.106026 137274321021824 utils.py:1231] [92100] uptime = 577379.46838713
+I1204 14:13:10.106102 137274321021824 utils.py:1231] [92100] examples_seen = 94310400.0
+I1204 14:13:10.106163 137274321021824 utils.py:1231] [92100] progress = 0.8179178174648988
+I1204 14:13:10.106231 137274321021824 utils.py:1231] [92100] epoch = 73.61288575181845
+I1204 14:13:10.106299 137274321021824 utils.py:1231] [92100] img/sec/core = 164.21146083650495
+I1204 14:13:10.106370 137274321021824 utils.py:1231] [92100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.348868060735
+I1204 14:13:10.106436 137274321021824 utils.py:1231] [92100] core_hours = 160.348868060735
+I1204 14:13:10.106501 137274321021824 train.py:125] NOTE: Steps:92100/112603 [81.8%]
+Walltime:6d16h22m (0s eval)
+ETA:1d11h41m
+Total train time:8d4h2m
+I1204 14:18:21.887727 137274321021824 utils.py:1231] [92150] l2_params = 245.1798081363223
+I1204 14:18:21.888006 137274321021824 utils.py:1231] [92150] train/loss = 3.0245987474918365
+I1204 14:18:21.888173 137274321021824 utils.py:1231] [92150] l2_grads = 2.357384443283081
+I1204 14:18:21.888266 137274321021824 utils.py:1231] [92150] lr = 9.48930326183334e-05
+I1204 14:18:21.888329 137274321021824 utils.py:1231] [92150] uptime = 577691.250691058
+I1204 14:18:21.888410 137274321021824 utils.py:1231] [92150] examples_seen = 94361600.0
+I1204 14:18:21.888490 137274321021824 utils.py:1231] [92150] progress = 0.8183618553679742
+I1204 14:18:21.888584 137274321021824 utils.py:1231] [92150] epoch = 73.65284931628742
+I1204 14:18:21.888662 137274321021824 utils.py:1231] [92150] img/sec/core = 164.21714560110408
+I1204 14:18:21.888739 137274321021824 utils.py:1231] [92150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.43547425627054
+I1204 14:18:21.888810 137274321021824 utils.py:1231] [92150] core_hours = 160.43547425627054
+I1204 14:18:21.888904 137274321021824 train.py:125] NOTE: Steps:92150/112603 [81.8%]
+Walltime:6d16h28m (0s eval)
+ETA:1d11h36m
+Total train time:8d4h2m
+I1204 14:23:33.596416 137274321021824 utils.py:1231] [92200] l2_params = 245.13417362560205
+I1204 14:23:33.596688 137274321021824 utils.py:1231] [92200] train/loss = 3.787135422229767
+I1204 14:23:33.596892 137274321021824 utils.py:1231] [92200] l2_grads = 2.5873517990112305
+I1204 14:23:33.596994 137274321021824 utils.py:1231] [92200] lr = 9.444483763740524e-05
+I1204 14:23:33.597072 137274321021824 utils.py:1231] [92200] uptime = 578002.959429694
+I1204 14:23:33.597147 137274321021824 utils.py:1231] [92200] examples_seen = 94412800.0
+I1204 14:23:33.597216 137274321021824 utils.py:1231] [92200] progress = 0.8188058932710496
+I1204 14:23:33.597285 137274321021824 utils.py:1231] [92200] epoch = 73.69281288075638
+I1204 14:23:33.597348 137274321021824 utils.py:1231] [92200] img/sec/core = 164.2559019167902
+I1204 14:23:33.597421 137274321021824 utils.py:1231] [92200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.5220600170028
+I1204 14:23:33.597491 137274321021824 utils.py:1231] [92200] core_hours = 160.5220600170028
+I1204 14:23:33.597563 137274321021824 train.py:125] NOTE: Steps:92200/112603 [81.9%]
+Walltime:6d16h33m (0s eval)
+ETA:1d11h31m
+Total train time:8d4h2m
+I1204 14:28:45.261258 137274321021824 utils.py:1231] [92250] l2_params = 245.08716876520293
+I1204 14:28:45.261469 137274321021824 utils.py:1231] [92250] train/loss = 1.6992948651313782
+I1204 14:28:45.261582 137274321021824 utils.py:1231] [92250] l2_grads = 2.5109217166900635
+I1204 14:28:45.261685 137274321021824 utils.py:1231] [92250] lr = 9.399759319447973e-05
+I1204 14:28:45.261770 137274321021824 utils.py:1231] [92250] uptime = 578314.624129234
+I1204 14:28:45.261835 137274321021824 utils.py:1231] [92250] examples_seen = 94464000.0
+I1204 14:28:45.261902 137274321021824 utils.py:1231] [92250] progress = 0.819249931174125
+I1204 14:28:45.261960 137274321021824 utils.py:1231] [92250] epoch = 73.73277644522533
+I1204 14:28:45.262013 137274321021824 utils.py:1231] [92250] img/sec/core = 164.27911173632143
+I1204 14:28:45.262073 137274321021824 utils.py:1231] [92250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.60863354465275
+I1204 14:28:45.262130 137274321021824 utils.py:1231] [92250] core_hours = 160.60863354465275
+I1204 14:28:45.262196 137274321021824 train.py:125] NOTE: Steps:92250/112603 [81.9%]
+Walltime:6d16h38m (0s eval)
+ETA:1d11h26m
+Total train time:8d4h2m
+I1204 14:33:57.061959 137274321021824 utils.py:1231] [92300] l2_params = 245.04305551657558
+I1204 14:33:57.062163 137274321021824 utils.py:1231] [92300] train/loss = 2.957980841398239
+I1204 14:33:57.062300 137274321021824 utils.py:1231] [92300] l2_grads = 2.47979736328125
+I1204 14:33:57.062428 137274321021824 utils.py:1231] [92300] lr = 9.355130033780596e-05
+I1204 14:33:57.062510 137274321021824 utils.py:1231] [92300] uptime = 578626.424869724
+I1204 14:33:57.062574 137274321021824 utils.py:1231] [92300] examples_seen = 94515200.0
+I1204 14:33:57.062629 137274321021824 utils.py:1231] [92300] progress = 0.8196939690772004
+I1204 14:33:57.062685 137274321021824 utils.py:1231] [92300] epoch = 73.77274000969429
+I1204 14:33:57.062740 137274321021824 utils.py:1231] [92300] img/sec/core = 164.20743555497475
+I1204 14:33:57.062802 137274321021824 utils.py:1231] [92300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.69524486145556
+I1204 14:33:57.062861 137274321021824 utils.py:1231] [92300] core_hours = 160.69524486145556
+I1204 14:33:57.062946 137274321021824 train.py:125] NOTE: Steps:92300/112603 [82.0%]
+Walltime:6d16h43m (0s eval)
+ETA:1d11h20m
+Total train time:8d4h2m
+I1204 14:39:08.857637 137274321021824 utils.py:1231] [92350] l2_params = 244.99992635127342
+I1204 14:39:08.857825 137274321021824 utils.py:1231] [92350] train/loss = 2.7511172592639923
+I1204 14:39:08.857922 137274321021824 utils.py:1231] [92350] l2_grads = 2.21348237991333
+I1204 14:39:08.857995 137274321021824 utils.py:1231] [92350] lr = 9.310596011340265e-05
+I1204 14:39:08.858048 137274321021824 utils.py:1231] [92350] uptime = 578938.220409319
+I1204 14:39:08.858099 137274321021824 utils.py:1231] [92350] examples_seen = 94566400.0
+I1204 14:39:08.858148 137274321021824 utils.py:1231] [92350] progress = 0.8201380069802758
+I1204 14:39:08.858196 137274321021824 utils.py:1231] [92350] epoch = 73.81270357416324
+I1204 14:39:08.858249 137274321021824 utils.py:1231] [92350] img/sec/core = 164.21017461152132
+I1204 14:39:08.858304 137274321021824 utils.py:1231] [92350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.78185473356527
+I1204 14:39:08.858356 137274321021824 utils.py:1231] [92350] core_hours = 160.78185473356527
+I1204 14:39:08.858416 137274321021824 train.py:125] NOTE: Steps:92350/112603 [82.0%]
+Walltime:6d16h48m (0s eval)
+ETA:1d11h15m
+Total train time:8d4h2m
+I1204 14:44:20.680821 137274321021824 utils.py:1231] [92400] l2_params = 244.95381616415935
+I1204 14:44:20.681059 137274321021824 utils.py:1231] [92400] train/loss = 1.676716148853302
+I1204 14:44:20.681158 137274321021824 utils.py:1231] [92400] l2_grads = 2.529646396636963
+I1204 14:44:20.681221 137274321021824 utils.py:1231] [92400] lr = 9.266157356505598e-05
+I1204 14:44:20.681276 137274321021824 utils.py:1231] [92400] uptime = 579250.0436376249
+I1204 14:44:20.681329 137274321021824 utils.py:1231] [92400] examples_seen = 94617600.0
+I1204 14:44:20.681379 137274321021824 utils.py:1231] [92400] progress = 0.8205820448833513
+I1204 14:44:20.681427 137274321021824 utils.py:1231] [92400] epoch = 73.8526671386322
+I1204 14:44:20.681479 137274321021824 utils.py:1231] [92400] img/sec/core = 164.1955933756541
+I1204 14:44:20.681540 137274321021824 utils.py:1231] [92400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.8684722969836
+I1204 14:44:20.681591 137274321021824 utils.py:1231] [92400] core_hours = 160.8684722969836
+I1204 14:44:20.681653 137274321021824 train.py:125] NOTE: Steps:92400/112603 [82.1%]
+Walltime:6d16h54m (0s eval)
+ETA:1d11h10m
+Total train time:8d4h2m
+I1204 14:49:32.420691 137274321021824 utils.py:1231] [92450] l2_params = 244.90791820141962
+I1204 14:49:32.420959 137274321021824 utils.py:1231] [92450] train/loss = 3.878270983695984
+I1204 14:49:32.421094 137274321021824 utils.py:1231] [92450] l2_grads = 2.5707626342773438
+I1204 14:49:32.421166 137274321021824 utils.py:1231] [92450] lr = 9.221814173431647e-05
+I1204 14:49:32.421227 137274321021824 utils.py:1231] [92450] uptime = 579561.78358294
+I1204 14:49:32.421288 137274321021824 utils.py:1231] [92450] examples_seen = 94668800.0
+I1204 14:49:32.421339 137274321021824 utils.py:1231] [92450] progress = 0.8210260827864266
+I1204 14:49:32.421391 137274321021824 utils.py:1231] [92450] epoch = 73.89263070310116
+I1204 14:49:32.421441 137274321021824 utils.py:1231] [92450] img/sec/core = 164.23945910510486
+I1204 14:49:32.421507 137274321021824 utils.py:1231] [92450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 160.9550667262378
+I1204 14:49:32.421557 137274321021824 utils.py:1231] [92450] core_hours = 160.9550667262378
+I1204 14:49:32.421614 137274321021824 train.py:125] NOTE: Steps:92450/112603 [82.1%]
+Walltime:6d16h59m (0s eval)
+ETA:1d11h5m
+Total train time:8d4h2m
+I1204 14:54:44.207596 137274321021824 utils.py:1231] [92500] l2_params = 244.86333345204375
+I1204 14:54:44.207843 137274321021824 utils.py:1231] [92500] train/loss = 3.875770092010498
+I1204 14:54:44.207998 137274321021824 utils.py:1231] [92500] l2_grads = 2.640855073928833
+I1204 14:54:44.208097 137274321021824 utils.py:1231] [92500] lr = 9.177566566049734e-05
+I1204 14:54:44.208168 137274321021824 utils.py:1231] [92500] uptime = 579873.570527636
+I1204 14:54:44.208242 137274321021824 utils.py:1231] [92500] examples_seen = 94720000.0
+I1204 14:54:44.208319 137274321021824 utils.py:1231] [92500] progress = 0.8214701206895021
+I1204 14:54:44.208383 137274321021824 utils.py:1231] [92500] epoch = 73.93259426757011
+I1204 14:54:44.208454 137274321021824 utils.py:1231] [92500] img/sec/core = 164.2147013240864
+I1204 14:54:44.208526 137274321021824 utils.py:1231] [92500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.04167421087556
+I1204 14:54:44.208594 137274321021824 utils.py:1231] [92500] core_hours = 161.04167421087556
+I1204 14:54:44.208664 137274321021824 train.py:125] NOTE: Steps:92500/112603 [82.1%]
+Walltime:6d17h4m (0s eval)
+ETA:1d10h59m
+Total train time:8d4h2m
+I1204 14:54:44.208787 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:92500/112603 [82.1%]
+Walltime:6d17h4m (0s eval)
+ETA:1d10h59m
+Total train time:8d4h2m
+I1204 14:56:22.446197 137274321021824 utils.py:1231] [92500] val/acc@1 = 0.7471898915816326
+I1204 14:56:22.446480 137274321021824 utils.py:1231] [92500] val/loss = 0.9926831212883093
+I1204 14:56:22.446654 137274321021824 utils.py:1231] [92500] z/secs/eval/val = 98.23778849595692
+I1204 14:56:22.446726 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.23778849595692
+I1204 15:01:34.234187 137274321021824 utils.py:1231] [92550] l2_params = 244.81793157124162
+I1204 15:01:34.234443 137274321021824 utils.py:1231] [92550] train/loss = 3.9441870152950287
+I1204 15:01:34.234562 137274321021824 utils.py:1231] [92550] l2_grads = 2.5447309017181396
+I1204 15:01:34.234655 137274321021824 utils.py:1231] [92550] lr = 9.133414638067184e-05
+I1204 15:01:34.234729 137274321021824 utils.py:1231] [92550] uptime = 580283.59708909
+I1204 15:01:34.234792 137274321021824 utils.py:1231] [92550] examples_seen = 94771200.0
+I1204 15:01:34.234851 137274321021824 utils.py:1231] [92550] progress = 0.8219141585925774
+I1204 15:01:34.234916 137274321021824 utils.py:1231] [92550] epoch = 73.97255783203907
+I1204 15:01:34.234975 137274321021824 utils.py:1231] [92550] img/sec/core = 124.86995920077037
+I1204 15:01:34.235039 137274321021824 utils.py:1231] [92550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.15557047794613
+I1204 15:01:34.235093 137274321021824 utils.py:1231] [92550] core_hours = 161.15557047794613
+I1204 15:01:34.235161 137274321021824 train.py:125] NOTE: Steps:92550/112603 [82.2%]
+Walltime:6d17h11m (0s eval)
+ETA:1d10h55m
+Total train time:8d4h4m
+I1204 15:06:46.013684 137274321021824 utils.py:1231] [92600] l2_params = 244.77180285360137
+I1204 15:06:46.013947 137274321021824 utils.py:1231] [92600] train/loss = 1.6013252139091492
+I1204 15:06:46.014135 137274321021824 utils.py:1231] [92600] l2_grads = 2.4285433292388916
+I1204 15:06:46.014219 137274321021824 utils.py:1231] [92600] lr = 9.08935849296702e-05
+I1204 15:06:46.014305 137274321021824 utils.py:1231] [92600] uptime = 580595.376657215
+I1204 15:06:46.014389 137274321021824 utils.py:1231] [92600] examples_seen = 94822400.0
+I1204 15:06:46.014460 137274321021824 utils.py:1231] [92600] progress = 0.8223581964956529
+I1204 15:06:46.014531 137274321021824 utils.py:1231] [92600] epoch = 74.01252139650802
+I1204 15:06:46.014596 137274321021824 utils.py:1231] [92600] img/sec/core = 164.2185865735986
+I1204 15:06:46.014667 137274321021824 utils.py:1231] [92600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.24217591353636
+I1204 15:06:46.014739 137274321021824 utils.py:1231] [92600] core_hours = 161.24217591353636
+I1204 15:06:46.014816 137274321021824 train.py:125] NOTE: Steps:92600/112603 [82.2%]
+Walltime:6d17h16m (0s eval)
+ETA:1d10h49m
+Total train time:8d4h4m
+I1204 15:11:57.785767 137274321021824 utils.py:1231] [92650] l2_params = 244.73165004462356
+I1204 15:11:57.786063 137274321021824 utils.py:1231] [92650] train/loss = 1.7553976476192474
+I1204 15:11:57.786247 137274321021824 utils.py:1231] [92650] l2_grads = 2.6499056816101074
+I1204 15:11:57.786357 137274321021824 utils.py:1231] [92650] lr = 9.045398234007823e-05
+I1204 15:11:57.786448 137274321021824 utils.py:1231] [92650] uptime = 580907.148808435
+I1204 15:11:57.786525 137274321021824 utils.py:1231] [92650] examples_seen = 94873600.0
+I1204 15:11:57.786596 137274321021824 utils.py:1231] [92650] progress = 0.8228022343987282
+I1204 15:11:57.786689 137274321021824 utils.py:1231] [92650] epoch = 74.05248496097698
+I1204 15:11:57.786760 137274321021824 utils.py:1231] [92650] img/sec/core = 164.2224932523552
+I1204 15:11:57.786845 137274321021824 utils.py:1231] [92650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.32877928887527
+I1204 15:11:57.786939 137274321021824 utils.py:1231] [92650] core_hours = 161.32877928887527
+I1204 15:11:57.787089 137274321021824 train.py:125] NOTE: Steps:92650/112603 [82.3%]
+Walltime:6d17h21m (0s eval)
+ETA:1d10h44m
+Total train time:8d4h4m
+I1204 15:17:09.566317 137274321021824 utils.py:1231] [92700] l2_params = 244.68664133194062
+I1204 15:17:09.566573 137274321021824 utils.py:1231] [92700] train/loss = 1.5686461478471756
+I1204 15:17:09.566720 137274321021824 utils.py:1231] [92700] l2_grads = 2.519192934036255
+I1204 15:17:09.566785 137274321021824 utils.py:1231] [92700] lr = 9.001533964223385e-05
+I1204 15:17:09.566838 137274321021824 utils.py:1231] [92700] uptime = 581218.929199856
+I1204 15:17:09.566897 137274321021824 utils.py:1231] [92700] examples_seen = 94924800.0
+I1204 15:17:09.566949 137274321021824 utils.py:1231] [92700] progress = 0.8232462723018037
+I1204 15:17:09.566998 137274321021824 utils.py:1231] [92700] epoch = 74.09244852544595
+I1204 15:17:09.567048 137274321021824 utils.py:1231] [92700] img/sec/core = 164.21815293334313
+I1204 15:17:09.567104 137274321021824 utils.py:1231] [92700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.4153849531589
+I1204 15:17:09.567158 137274321021824 utils.py:1231] [92700] core_hours = 161.4153849531589
+I1204 15:17:09.567219 137274321021824 train.py:125] NOTE: Steps:92700/112603 [82.3%]
+Walltime:6d17h26m (0s eval)
+ETA:1d10h39m
+Total train time:8d4h4m
+I1204 15:22:21.356161 137274321021824 utils.py:1231] [92750] l2_params = 244.64365788858026
+I1204 15:22:21.356457 137274321021824 utils.py:1231] [92750] train/loss = 1.4933629930019379
+I1204 15:22:21.356629 137274321021824 utils.py:1231] [92750] l2_grads = 2.572333574295044
+I1204 15:22:21.356714 137274321021824 utils.py:1231] [92750] lr = 8.957765786422552e-05
+I1204 15:22:21.356773 137274321021824 utils.py:1231] [92750] uptime = 581530.719134861
+I1204 15:22:21.356834 137274321021824 utils.py:1231] [92750] examples_seen = 94976000.0
+I1204 15:22:21.356893 137274321021824 utils.py:1231] [92750] progress = 0.823690310204879
+I1204 15:22:21.356944 137274321021824 utils.py:1231] [92750] epoch = 74.1324120899149
+I1204 15:22:21.356995 137274321021824 utils.py:1231] [92750] img/sec/core = 164.21312637682377
+I1204 15:22:21.357052 137274321021824 utils.py:1231] [92750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.50199326843807
+I1204 15:22:21.357102 137274321021824 utils.py:1231] [92750] core_hours = 161.50199326843807
+I1204 15:22:21.357164 137274321021824 train.py:125] NOTE: Steps:92750/112603 [82.4%]
+Walltime:6d17h32m (0s eval)
+ETA:1d10h34m
+Total train time:8d4h4m
+I1204 15:27:33.174638 137274321021824 utils.py:1231] [92800] l2_params = 244.59878352166643
+I1204 15:27:33.174937 137274321021824 utils.py:1231] [92800] train/loss = 2.903464615345001
+I1204 15:27:33.175099 137274321021824 utils.py:1231] [92800] l2_grads = 2.282848596572876
+I1204 15:27:33.175182 137274321021824 utils.py:1231] [92800] lr = 8.91409380318896e-05
+I1204 15:27:33.175249 137274321021824 utils.py:1231] [92800] uptime = 581842.537610703
+I1204 15:27:33.175305 137274321021824 utils.py:1231] [92800] examples_seen = 95027200.0
+I1204 15:27:33.175361 137274321021824 utils.py:1231] [92800] progress = 0.8241343481079545
+I1204 15:27:33.175417 137274321021824 utils.py:1231] [92800] epoch = 74.17237565438386
+I1204 15:27:33.175467 137274321021824 utils.py:1231] [92800] img/sec/core = 164.1980959009678
+I1204 15:27:33.175534 137274321021824 utils.py:1231] [92800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.5886095117275
+I1204 15:27:33.175591 137274321021824 utils.py:1231] [92800] core_hours = 161.5886095117275
+I1204 15:27:33.175653 137274321021824 train.py:125] NOTE: Steps:92800/112603 [82.4%]
+Walltime:6d17h37m (0s eval)
+ETA:1d10h28m
+Total train time:8d4h4m
+I1204 15:32:44.949021 137274321021824 utils.py:1231] [92850] l2_params = 244.5541512735972
+I1204 15:32:44.949267 137274321021824 utils.py:1231] [92850] train/loss = 1.6604070365428925
+I1204 15:32:44.949499 137274321021824 utils.py:1231] [92850] l2_grads = 2.538813591003418
+I1204 15:32:44.949624 137274321021824 utils.py:1231] [92850] lr = 8.87051811688075e-05
+I1204 15:32:44.949705 137274321021824 utils.py:1231] [92850] uptime = 582154.312057814
+I1204 15:32:44.949783 137274321021824 utils.py:1231] [92850] examples_seen = 95078400.0
+I1204 15:32:44.949850 137274321021824 utils.py:1231] [92850] progress = 0.82457838601103
+I1204 15:32:44.949922 137274321021824 utils.py:1231] [92850] epoch = 74.21233921885282
+I1204 15:32:44.949988 137274321021824 utils.py:1231] [92850] img/sec/core = 164.2212839263896
+I1204 15:32:44.950052 137274321021824 utils.py:1231] [92850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.67521352481387
+I1204 15:32:44.950108 137274321021824 utils.py:1231] [92850] core_hours = 161.67521352481387
+I1204 15:32:44.950175 137274321021824 train.py:125] NOTE: Steps:92850/112603 [82.5%]
+Walltime:6d17h42m (0s eval)
+ETA:1d10h23m
+Total train time:8d4h4m
+I1204 15:37:56.690245 137274321021824 utils.py:1231] [92900] l2_params = 244.50765262565258
+I1204 15:37:56.690512 137274321021824 utils.py:1231] [92900] train/loss = 1.7645489871501923
+I1204 15:37:56.690633 137274321021824 utils.py:1231] [92900] l2_grads = 2.401473045349121
+I1204 15:37:56.690728 137274321021824 utils.py:1231] [92900] lr = 8.827038829630394e-05
+I1204 15:37:56.690798 137274321021824 utils.py:1231] [92900] uptime = 582466.053156196
+I1204 15:37:56.690859 137274321021824 utils.py:1231] [92900] examples_seen = 95129600.0
+I1204 15:37:56.690927 137274321021824 utils.py:1231] [92900] progress = 0.8250224239141053
+I1204 15:37:56.690983 137274321021824 utils.py:1231] [92900] epoch = 74.25230278332177
+I1204 15:37:56.691041 137274321021824 utils.py:1231] [92900] img/sec/core = 164.23885161671382
+I1204 15:37:56.691100 137274321021824 utils.py:1231] [92900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.76180827436443
+I1204 15:37:56.691156 137274321021824 utils.py:1231] [92900] core_hours = 161.76180827436443
+I1204 15:37:56.691224 137274321021824 train.py:125] NOTE: Steps:92900/112603 [82.5%]
+Walltime:6d17h47m (0s eval)
+ETA:1d10h18m
+Total train time:8d4h4m
+I1204 15:43:08.478106 137274321021824 utils.py:1231] [92950] l2_params = 244.46348518394066
+I1204 15:43:08.478338 137274321021824 utils.py:1231] [92950] train/loss = 3.099687486886978
+I1204 15:43:08.478470 137274321021824 utils.py:1231] [92950] l2_grads = 2.286783218383789
+I1204 15:43:08.478545 137274321021824 utils.py:1231] [92950] lr = 8.783656043344395e-05
+I1204 15:43:08.478608 137274321021824 utils.py:1231] [92950] uptime = 582777.840968898
+I1204 15:43:08.478671 137274321021824 utils.py:1231] [92950] examples_seen = 95180800.0
+I1204 15:43:08.478730 137274321021824 utils.py:1231] [92950] progress = 0.8254664618171808
+I1204 15:43:08.478788 137274321021824 utils.py:1231] [92950] epoch = 74.29226634779073
+I1204 15:43:08.478847 137274321021824 utils.py:1231] [92950] img/sec/core = 164.2142441562796
+I1204 15:43:08.478920 137274321021824 utils.py:1231] [92950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.848416000115
+I1204 15:43:08.478979 137274321021824 utils.py:1231] [92950] core_hours = 161.848416000115
+I1204 15:43:08.479047 137274321021824 train.py:125] NOTE: Steps:92950/112603 [82.5%]
+Walltime:6d17h52m (0s eval)
+ETA:1d10h13m
+Total train time:8d4h4m
+I1204 15:48:20.259135 137274321021824 utils.py:1231] [93000] l2_params = 244.41936361347678
+I1204 15:48:20.259342 137274321021824 utils.py:1231] [93000] train/loss = 3.673387110233307
+I1204 15:48:20.259473 137274321021824 utils.py:1231] [93000] l2_grads = 2.460174798965454
+I1204 15:48:20.259564 137274321021824 utils.py:1231] [93000] lr = 8.740369859703118e-05
+I1204 15:48:20.259637 137274321021824 utils.py:1231] [93000] uptime = 583089.621995898
+I1204 15:48:20.259719 137274321021824 utils.py:1231] [93000] examples_seen = 95232000.0
+I1204 15:48:20.259791 137274321021824 utils.py:1231] [93000] progress = 0.8259104997202561
+I1204 15:48:20.259860 137274321021824 utils.py:1231] [93000] epoch = 74.33222991225968
+I1204 15:48:20.259948 137274321021824 utils.py:1231] [93000] img/sec/core = 164.2178181676565
+I1204 15:48:20.260023 137274321021824 utils.py:1231] [93000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 161.9350218409483
+I1204 15:48:20.260112 137274321021824 utils.py:1231] [93000] core_hours = 161.9350218409483
+I1204 15:48:20.260180 137274321021824 train.py:125] NOTE: Steps:93000/112603 [82.6%]
+Walltime:6d17h58m (0s eval)
+ETA:1d10h8m
+Total train time:8d4h4m
+I1204 15:53:32.397732 137274321021824 utils.py:1231] [93050] l2_params = 244.37683387943866
+I1204 15:53:32.398043 137274321021824 utils.py:1231] [93050] train/loss = 3.9299648106098175
+I1204 15:53:32.398212 137274321021824 utils.py:1231] [93050] l2_grads = 2.5416672229766846
+I1204 15:53:32.398295 137274321021824 utils.py:1231] [93050] lr = 8.697180380160461e-05
+I1204 15:53:32.398353 137274321021824 utils.py:1231] [93050] uptime = 583401.760714379
+I1204 15:53:32.398430 137274321021824 utils.py:1231] [93050] examples_seen = 95283200.0
+I1204 15:53:32.398479 137274321021824 utils.py:1231] [93050] progress = 0.8263545376233316
+I1204 15:53:32.398528 137274321021824 utils.py:1231] [93050] epoch = 74.37219347672864
+I1204 15:53:32.398579 137274321021824 utils.py:1231] [93050] img/sec/core = 164.02963480196019
+I1204 15:53:32.398636 137274321021824 utils.py:1231] [93050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.0217270405264
+I1204 15:53:32.398689 137274321021824 utils.py:1231] [93050] core_hours = 162.0217270405264
+I1204 15:53:32.398749 137274321021824 train.py:125] NOTE: Steps:93050/112603 [82.6%]
+Walltime:6d18h3m (0s eval)
+ETA:1d10h2m
+Total train time:8d4h4m
+I1204 15:58:44.187097 137274321021824 utils.py:1231] [93100] l2_params = 244.33553212828213
+I1204 15:58:44.187461 137274321021824 utils.py:1231] [93100] train/loss = 3.539644777774811
+I1204 15:58:44.187689 137274321021824 utils.py:1231] [93100] l2_grads = 2.4676601886749268
+I1204 15:58:44.187810 137274321021824 utils.py:1231] [93100] lr = 8.654087705943698e-05
+I1204 15:58:44.187908 137274321021824 utils.py:1231] [93100] uptime = 583713.550263139
+I1204 15:58:44.188033 137274321021824 utils.py:1231] [93100] examples_seen = 95334400.0
+I1204 15:58:44.188129 137274321021824 utils.py:1231] [93100] progress = 0.8267985755264069
+I1204 15:58:44.188234 137274321021824 utils.py:1231] [93100] epoch = 74.4121570411976
+I1204 15:58:44.188314 137274321021824 utils.py:1231] [93100] img/sec/core = 164.21332980414482
+I1204 15:58:44.188410 137274321021824 utils.py:1231] [93100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.10833524851526
+I1204 15:58:44.188503 137274321021824 utils.py:1231] [93100] core_hours = 162.10833524851526
+I1204 15:58:44.188603 137274321021824 train.py:125] NOTE: Steps:93100/112603 [82.7%]
+Walltime:6d18h8m (0s eval)
+ETA:1d9h57m
+Total train time:8d4h4m
+I1204 16:03:55.979933 137274321021824 utils.py:1231] [93150] l2_params = 244.29266474041603
+I1204 16:03:55.980174 137274321021824 utils.py:1231] [93150] train/loss = 3.008532464504242
+I1204 16:03:55.980318 137274321021824 utils.py:1231] [93150] l2_grads = 2.4129202365875244
+I1204 16:03:55.980416 137274321021824 utils.py:1231] [93150] lr = 8.611091938053228e-05
+I1204 16:03:55.980485 137274321021824 utils.py:1231] [93150] uptime = 584025.342846206
+I1204 16:03:55.980549 137274321021824 utils.py:1231] [93150] examples_seen = 95385600.0
+I1204 16:03:55.980608 137274321021824 utils.py:1231] [93150] progress = 0.8272426134294824
+I1204 16:03:55.980674 137274321021824 utils.py:1231] [93150] epoch = 74.45212060566655
+I1204 16:03:55.980742 137274321021824 utils.py:1231] [93150] img/sec/core = 164.21173171070464
+I1204 16:03:55.980808 137274321021824 utils.py:1231] [93150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.19494429936722
+I1204 16:03:55.980867 137274321021824 utils.py:1231] [93150] core_hours = 162.19494429936722
+I1204 16:03:55.980946 137274321021824 train.py:125] NOTE: Steps:93150/112603 [82.7%]
+Walltime:6d18h13m (0s eval)
+ETA:1d9h52m
+Total train time:8d4h4m
+I1204 16:09:07.772562 137274321021824 utils.py:1231] [93200] l2_params = 244.24906090047782
+I1204 16:09:07.772781 137274321021824 utils.py:1231] [93200] train/loss = 1.5425866097211838
+I1204 16:09:07.772894 137274321021824 utils.py:1231] [93200] l2_grads = 2.475569486618042
+I1204 16:09:07.772970 137274321021824 utils.py:1231] [93200] lr = 8.568193177262272e-05
+I1204 16:09:07.773030 137274321021824 utils.py:1231] [93200] uptime = 584337.1353917
+I1204 16:09:07.773094 137274321021824 utils.py:1231] [93200] examples_seen = 95436800.0
+I1204 16:09:07.773154 137274321021824 utils.py:1231] [93200] progress = 0.8276866513325577
+I1204 16:09:07.773210 137274321021824 utils.py:1231] [93200] epoch = 74.49208417013551
+I1204 16:09:07.773268 137274321021824 utils.py:1231] [93200] img/sec/core = 164.2117514993558
+I1204 16:09:07.773329 137274321021824 utils.py:1231] [93200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.2815533397822
+I1204 16:09:07.773386 137274321021824 utils.py:1231] [93200] core_hours = 162.2815533397822
+I1204 16:09:07.773447 137274321021824 train.py:125] NOTE: Steps:93200/112603 [82.8%]
+Walltime:6d18h18m (0s eval)
+ETA:1d9h47m
+Total train time:8d4h4m
+I1204 16:14:19.556770 137274321021824 utils.py:1231] [93250] l2_params = 244.20606315666566
+I1204 16:14:19.556973 137274321021824 utils.py:1231] [93250] train/loss = 1.6290929913520813
+I1204 16:14:19.557080 137274321021824 utils.py:1231] [93250] l2_grads = 2.5209758281707764
+I1204 16:14:19.557144 137274321021824 utils.py:1231] [93250] lr = 8.525391524116735e-05
+I1204 16:14:19.557197 137274321021824 utils.py:1231] [93250] uptime = 584648.91955861
+I1204 16:14:19.557252 137274321021824 utils.py:1231] [93250] examples_seen = 95488000.0
+I1204 16:14:19.557304 137274321021824 utils.py:1231] [93250] progress = 0.8281306892356332
+I1204 16:14:19.557353 137274321021824 utils.py:1231] [93250] epoch = 74.53204773460446
+I1204 16:14:19.557405 137274321021824 utils.py:1231] [93250] img/sec/core = 164.21616436597813
+I1204 16:14:19.557463 137274321021824 utils.py:1231] [93250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.36816005281278
+I1204 16:14:19.557516 137274321021824 utils.py:1231] [93250] core_hours = 162.36816005281278
+I1204 16:14:19.557579 137274321021824 train.py:125] NOTE: Steps:93250/112603 [82.8%]
+Walltime:6d18h24m (0s eval)
+ETA:1d9h41m
+Total train time:8d4h4m
+I1204 16:19:31.348110 137274321021824 utils.py:1231] [93300] l2_params = 244.1641841342889
+I1204 16:19:31.348334 137274321021824 utils.py:1231] [93300] train/loss = 1.730505645275116
+I1204 16:19:31.348438 137274321021824 utils.py:1231] [93300] l2_grads = 2.5153932571411133
+I1204 16:19:31.348503 137274321021824 utils.py:1231] [93300] lr = 8.482687078934878e-05
+I1204 16:19:31.348555 137274321021824 utils.py:1231] [93300] uptime = 584960.710916716
+I1204 16:19:31.348608 137274321021824 utils.py:1231] [93300] examples_seen = 95539200.0
+I1204 16:19:31.348659 137274321021824 utils.py:1231] [93300] progress = 0.8285747271387086
+I1204 16:19:31.348707 137274321021824 utils.py:1231] [93300] epoch = 74.57201129907342
+I1204 16:19:31.348757 137274321021824 utils.py:1231] [93300] img/sec/core = 164.21237686319006
+I1204 16:19:31.348813 137274321021824 utils.py:1231] [93300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.45476876339777
+I1204 16:19:31.348864 137274321021824 utils.py:1231] [93300] core_hours = 162.45476876339777
+I1204 16:19:31.348938 137274321021824 train.py:125] NOTE: Steps:93300/112603 [82.9%]
+Walltime:6d18h29m (0s eval)
+ETA:1d9h36m
+Total train time:8d4h4m
+I1204 16:24:43.135593 137274321021824 utils.py:1231] [93350] l2_params = 244.12532914706424
+I1204 16:24:43.135868 137274321021824 utils.py:1231] [93350] train/loss = 1.7518105208873749
+I1204 16:24:43.136055 137274321021824 utils.py:1231] [93350] l2_grads = 2.448993444442749
+I1204 16:24:43.136169 137274321021824 utils.py:1231] [93350] lr = 8.440079941807163e-05
+I1204 16:24:43.136268 137274321021824 utils.py:1231] [93350] uptime = 585272.498619273
+I1204 16:24:43.136390 137274321021824 utils.py:1231] [93350] examples_seen = 95590400.0
+I1204 16:24:43.136461 137274321021824 utils.py:1231] [93350] progress = 0.829018765041784
+I1204 16:24:43.136526 137274321021824 utils.py:1231] [93350] epoch = 74.61197486354239
+I1204 16:24:43.136596 137274321021824 utils.py:1231] [93350] img/sec/core = 164.21430216815847
+I1204 16:24:43.136669 137274321021824 utils.py:1231] [93350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.54137645855246
+I1204 16:24:43.136742 137274321021824 utils.py:1231] [93350] core_hours = 162.54137645855246
+I1204 16:24:43.136837 137274321021824 train.py:125] NOTE: Steps:93350/112603 [82.9%]
+Walltime:6d18h34m (0s eval)
+ETA:1d9h31m
+Total train time:8d4h4m
+I1204 16:29:54.925717 137274321021824 utils.py:1231] [93400] l2_params = 244.08353992907678
+I1204 16:29:54.925976 137274321021824 utils.py:1231] [93400] train/loss = 1.5284664034843445
+I1204 16:29:54.926098 137274321021824 utils.py:1231] [93400] l2_grads = 2.608098030090332
+I1204 16:29:54.926196 137274321021824 utils.py:1231] [93400] lr = 8.397570212595977e-05
+I1204 16:29:54.926308 137274321021824 utils.py:1231] [93400] uptime = 585584.288664479
+I1204 16:29:54.926405 137274321021824 utils.py:1231] [93400] examples_seen = 95641600.0
+I1204 16:29:54.926483 137274321021824 utils.py:1231] [93400] progress = 0.8294628029448594
+I1204 16:29:54.926568 137274321021824 utils.py:1231] [93400] epoch = 74.65193842801133
+I1204 16:29:54.926650 137274321021824 utils.py:1231] [93400] img/sec/core = 164.21306833628387
+I1204 16:29:54.926762 137274321021824 utils.py:1231] [93400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.6279848044431
+I1204 16:29:54.926829 137274321021824 utils.py:1231] [93400] core_hours = 162.6279848044431
+I1204 16:29:54.926913 137274321021824 train.py:125] NOTE: Steps:93400/112603 [82.9%]
+Walltime:6d18h39m (0s eval)
+ETA:1d9h26m
+Total train time:8d4h4m
+I1204 16:35:06.651880 137274321021824 utils.py:1231] [93450] l2_params = 244.03792844918496
+I1204 16:35:06.652101 137274321021824 utils.py:1231] [93450] train/loss = 3.8971880972385406
+I1204 16:35:06.652213 137274321021824 utils.py:1231] [93450] l2_grads = 2.6329963207244873
+I1204 16:35:06.652308 137274321021824 utils.py:1231] [93450] lr = 8.35515799093538e-05
+I1204 16:35:06.652376 137274321021824 utils.py:1231] [93450] uptime = 585896.014734002
+I1204 16:35:06.652440 137274321021824 utils.py:1231] [93450] examples_seen = 95692800.0
+I1204 16:35:06.652506 137274321021824 utils.py:1231] [93450] progress = 0.8299068408479348
+I1204 16:35:06.652574 137274321021824 utils.py:1231] [93450] epoch = 74.6919019924803
+I1204 16:35:06.652638 137274321021824 utils.py:1231] [93450] img/sec/core = 164.2467698590666
+I1204 16:35:06.652709 137274321021824 utils.py:1231] [93450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.71457537931053
+I1204 16:35:06.652772 137274321021824 utils.py:1231] [93450] core_hours = 162.71457537931053
+I1204 16:35:06.652845 137274321021824 train.py:125] NOTE: Steps:93450/112603 [83.0%]
+Walltime:6d18h44m (0s eval)
+ETA:1d9h20m
+Total train time:8d4h4m
+I1204 16:40:18.429058 137274321021824 utils.py:1231] [93500] l2_params = 243.99837749937933
+I1204 16:40:18.429272 137274321021824 utils.py:1231] [93500] train/loss = 2.9434246122837067
+I1204 16:40:18.429380 137274321021824 utils.py:1231] [93500] l2_grads = 2.2795298099517822
+I1204 16:40:18.429474 137274321021824 utils.py:1231] [93500] lr = 8.312843376230925e-05
+I1204 16:40:18.429538 137274321021824 utils.py:1231] [93500] uptime = 586207.791898764
+I1204 16:40:18.429601 137274321021824 utils.py:1231] [93500] examples_seen = 95744000.0
+I1204 16:40:18.429661 137274321021824 utils.py:1231] [93500] progress = 0.8303508787510102
+I1204 16:40:18.429724 137274321021824 utils.py:1231] [93500] epoch = 74.73186555694924
+I1204 16:40:18.429785 137274321021824 utils.py:1231] [93500] img/sec/core = 164.21985246764388
+I1204 16:40:18.429852 137274321021824 utils.py:1231] [93500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.8011801473
+I1204 16:40:18.429921 137274321021824 utils.py:1231] [93500] core_hours = 162.8011801473
+I1204 16:40:18.429991 137274321021824 train.py:125] NOTE: Steps:93500/112603 [83.0%]
+Walltime:6d18h50m (0s eval)
+ETA:1d9h15m
+Total train time:8d4h4m
+I1204 16:45:30.215425 137274321021824 utils.py:1231] [93550] l2_params = 243.9572908499884
+I1204 16:45:30.215632 137274321021824 utils.py:1231] [93550] train/loss = 1.5995029658079147
+I1204 16:45:30.215728 137274321021824 utils.py:1231] [93550] l2_grads = 2.379781723022461
+I1204 16:45:30.215790 137274321021824 utils.py:1231] [93550] lr = 8.270626467659365e-05
+I1204 16:45:30.215841 137274321021824 utils.py:1231] [93550] uptime = 586519.578202953
+I1204 16:45:30.215899 137274321021824 utils.py:1231] [93550] examples_seen = 95795200.0
+I1204 16:45:30.215949 137274321021824 utils.py:1231] [93550] progress = 0.8307949166540856
+I1204 16:45:30.215997 137274321021824 utils.py:1231] [93550] epoch = 74.77182912141821
+I1204 16:45:30.216047 137274321021824 utils.py:1231] [93550] img/sec/core = 164.21503867264784
+I1204 16:45:30.216103 137274321021824 utils.py:1231] [93550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.88778745401913
+I1204 16:45:30.216153 137274321021824 utils.py:1231] [93550] core_hours = 162.88778745401913
+I1204 16:45:30.216213 137274321021824 train.py:125] NOTE: Steps:93550/112603 [83.1%]
+Walltime:6d18h55m (0s eval)
+ETA:1d9h10m
+Total train time:8d4h4m
+I1204 16:50:42.096729 137274321021824 utils.py:1231] [93600] l2_params = 243.91489629259942
+I1204 16:50:42.097026 137274321021824 utils.py:1231] [93600] train/loss = 1.5575932562351227
+I1204 16:50:42.097213 137274321021824 utils.py:1231] [93600] l2_grads = 2.509843111038208
+I1204 16:50:42.097319 137274321021824 utils.py:1231] [93600] lr = 8.228507364168441e-05
+I1204 16:50:42.097413 137274321021824 utils.py:1231] [93600] uptime = 586831.459765312
+I1204 16:50:42.097496 137274321021824 utils.py:1231] [93600] examples_seen = 95846400.0
+I1204 16:50:42.097579 137274321021824 utils.py:1231] [93600] progress = 0.831238954557161
+I1204 16:50:42.097642 137274321021824 utils.py:1231] [93600] epoch = 74.81179268588717
+I1204 16:50:42.097720 137274321021824 utils.py:1231] [93600] img/sec/core = 164.1648823762681
+I1204 16:50:42.097784 137274321021824 utils.py:1231] [93600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 162.9744212213411
+I1204 16:50:42.097854 137274321021824 utils.py:1231] [93600] core_hours = 162.9744212213411
+I1204 16:50:42.097977 137274321021824 train.py:125] NOTE: Steps:93600/112603 [83.1%]
+Walltime:6d19h0m (0s eval)
+ETA:1d9h5m
+Total train time:8d4h3m
+I1204 16:55:53.891560 137274321021824 utils.py:1231] [93650] l2_params = 243.8718698360342
+I1204 16:55:53.891855 137274321021824 utils.py:1231] [93650] train/loss = 1.6828245520591736
+I1204 16:55:53.892001 137274321021824 utils.py:1231] [93650] l2_grads = 2.7037758827209473
+I1204 16:55:53.892068 137274321021824 utils.py:1231] [93650] lr = 8.186486164476726e-05
+I1204 16:55:53.892122 137274321021824 utils.py:1231] [93650] uptime = 587143.254483591
+I1204 16:55:53.892183 137274321021824 utils.py:1231] [93650] examples_seen = 95897600.0
+I1204 16:55:53.892235 137274321021824 utils.py:1231] [93650] progress = 0.8316829924602364
+I1204 16:55:53.892287 137274321021824 utils.py:1231] [93650] epoch = 74.85175625035612
+I1204 16:55:53.892343 137274321021824 utils.py:1231] [93650] img/sec/core = 164.21060716685452
+I1204 16:55:53.892415 137274321021824 utils.py:1231] [93650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.0610308653075
+I1204 16:55:53.892474 137274321021824 utils.py:1231] [93650] core_hours = 163.0610308653075
+I1204 16:55:53.892539 137274321021824 train.py:125] NOTE: Steps:93650/112603 [83.2%]
+Walltime:6d19h5m (0s eval)
+ETA:1d9h0m
+Total train time:8d4h3m
+I1204 17:01:05.691008 137274321021824 utils.py:1231] [93700] l2_params = 243.8338944912154
+I1204 17:01:05.691317 137274321021824 utils.py:1231] [93700] train/loss = 1.6242075115442276
+I1204 17:01:05.691501 137274321021824 utils.py:1231] [93700] l2_grads = 2.4624054431915283
+I1204 17:01:05.691614 137274321021824 utils.py:1231] [93700] lr = 8.144562967073252e-05
+I1204 17:01:05.691683 137274321021824 utils.py:1231] [93700] uptime = 587455.0540442329
+I1204 17:01:05.691753 137274321021824 utils.py:1231] [93700] examples_seen = 95948800.0
+I1204 17:01:05.691812 137274321021824 utils.py:1231] [93700] progress = 0.8321270303633118
+I1204 17:01:05.691875 137274321021824 utils.py:1231] [93700] epoch = 74.89171981482508
+I1204 17:01:05.691950 137274321021824 utils.py:1231] [93700] img/sec/core = 164.2080569150921
+I1204 17:01:05.692012 137274321021824 utils.py:1231] [93700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.1476418543747
+I1204 17:01:05.692070 137274321021824 utils.py:1231] [93700] core_hours = 163.1476418543747
+I1204 17:01:05.692139 137274321021824 train.py:125] NOTE: Steps:93700/112603 [83.2%]
+Walltime:6d19h10m (0s eval)
+ETA:1d8h54m
+Total train time:8d4h3m
+I1204 17:06:17.496247 137274321021824 utils.py:1231] [93750] l2_params = 243.79395935624544
+I1204 17:06:17.496533 137274321021824 utils.py:1231] [93750] train/loss = 1.6358218044042587
+I1204 17:06:17.496706 137274321021824 utils.py:1231] [93750] l2_grads = 2.686277151107788
+I1204 17:06:17.496795 137274321021824 utils.py:1231] [93750] lr = 8.10273787021741e-05
+I1204 17:06:17.496876 137274321021824 utils.py:1231] [93750] uptime = 587766.85923443
+I1204 17:06:17.496957 137274321021824 utils.py:1231] [93750] examples_seen = 96000000.0
+I1204 17:06:17.497026 137274321021824 utils.py:1231] [93750] progress = 0.8325710682663873
+I1204 17:06:17.497091 137274321021824 utils.py:1231] [93750] epoch = 74.93168337929403
+I1204 17:06:17.497159 137274321021824 utils.py:1231] [93750] img/sec/core = 164.2050921847682
+I1204 17:06:17.497235 137274321021824 utils.py:1231] [93750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.2342544072072
+I1204 17:06:17.497297 137274321021824 utils.py:1231] [93750] core_hours = 163.2342544072072
+I1204 17:06:17.497377 137274321021824 train.py:125] NOTE: Steps:93750/112603 [83.3%]
+Walltime:6d19h16m (0s eval)
+ETA:1d8h49m
+Total train time:8d4h3m
+I1204 17:11:29.289712 137274321021824 utils.py:1231] [93800] l2_params = 243.7511782819439
+I1204 17:11:29.289913 137274321021824 utils.py:1231] [93800] train/loss = 1.6797759979963303
+I1204 17:11:29.290009 137274321021824 utils.py:1231] [93800] l2_grads = 2.5885517597198486
+I1204 17:11:29.290070 137274321021824 utils.py:1231] [93800] lr = 8.061010971938606e-05
+I1204 17:11:29.290121 137274321021824 utils.py:1231] [93800] uptime = 588078.652483109
+I1204 17:11:29.290190 137274321021824 utils.py:1231] [93800] examples_seen = 96051200.0
+I1204 17:11:29.290238 137274321021824 utils.py:1231] [93800] progress = 0.8330151061694626
+I1204 17:11:29.290286 137274321021824 utils.py:1231] [93800] epoch = 74.97164694376299
+I1204 17:11:29.290338 137274321021824 utils.py:1231] [93800] img/sec/core = 164.21138115375948
+I1204 17:11:29.290394 137274321021824 utils.py:1231] [93800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.3208636429514
+I1204 17:11:29.290444 137274321021824 utils.py:1231] [93800] core_hours = 163.3208636429514
+I1204 17:11:29.290504 137274321021824 train.py:125] NOTE: Steps:93800/112603 [83.3%]
+Walltime:6d19h21m (0s eval)
+ETA:1d8h44m
+Total train time:8d4h3m
+I1204 17:16:41.109111 137274321021824 utils.py:1231] [93850] l2_params = 243.70881693939006
+I1204 17:16:41.109349 137274321021824 utils.py:1231] [93850] train/loss = 2.2134194523096085
+I1204 17:16:41.109493 137274321021824 utils.py:1231] [93850] l2_grads = 2.4479331970214844
+I1204 17:16:41.109587 137274321021824 utils.py:1231] [93850] lr = 8.019382370036102e-05
+I1204 17:16:41.109668 137274321021824 utils.py:1231] [93850] uptime = 588390.472024607
+I1204 17:16:41.109743 137274321021824 utils.py:1231] [93850] examples_seen = 96102400.0
+I1204 17:16:41.109809 137274321021824 utils.py:1231] [93850] progress = 0.8334591440725381
+I1204 17:16:41.109894 137274321021824 utils.py:1231] [93850] epoch = 75.01161050823195
+I1204 17:16:41.109955 137274321021824 utils.py:1231] [93850] img/sec/core = 164.19753474729924
+I1204 17:16:41.110018 137274321021824 utils.py:1231] [93850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.40748018225636
+I1204 17:16:41.110073 137274321021824 utils.py:1231] [93850] core_hours = 163.40748018225636
+I1204 17:16:41.110141 137274321021824 train.py:125] NOTE: Steps:93850/112603 [83.3%]
+Walltime:6d19h26m (0s eval)
+ETA:1d8h39m
+Total train time:8d4h3m
+I1204 17:21:52.912769 137274321021824 utils.py:1231] [93900] l2_params = 243.66824516303092
+I1204 17:21:52.913002 137274321021824 utils.py:1231] [93900] train/loss = 1.9308399260044098
+I1204 17:21:52.913117 137274321021824 utils.py:1231] [93900] l2_grads = 2.402312994003296
+I1204 17:21:52.913191 137274321021824 utils.py:1231] [93900] lr = 7.977852162078848e-05
+I1204 17:21:52.913252 137274321021824 utils.py:1231] [93900] uptime = 588702.275613208
+I1204 17:21:52.913316 137274321021824 utils.py:1231] [93900] examples_seen = 96153600.0
+I1204 17:21:52.913374 137274321021824 utils.py:1231] [93900] progress = 0.8339031819756134
+I1204 17:21:52.913431 137274321021824 utils.py:1231] [93900] epoch = 75.0515740727009
+I1204 17:21:52.913490 137274321021824 utils.py:1231] [93900] img/sec/core = 164.20593563317544
+I1204 17:21:52.913553 137274321021824 utils.py:1231] [93900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.49409229020108
+I1204 17:21:52.913615 137274321021824 utils.py:1231] [93900] core_hours = 163.49409229020108
+I1204 17:21:52.913686 137274321021824 train.py:125] NOTE: Steps:93900/112603 [83.4%]
+Walltime:6d19h31m (0s eval)
+ETA:1d8h33m
+Total train time:8d4h3m
+I1204 17:27:04.705280 137274321021824 utils.py:1231] [93950] l2_params = 243.62491405540837
+I1204 17:27:04.705513 137274321021824 utils.py:1231] [93950] train/loss = 2.492500752210617
+I1204 17:27:04.705612 137274321021824 utils.py:1231] [93950] l2_grads = 2.1950106620788574
+I1204 17:27:04.705686 137274321021824 utils.py:1231] [93950] lr = 7.936420445405077e-05
+I1204 17:27:04.705740 137274321021824 utils.py:1231] [93950] uptime = 589014.068102224
+I1204 17:27:04.705815 137274321021824 utils.py:1231] [93950] examples_seen = 96204800.0
+I1204 17:27:04.705869 137274321021824 utils.py:1231] [93950] progress = 0.8343472198786889
+I1204 17:27:04.705925 137274321021824 utils.py:1231] [93950] epoch = 75.09153763716986
+I1204 17:27:04.705977 137274321021824 utils.py:1231] [93950] img/sec/core = 164.211781244543
+I1204 17:27:04.706034 137274321021824 utils.py:1231] [93950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.58070131492778
+I1204 17:27:04.706086 137274321021824 utils.py:1231] [93950] core_hours = 163.58070131492778
+I1204 17:27:04.706147 137274321021824 train.py:125] NOTE: Steps:93950/112603 [83.4%]
+Walltime:6d19h36m (0s eval)
+ETA:1d8h28m
+Total train time:8d4h3m
+I1204 17:32:16.433475 137274321021824 utils.py:1231] [94000] l2_params = 243.5848736390617
+I1204 17:32:16.433716 137274321021824 utils.py:1231] [94000] train/loss = 1.6086394786834717
+I1204 17:32:16.433838 137274321021824 utils.py:1231] [94000] l2_grads = 2.4997425079345703
+I1204 17:32:16.433985 137274321021824 utils.py:1231] [94000] lr = 7.895087317122258e-05
+I1204 17:32:16.434087 137274321021824 utils.py:1231] [94000] uptime = 589325.796445298
+I1204 17:32:16.434149 137274321021824 utils.py:1231] [94000] examples_seen = 96256000.0
+I1204 17:32:16.434208 137274321021824 utils.py:1231] [94000] progress = 0.8347912577817642
+I1204 17:32:16.434264 137274321021824 utils.py:1231] [94000] epoch = 75.13150120163881
+I1204 17:32:16.434321 137274321021824 utils.py:1231] [94000] img/sec/core = 164.24557194611458
+I1204 17:32:16.434381 137274321021824 utils.py:1231] [94000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.6672925213372
+I1204 17:32:16.434470 137274321021824 utils.py:1231] [94000] core_hours = 163.6672925213372
+I1204 17:32:16.434567 137274321021824 train.py:125] NOTE: Steps:94000/112603 [83.5%]
+Walltime:6d19h42m (0s eval)
+ETA:1d8h23m
+Total train time:8d4h3m
+I1204 17:37:28.584607 137274321021824 utils.py:1231] [94050] l2_params = 243.545059649363
+I1204 17:37:28.584832 137274321021824 utils.py:1231] [94050] train/loss = 2.017798289656639
+I1204 17:37:28.584942 137274321021824 utils.py:1231] [94050] l2_grads = 2.45885968208313
+I1204 17:37:28.585018 137274321021824 utils.py:1231] [94050] lr = 7.85385287410671e-05
+I1204 17:37:28.585080 137274321021824 utils.py:1231] [94050] uptime = 589637.947441263
+I1204 17:37:28.585145 137274321021824 utils.py:1231] [94050] examples_seen = 96307200.0
+I1204 17:37:28.585204 137274321021824 utils.py:1231] [94050] progress = 0.8352352956848397
+I1204 17:37:28.585261 137274321021824 utils.py:1231] [94050] epoch = 75.17146476610777
+I1204 17:37:28.585328 137274321021824 utils.py:1231] [94050] img/sec/core = 164.02318320887537
+I1204 17:37:28.585387 137274321021824 utils.py:1231] [94050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.7540011313275
+I1204 17:37:28.585445 137274321021824 utils.py:1231] [94050] core_hours = 163.7540011313275
+I1204 17:37:28.585513 137274321021824 train.py:125] NOTE: Steps:94050/112603 [83.5%]
+Walltime:6d19h47m (0s eval)
+ETA:1d8h18m
+Total train time:8d4h3m
+I1204 17:42:40.364229 137274321021824 utils.py:1231] [94100] l2_params = 243.50680673400043
+I1204 17:42:40.364479 137274321021824 utils.py:1231] [94100] train/loss = 1.55735245347023
+I1204 17:42:40.364615 137274321021824 utils.py:1231] [94100] l2_grads = 2.401942014694214
+I1204 17:42:40.364716 137274321021824 utils.py:1231] [94100] lr = 7.812717213003548e-05
+I1204 17:42:40.364795 137274321021824 utils.py:1231] [94100] uptime = 589949.727148204
+I1204 17:42:40.364862 137274321021824 utils.py:1231] [94100] examples_seen = 96358400.0
+I1204 17:42:40.364936 137274321021824 utils.py:1231] [94100] progress = 0.835679333587915
+I1204 17:42:40.364999 137274321021824 utils.py:1231] [94100] epoch = 75.21142833057674
+I1204 17:42:40.365063 137274321021824 utils.py:1231] [94100] img/sec/core = 164.21851345729553
+I1204 17:42:40.365133 137274321021824 utils.py:1231] [94100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.84060660547777
+I1204 17:42:40.365204 137274321021824 utils.py:1231] [94100] core_hours = 163.84060660547777
+I1204 17:42:40.365283 137274321021824 train.py:125] NOTE: Steps:94100/112603 [83.6%]
+Walltime:6d19h52m (0s eval)
+ETA:1d8h13m
+Total train time:8d4h3m
+I1204 17:47:52.160692 137274321021824 utils.py:1231] [94150] l2_params = 243.46631605062507
+I1204 17:47:52.160994 137274321021824 utils.py:1231] [94150] train/loss = 1.583529219031334
+I1204 17:47:52.161146 137274321021824 utils.py:1231] [94150] l2_grads = 2.4225552082061768
+I1204 17:47:52.161231 137274321021824 utils.py:1231] [94150] lr = 7.771680430226272e-05
+I1204 17:47:52.161292 137274321021824 utils.py:1231] [94150] uptime = 590261.523652902
+I1204 17:47:52.161351 137274321021824 utils.py:1231] [94150] examples_seen = 96409600.0
+I1204 17:47:52.161401 137274321021824 utils.py:1231] [94150] progress = 0.8361233714909905
+I1204 17:47:52.161450 137274321021824 utils.py:1231] [94150] epoch = 75.25139189504569
+I1204 17:47:52.161502 137274321021824 utils.py:1231] [94150] img/sec/core = 164.20966633218708
+I1204 17:47:52.161571 137274321021824 utils.py:1231] [94150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 163.92721674567164
+I1204 17:47:52.161623 137274321021824 utils.py:1231] [94150] core_hours = 163.92721674567164
+I1204 17:47:52.161685 137274321021824 train.py:125] NOTE: Steps:94150/112603 [83.6%]
+Walltime:6d19h57m (0s eval)
+ETA:1d8h7m
+Total train time:8d4h3m
+I1204 17:53:03.892485 137274321021824 utils.py:1231] [94200] l2_params = 243.4253656750548
+I1204 17:53:03.892721 137274321021824 utils.py:1231] [94200] train/loss = 3.2738084197044373
+I1204 17:53:03.892821 137274321021824 utils.py:1231] [94200] l2_grads = 2.474365711212158
+I1204 17:53:03.892894 137274321021824 utils.py:1231] [94200] lr = 7.73074262195669e-05
+I1204 17:53:03.892953 137274321021824 utils.py:1231] [94200] uptime = 590573.255311536
+I1204 17:53:03.893034 137274321021824 utils.py:1231] [94200] examples_seen = 96460800.0
+I1204 17:53:03.893094 137274321021824 utils.py:1231] [94200] progress = 0.8365674093940659
+I1204 17:53:03.893144 137274321021824 utils.py:1231] [94200] epoch = 75.29135545951465
+I1204 17:53:03.893199 137274321021824 utils.py:1231] [94200] img/sec/core = 164.24382503962525
+I1204 17:53:03.893256 137274321021824 utils.py:1231] [94200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.01380887307002
+I1204 17:53:03.893310 137274321021824 utils.py:1231] [94200] core_hours = 164.01380887307002
+I1204 17:53:03.893371 137274321021824 train.py:125] NOTE: Steps:94200/112603 [83.7%]
+Walltime:6d20h2m (0s eval)
+ETA:1d8h2m
+Total train time:8d4h3m
+I1204 17:58:15.681432 137274321021824 utils.py:1231] [94250] l2_params = 243.3840460150456
+I1204 17:58:15.681702 137274321021824 utils.py:1231] [94250] train/loss = 2.6381229162216187
+I1204 17:58:15.681864 137274321021824 utils.py:1231] [94250] l2_grads = 2.383042335510254
+I1204 17:58:15.681972 137274321021824 utils.py:1231] [94250] lr = 7.689903884144618e-05
+I1204 17:58:15.682044 137274321021824 utils.py:1231] [94250] uptime = 590885.044404031
+I1204 17:58:15.682113 137274321021824 utils.py:1231] [94250] examples_seen = 96512000.0
+I1204 17:58:15.682179 137274321021824 utils.py:1231] [94250] progress = 0.8370114472971413
+I1204 17:58:15.682245 137274321021824 utils.py:1231] [94250] epoch = 75.3313190239836
+I1204 17:58:15.682311 137274321021824 utils.py:1231] [94250] img/sec/core = 164.21357011013063
+I1204 17:58:15.682379 137274321021824 utils.py:1231] [94250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.1004169543186
+I1204 17:58:15.682450 137274321021824 utils.py:1231] [94250] core_hours = 164.1004169543186
+I1204 17:58:15.682521 137274321021824 train.py:125] NOTE: Steps:94250/112603 [83.7%]
+Walltime:6d20h8m (0s eval)
+ETA:1d7h57m
+Total train time:8d4h3m
+I1204 18:03:27.473701 137274321021824 utils.py:1231] [94300] l2_params = 243.34459909014416
+I1204 18:03:27.473926 137274321021824 utils.py:1231] [94300] train/loss = 2.569666802883148
+I1204 18:03:27.474067 137274321021824 utils.py:1231] [94300] l2_grads = 2.541794538497925
+I1204 18:03:27.474149 137274321021824 utils.py:1231] [94300] lr = 7.649164312507661e-05
+I1204 18:03:27.474219 137274321021824 utils.py:1231] [94300] uptime = 591196.836578955
+I1204 18:03:27.474285 137274321021824 utils.py:1231] [94300] examples_seen = 96563200.0
+I1204 18:03:27.474349 137274321021824 utils.py:1231] [94300] progress = 0.8374554852002167
+I1204 18:03:27.474412 137274321021824 utils.py:1231] [94300] epoch = 75.37128258845256
+I1204 18:03:27.474478 137274321021824 utils.py:1231] [94300] img/sec/core = 164.2119466676521
+I1204 18:03:27.474546 137274321021824 utils.py:1231] [94300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.18702589179748
+I1204 18:03:27.474608 137274321021824 utils.py:1231] [94300] core_hours = 164.18702589179748
+I1204 18:03:27.474684 137274321021824 train.py:125] NOTE: Steps:94300/112603 [83.7%]
+Walltime:6d20h13m (0s eval)
+ETA:1d7h52m
+Total train time:8d4h3m
+I1204 18:08:39.205098 137274321021824 utils.py:1231] [94350] l2_params = 243.305613133067
+I1204 18:08:39.205294 137274321021824 utils.py:1231] [94350] train/loss = 3.6402194797992706
+I1204 18:08:39.205399 137274321021824 utils.py:1231] [94350] l2_grads = 2.5911970138549805
+I1204 18:08:39.205462 137274321021824 utils.py:1231] [94350] lr = 7.608524002531023e-05
+I1204 18:08:39.205514 137274321021824 utils.py:1231] [94350] uptime = 591508.567876426
+I1204 18:08:39.205574 137274321021824 utils.py:1231] [94350] examples_seen = 96614400.0
+I1204 18:08:39.205625 137274321021824 utils.py:1231] [94350] progress = 0.8378995231032921
+I1204 18:08:39.205675 137274321021824 utils.py:1231] [94350] epoch = 75.41124615292152
+I1204 18:08:39.205728 137274321021824 utils.py:1231] [94350] img/sec/core = 164.24401532780806
+I1204 18:08:39.205786 137274321021824 utils.py:1231] [94350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.2736179188728
+I1204 18:08:39.205843 137274321021824 utils.py:1231] [94350] core_hours = 164.2736179188728
+I1204 18:08:39.205912 137274321021824 train.py:125] NOTE: Steps:94350/112603 [83.8%]
+Walltime:6d20h18m (0s eval)
+ETA:1d7h46m
+Total train time:8d4h3m
+I1204 18:13:51.008893 137274321021824 utils.py:1231] [94400] l2_params = 243.26918569859072
+I1204 18:13:51.009154 137274321021824 utils.py:1231] [94400] train/loss = 2.1688483357429504
+I1204 18:13:51.009372 137274321021824 utils.py:1231] [94400] l2_grads = 2.4789235591888428
+I1204 18:13:51.009519 137274321021824 utils.py:1231] [94400] lr = 7.56798304946722e-05
+I1204 18:13:51.009621 137274321021824 utils.py:1231] [94400] uptime = 591820.3719715719
+I1204 18:13:51.009705 137274321021824 utils.py:1231] [94400] examples_seen = 96665600.0
+I1204 18:13:51.009784 137274321021824 utils.py:1231] [94400] progress = 0.8383435610063675
+I1204 18:13:51.009868 137274321021824 utils.py:1231] [94400] epoch = 75.45120971739047
+I1204 18:13:51.009951 137274321021824 utils.py:1231] [94400] img/sec/core = 164.20566887052232
+I1204 18:13:51.010035 137274321021824 utils.py:1231] [94400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.36023016752443
+I1204 18:13:51.010101 137274321021824 utils.py:1231] [94400] core_hours = 164.36023016752443
+I1204 18:13:51.010180 137274321021824 train.py:125] NOTE: Steps:94400/112603 [83.8%]
+Walltime:6d20h23m (0s eval)
+ETA:1d7h41m
+Total train time:8d4h3m
+I1204 18:19:02.806830 137274321021824 utils.py:1231] [94450] l2_params = 243.2328386858005
+I1204 18:19:02.807115 137274321021824 utils.py:1231] [94450] train/loss = 1.6738500446081161
+I1204 18:19:02.807251 137274321021824 utils.py:1231] [94450] l2_grads = 2.6335794925689697
+I1204 18:19:02.807363 137274321021824 utils.py:1231] [94450] lr = 7.527541548335916e-05
+I1204 18:19:02.807436 137274321021824 utils.py:1231] [94450] uptime = 592132.169797208
+I1204 18:19:02.807491 137274321021824 utils.py:1231] [94450] examples_seen = 96716800.0
+I1204 18:19:02.807541 137274321021824 utils.py:1231] [94450] progress = 0.8387875989094429
+I1204 18:19:02.807591 137274321021824 utils.py:1231] [94450] epoch = 75.49117328185943
+I1204 18:19:02.807643 137274321021824 utils.py:1231] [94450] img/sec/core = 164.20897065444618
+I1204 18:19:02.807719 137274321021824 utils.py:1231] [94450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.44684067464556
+I1204 18:19:02.807779 137274321021824 utils.py:1231] [94450] core_hours = 164.44684067464556
+I1204 18:19:02.807842 137274321021824 train.py:125] NOTE: Steps:94450/112603 [83.9%]
+Walltime:6d20h28m (0s eval)
+ETA:1d7h36m
+Total train time:8d4h3m
+I1204 18:24:14.561053 137274321021824 utils.py:1231] [94500] l2_params = 243.19314944710828
+I1204 18:24:14.561273 137274321021824 utils.py:1231] [94500] train/loss = 1.6209094524383545
+I1204 18:24:14.561386 137274321021824 utils.py:1231] [94500] l2_grads = 2.5394246578216553
+I1204 18:24:14.561462 137274321021824 utils.py:1231] [94500] lr = 7.48719959392372e-05
+I1204 18:24:14.561543 137274321021824 utils.py:1231] [94500] uptime = 592443.923899659
+I1204 18:24:14.561615 137274321021824 utils.py:1231] [94500] examples_seen = 96768000.0
+I1204 18:24:14.561677 137274321021824 utils.py:1231] [94500] progress = 0.8392316368125183
+I1204 18:24:14.561732 137274321021824 utils.py:1231] [94500] epoch = 75.53113684632838
+I1204 18:24:14.561788 137274321021824 utils.py:1231] [94500] img/sec/core = 164.2320007899875
+I1204 18:24:14.561848 137274321021824 utils.py:1231] [94500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.53343903643747
+I1204 18:24:14.561915 137274321021824 utils.py:1231] [94500] core_hours = 164.53343903643747
+I1204 18:24:14.561983 137274321021824 train.py:125] NOTE: Steps:94500/112603 [83.9%]
+Walltime:6d20h34m (0s eval)
+ETA:1d7h31m
+Total train time:8d4h3m
+I1204 18:29:26.360988 137274321021824 utils.py:1231] [94550] l2_params = 243.155062812424
+I1204 18:29:26.361208 137274321021824 utils.py:1231] [94550] train/loss = 1.6640934497117996
+I1204 18:29:26.361329 137274321021824 utils.py:1231] [94550] l2_grads = 2.608393907546997
+I1204 18:29:26.361409 137274321021824 utils.py:1231] [94550] lr = 7.446957280783852e-05
+I1204 18:29:26.361473 137274321021824 utils.py:1231] [94550] uptime = 592755.723834285
+I1204 18:29:26.361538 137274321021824 utils.py:1231] [94550] examples_seen = 96819200.0
+I1204 18:29:26.361615 137274321021824 utils.py:1231] [94550] progress = 0.8396756747155937
+I1204 18:29:26.361676 137274321021824 utils.py:1231] [94550] epoch = 75.57110041079734
+I1204 18:29:26.361742 137274321021824 utils.py:1231] [94550] img/sec/core = 164.20785995801683
+I1204 18:29:26.361809 137274321021824 utils.py:1231] [94550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.62005012938917
+I1204 18:29:26.361870 137274321021824 utils.py:1231] [94550] core_hours = 164.62005012938917
+I1204 18:29:26.361952 137274321021824 train.py:125] NOTE: Steps:94550/112603 [84.0%]
+Walltime:6d20h39m (0s eval)
+ETA:1d7h25m
+Total train time:8d4h3m
+I1204 18:34:38.150621 137274321021824 utils.py:1231] [94600] l2_params = 243.11664834203768
+I1204 18:34:38.150846 137274321021824 utils.py:1231] [94600] train/loss = 1.8554718494415283
+I1204 18:34:38.150969 137274321021824 utils.py:1231] [94600] l2_grads = 2.4434432983398438
+I1204 18:34:38.151046 137274321021824 utils.py:1231] [94600] lr = 7.406814703236053e-05
+I1204 18:34:38.151107 137274321021824 utils.py:1231] [94600] uptime = 593067.513468694
+I1204 18:34:38.151172 137274321021824 utils.py:1231] [94600] examples_seen = 96870400.0
+I1204 18:34:38.151241 137274321021824 utils.py:1231] [94600] progress = 0.8401197126186691
+I1204 18:34:38.151304 137274321021824 utils.py:1231] [94600] epoch = 75.6110639752663
+I1204 18:34:38.151371 137274321021824 utils.py:1231] [94600] img/sec/core = 164.21328469455793
+I1204 18:34:38.151443 137274321021824 utils.py:1231] [94600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.70665836116942
+I1204 18:34:38.151508 137274321021824 utils.py:1231] [94600] core_hours = 164.70665836116942
+I1204 18:34:38.151574 137274321021824 train.py:125] NOTE: Steps:94600/112603 [84.0%]
+Walltime:6d20h44m (0s eval)
+ETA:1d7h20m
+Total train time:8d4h3m
+I1204 18:39:49.950195 137274321021824 utils.py:1231] [94650] l2_params = 243.07833633050757
+I1204 18:39:49.950435 137274321021824 utils.py:1231] [94650] train/loss = 3.746070623397827
+I1204 18:39:49.950535 137274321021824 utils.py:1231] [94650] l2_grads = 2.7134289741516113
+I1204 18:39:49.950596 137274321021824 utils.py:1231] [94650] lr = 7.366771955366269e-05
+I1204 18:39:49.950645 137274321021824 utils.py:1231] [94650] uptime = 593379.313008234
+I1204 18:39:49.950698 137274321021824 utils.py:1231] [94650] examples_seen = 96921600.0
+I1204 18:39:49.950748 137274321021824 utils.py:1231] [94650] progress = 0.8405637505217446
+I1204 18:39:49.950796 137274321021824 utils.py:1231] [94650] epoch = 75.65102753973525
+I1204 18:39:49.950845 137274321021824 utils.py:1231] [94650] img/sec/core = 164.20806802833215
+I1204 18:39:49.950904 137274321021824 utils.py:1231] [94650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.793269344375
+I1204 18:39:49.950953 137274321021824 utils.py:1231] [94650] core_hours = 164.793269344375
+I1204 18:39:49.951013 137274321021824 train.py:125] NOTE: Steps:94650/112603 [84.1%]
+Walltime:6d20h49m (0s eval)
+ETA:1d7h15m
+Total train time:8d4h3m
+I1204 18:45:01.745237 137274321021824 utils.py:1231] [94700] l2_params = 243.04281277684427
+I1204 18:45:01.745465 137274321021824 utils.py:1231] [94700] train/loss = 1.6422466039657593
+I1204 18:45:01.745567 137274321021824 utils.py:1231] [94700] l2_grads = 2.6101503372192383
+I1204 18:45:01.745634 137274321021824 utils.py:1231] [94700] lr = 7.326829131026457e-05
+I1204 18:45:01.745687 137274321021824 utils.py:1231] [94700] uptime = 593691.108049318
+I1204 18:45:01.745742 137274321021824 utils.py:1231] [94700] examples_seen = 96972800.0
+I1204 18:45:01.745792 137274321021824 utils.py:1231] [94700] progress = 0.8410077884248199
+I1204 18:45:01.745841 137274321021824 utils.py:1231] [94700] epoch = 75.69099110420422
+I1204 18:45:01.745901 137274321021824 utils.py:1231] [94700] img/sec/core = 164.21043715771003
+I1204 18:45:01.745960 137274321021824 utils.py:1231] [94700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.87987907800942
+I1204 18:45:01.746013 137274321021824 utils.py:1231] [94700] core_hours = 164.87987907800942
+I1204 18:45:01.746074 137274321021824 train.py:125] NOTE: Steps:94700/112603 [84.1%]
+Walltime:6d20h54m (0s eval)
+ETA:1d7h10m
+Total train time:8d4h3m
+I1204 18:50:13.522356 137274321021824 utils.py:1231] [94750] l2_params = 243.00460069474946
+I1204 18:50:13.522616 137274321021824 utils.py:1231] [94750] train/loss = 1.9683443307876587
+I1204 18:50:13.522742 137274321021824 utils.py:1231] [94750] l2_grads = 2.5653090476989746
+I1204 18:50:13.522840 137274321021824 utils.py:1231] [94750] lr = 7.286986323834431e-05
+I1204 18:50:13.522920 137274321021824 utils.py:1231] [94750] uptime = 594002.885281342
+I1204 18:50:13.522982 137274321021824 utils.py:1231] [94750] examples_seen = 97024000.0
+I1204 18:50:13.523040 137274321021824 utils.py:1231] [94750] progress = 0.8414518263278954
+I1204 18:50:13.523097 137274321021824 utils.py:1231] [94750] epoch = 75.73095466867316
+I1204 18:50:13.523157 137274321021824 utils.py:1231] [94750] img/sec/core = 164.2198170392959
+I1204 18:50:13.523220 137274321021824 utils.py:1231] [94750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 164.96648386468277
+I1204 18:50:13.523279 137274321021824 utils.py:1231] [94750] core_hours = 164.96648386468277
+I1204 18:50:13.523343 137274321021824 train.py:125] NOTE: Steps:94750/112603 [84.1%]
+Walltime:6d21h0m (0s eval)
+ETA:1d7h5m
+Total train time:8d4h3m
+I1204 18:55:25.545513 137274321021824 utils.py:1231] [94800] l2_params = 242.96636892458326
+I1204 18:55:25.545716 137274321021824 utils.py:1231] [94800] train/loss = 2.0312767326831818
+I1204 18:55:25.545816 137274321021824 utils.py:1231] [94800] l2_grads = 2.491945743560791
+I1204 18:55:25.545890 137274321021824 utils.py:1231] [94800] lr = 7.247243627173512e-05
+I1204 18:55:25.545949 137274321021824 utils.py:1231] [94800] uptime = 594314.908310705
+I1204 18:55:25.546010 137274321021824 utils.py:1231] [94800] examples_seen = 97075200.0
+I1204 18:55:25.546065 137274321021824 utils.py:1231] [94800] progress = 0.8418958642309707
+I1204 18:55:25.546123 137274321021824 utils.py:1231] [94800] epoch = 75.77091823314213
+I1204 18:55:25.546174 137274321021824 utils.py:1231] [94800] img/sec/core = 164.09045224810183
+I1204 18:55:25.546234 137274321021824 utils.py:1231] [94800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.0531569283947
+I1204 18:55:25.546291 137274321021824 utils.py:1231] [94800] core_hours = 165.0531569283947
+I1204 18:55:25.546354 137274321021824 train.py:125] NOTE: Steps:94800/112603 [84.2%]
+Walltime:6d21h5m (0s eval)
+ETA:1d6h59m
+Total train time:8d4h3m
+I1204 19:00:37.347279 137274321021824 utils.py:1231] [94850] l2_params = 242.93126152683047
+I1204 19:00:37.347493 137274321021824 utils.py:1231] [94850] train/loss = 1.8242932856082916
+I1204 19:00:37.347596 137274321021824 utils.py:1231] [94850] l2_grads = 2.4571034908294678
+I1204 19:00:37.347676 137274321021824 utils.py:1231] [94850] lr = 7.207601134192462e-05
+I1204 19:00:37.347738 137274321021824 utils.py:1231] [94850] uptime = 594626.710098488
+I1204 19:00:37.347800 137274321021824 utils.py:1231] [94850] examples_seen = 97126400.0
+I1204 19:00:37.347859 137274321021824 utils.py:1231] [94850] progress = 0.8423399021340462
+I1204 19:00:37.347923 137274321021824 utils.py:1231] [94850] epoch = 75.81088179761109
+I1204 19:00:37.347985 137274321021824 utils.py:1231] [94850] img/sec/core = 164.20688400807344
+I1204 19:00:37.348049 137274321021824 utils.py:1231] [94850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.13976853611223
+I1204 19:00:37.348103 137274321021824 utils.py:1231] [94850] core_hours = 165.13976853611223
+I1204 19:00:37.348191 137274321021824 train.py:125] NOTE: Steps:94850/112603 [84.2%]
+Walltime:6d21h10m (0s eval)
+ETA:1d6h54m
+Total train time:8d4h3m
+I1204 19:05:49.138018 137274321021824 utils.py:1231] [94900] l2_params = 242.89327135972624
+I1204 19:05:49.138285 137274321021824 utils.py:1231] [94900] train/loss = 2.4219528138637543
+I1204 19:05:49.138407 137274321021824 utils.py:1231] [94900] l2_grads = 2.3515303134918213
+I1204 19:05:49.138498 137274321021824 utils.py:1231] [94900] lr = 7.168058937805106e-05
+I1204 19:05:49.138561 137274321021824 utils.py:1231] [94900] uptime = 594938.500923159
+I1204 19:05:49.138620 137274321021824 utils.py:1231] [94900] examples_seen = 97177600.0
+I1204 19:05:49.138670 137274321021824 utils.py:1231] [94900] progress = 0.8427839400371215
+I1204 19:05:49.138719 137274321021824 utils.py:1231] [94900] epoch = 75.85084536208004
+I1204 19:05:49.138772 137274321021824 utils.py:1231] [94900] img/sec/core = 164.21265781004942
+I1204 19:05:49.138835 137274321021824 utils.py:1231] [94900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.22637709852083
+I1204 19:05:49.138896 137274321021824 utils.py:1231] [94900] core_hours = 165.22637709852083
+I1204 19:05:49.138960 137274321021824 train.py:125] NOTE: Steps:94900/112603 [84.3%]
+Walltime:6d21h15m (0s eval)
+ETA:1d6h49m
+Total train time:8d4h3m
+I1204 19:11:00.942259 137274321021824 utils.py:1231] [94950] l2_params = 242.8567016877547
+I1204 19:11:00.942546 137274321021824 utils.py:1231] [94950] train/loss = 1.5862731784582138
+I1204 19:11:00.942701 137274321021824 utils.py:1231] [94950] l2_grads = 2.4920005798339844
+I1204 19:11:00.942778 137274321021824 utils.py:1231] [94950] lr = 7.12861713069022e-05
+I1204 19:11:00.942843 137274321021824 utils.py:1231] [94950] uptime = 595250.305205065
+I1204 19:11:00.942903 137274321021824 utils.py:1231] [94950] examples_seen = 97228800.0
+I1204 19:11:00.942965 137274321021824 utils.py:1231] [94950] progress = 0.843227977940197
+I1204 19:11:00.943019 137274321021824 utils.py:1231] [94950] epoch = 75.890808926549
+I1204 19:11:00.943072 137274321021824 utils.py:1231] [94950] img/sec/core = 164.20557051694576
+I1204 19:11:00.943128 137274321021824 utils.py:1231] [94950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.31298939905025
+I1204 19:11:00.943188 137274321021824 utils.py:1231] [94950] core_hours = 165.31298939905025
+I1204 19:11:00.943261 137274321021824 train.py:125] NOTE: Steps:94950/112603 [84.3%]
+Walltime:6d21h20m (0s eval)
+ETA:1d6h44m
+Total train time:8d4h3m
+I1204 19:16:12.743689 137274321021824 utils.py:1231] [95000] l2_params = 242.81711511141333
+I1204 19:16:12.743899 137274321021824 utils.py:1231] [95000] train/loss = 2.3450523912906647
+I1204 19:16:12.744017 137274321021824 utils.py:1231] [95000] l2_grads = 2.392606258392334
+I1204 19:16:12.744092 137274321021824 utils.py:1231] [95000] lr = 7.089275805291342e-05
+I1204 19:16:12.744150 137274321021824 utils.py:1231] [95000] uptime = 595562.106511611
+I1204 19:16:12.744209 137274321021824 utils.py:1231] [95000] examples_seen = 97280000.0
+I1204 19:16:12.744264 137274321021824 utils.py:1231] [95000] progress = 0.8436720158432723
+I1204 19:16:12.744318 137274321021824 utils.py:1231] [95000] epoch = 75.93077249101796
+I1204 19:16:12.744374 137274321021824 utils.py:1231] [95000] img/sec/core = 164.2071374464646
+I1204 19:16:12.744436 137274321021824 utils.py:1231] [95000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.39960087309086
+I1204 19:16:12.744491 137274321021824 utils.py:1231] [95000] core_hours = 165.39960087309086
+I1204 19:16:12.744554 137274321021824 train.py:125] NOTE: Steps:95000/112603 [84.4%]
+Walltime:6d21h26m (0s eval)
+ETA:1d6h38m
+Total train time:8d4h3m
+I1204 19:16:13.100414 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:95000/112603 [84.4%]
+Walltime:6d21h26m (0s eval)
+ETA:1d6h38m
+Total train time:8d4h3m
+I1204 19:17:51.431739 137274321021824 utils.py:1231] [95000] val/acc@1 = 0.7509566326530612
+I1204 19:17:51.432051 137274321021824 utils.py:1231] [95000] val/loss = 0.9899900467420111
+I1204 19:17:51.432227 137274321021824 utils.py:1231] [95000] z/secs/eval/val = 98.33155454101507
+I1204 19:17:51.432304 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.33155454101507
+I1204 19:23:03.223859 137274321021824 utils.py:1231] [95050] l2_params = 242.78073756860752
+I1204 19:23:03.224102 137274321021824 utils.py:1231] [95050] train/loss = 1.582329973578453
+I1204 19:23:03.224248 137274321021824 utils.py:1231] [95050] l2_grads = 2.43939208984375
+I1204 19:23:03.224360 137274321021824 utils.py:1231] [95050] lr = 7.050035053816422e-05
+I1204 19:23:03.224437 137274321021824 utils.py:1231] [95050] uptime = 595972.586795128
+I1204 19:23:03.224510 137274321021824 utils.py:1231] [95050] examples_seen = 97331200.0
+I1204 19:23:03.224578 137274321021824 utils.py:1231] [95050] progress = 0.8441160537463478
+I1204 19:23:03.224641 137274321021824 utils.py:1231] [95050] epoch = 75.97073605548691
+I1204 19:23:03.224703 137274321021824 utils.py:1231] [95050] img/sec/core = 124.73193489667071
+I1204 19:23:03.224765 137274321021824 utils.py:1231] [95050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.51362317406776
+I1204 19:23:03.224832 137274321021824 utils.py:1231] [95050] core_hours = 165.51362317406776
+I1204 19:23:03.224914 137274321021824 train.py:125] NOTE: Steps:95050/112603 [84.4%]
+Walltime:6d21h32m (0s eval)
+ETA:1d6h33m
+Total train time:8d4h5m
+I1204 19:28:14.963418 137274321021824 utils.py:1231] [95100] l2_params = 242.7415230150696
+I1204 19:28:14.963689 137274321021824 utils.py:1231] [95100] train/loss = 1.6123383790254593
+I1204 19:28:14.963820 137274321021824 utils.py:1231] [95100] l2_grads = 2.5527682304382324
+I1204 19:28:14.963931 137274321021824 utils.py:1231] [95100] lr = 7.010894968237751e-05
+I1204 19:28:14.964021 137274321021824 utils.py:1231] [95100] uptime = 596284.326380218
+I1204 19:28:14.964087 137274321021824 utils.py:1231] [95100] examples_seen = 97382400.0
+I1204 19:28:14.964145 137274321021824 utils.py:1231] [95100] progress = 0.8445600916494233
+I1204 19:28:14.964201 137274321021824 utils.py:1231] [95100] epoch = 76.01069961995587
+I1204 19:28:14.964260 137274321021824 utils.py:1231] [95100] img/sec/core = 164.23964888900906
+I1204 19:28:14.964322 137274321021824 utils.py:1231] [95100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.60021750325944
+I1204 19:28:14.964382 137274321021824 utils.py:1231] [95100] core_hours = 165.60021750325944
+I1204 19:28:14.964444 137274321021824 train.py:125] NOTE: Steps:95100/112603 [84.5%]
+Walltime:6d21h38m (0s eval)
+ETA:1d6h28m
+Total train time:8d4h4m
+I1204 19:33:26.770723 137274321021824 utils.py:1231] [95150] l2_params = 242.70547491127206
+I1204 19:33:26.770941 137274321021824 utils.py:1231] [95150] train/loss = 1.5253791213035583
+I1204 19:33:26.771056 137274321021824 utils.py:1231] [95150] l2_grads = 2.600799083709717
+I1204 19:33:26.771132 137274321021824 utils.py:1231] [95150] lr = 6.97185564029163e-05
+I1204 19:33:26.771198 137274321021824 utils.py:1231] [95150] uptime = 596596.13355826
+I1204 19:33:26.771260 137274321021824 utils.py:1231] [95150] examples_seen = 97433600.0
+I1204 19:33:26.771323 137274321021824 utils.py:1231] [95150] progress = 0.8450041295524986
+I1204 19:33:26.771382 137274321021824 utils.py:1231] [95150] epoch = 76.05066318442482
+I1204 19:33:26.771442 137274321021824 utils.py:1231] [95150] img/sec/core = 164.2040453382414
+I1204 19:33:26.771502 137274321021824 utils.py:1231] [95150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.68683060827112
+I1204 19:33:26.771564 137274321021824 utils.py:1231] [95150] core_hours = 165.68683060827112
+I1204 19:33:26.771629 137274321021824 train.py:125] NOTE: Steps:95150/112603 [84.5%]
+Walltime:6d21h43m (0s eval)
+ETA:1d6h23m
+Total train time:8d4h4m
+I1204 19:38:38.568722 137274321021824 utils.py:1231] [95200] l2_params = 242.67023839270354
+I1204 19:38:38.569024 137274321021824 utils.py:1231] [95200] train/loss = 3.724862813949585
+I1204 19:38:38.569210 137274321021824 utils.py:1231] [95200] l2_grads = 2.5933837890625
+I1204 19:38:38.569303 137274321021824 utils.py:1231] [95200] lr = 6.932917161478221e-05
+I1204 19:38:38.569375 137274321021824 utils.py:1231] [95200] uptime = 596907.931735861
+I1204 19:38:38.569456 137274321021824 utils.py:1231] [95200] examples_seen = 97484800.0
+I1204 19:38:38.569522 137274321021824 utils.py:1231] [95200] progress = 0.845448167455574
+I1204 19:38:38.569597 137274321021824 utils.py:1231] [95200] epoch = 76.09062674889378
+I1204 19:38:38.569665 137274321021824 utils.py:1231] [95200] img/sec/core = 164.20878529164065
+I1204 19:38:38.569731 137274321021824 utils.py:1231] [95200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.77344121316028
+I1204 19:38:38.569790 137274321021824 utils.py:1231] [95200] core_hours = 165.77344121316028
+I1204 19:38:38.569860 137274321021824 train.py:125] NOTE: Steps:95200/112603 [84.5%]
+Walltime:6d21h48m (0s eval)
+ETA:1d6h18m
+Total train time:8d4h4m
+I1204 19:43:50.354933 137274321021824 utils.py:1231] [95250] l2_params = 242.6330038836929
+I1204 19:43:50.355151 137274321021824 utils.py:1231] [95250] train/loss = 1.708649754524231
+I1204 19:43:50.355288 137274321021824 utils.py:1231] [95250] l2_grads = 2.8358652591705322
+I1204 19:43:50.355379 137274321021824 utils.py:1231] [95250] lr = 6.894079623061334e-05
+I1204 19:43:50.355453 137274321021824 utils.py:1231] [95250] uptime = 597219.717810293
+I1204 19:43:50.355533 137274321021824 utils.py:1231] [95250] examples_seen = 97536000.0
+I1204 19:43:50.355604 137274321021824 utils.py:1231] [95250] progress = 0.8458922053586494
+I1204 19:43:50.355670 137274321021824 utils.py:1231] [95250] epoch = 76.13059031336275
+I1204 19:43:50.355745 137274321021824 utils.py:1231] [95250] img/sec/core = 164.21515968370426
+I1204 19:43:50.355824 137274321021824 utils.py:1231] [95250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.86004845605802
+I1204 19:43:50.355894 137274321021824 utils.py:1231] [95250] core_hours = 165.86004845605802
+I1204 19:43:50.355973 137274321021824 train.py:125] NOTE: Steps:95250/112603 [84.6%]
+Walltime:6d21h53m (0s eval)
+ETA:1d6h13m
+Total train time:8d4h4m
+I1204 19:49:02.147265 137274321021824 utils.py:1231] [95300] l2_params = 242.59810904618854
+I1204 19:49:02.147534 137274321021824 utils.py:1231] [95300] train/loss = 1.5376231670379639
+I1204 19:49:02.147643 137274321021824 utils.py:1231] [95300] l2_grads = 2.4691967964172363
+I1204 19:49:02.147737 137274321021824 utils.py:1231] [95300] lr = 6.855343116068173e-05
+I1204 19:49:02.147804 137274321021824 utils.py:1231] [95300] uptime = 597531.510164906
+I1204 19:49:02.147865 137274321021824 utils.py:1231] [95300] examples_seen = 97587200.0
+I1204 19:49:02.147930 137274321021824 utils.py:1231] [95300] progress = 0.8463362432617249
+I1204 19:49:02.147983 137274321021824 utils.py:1231] [95300] epoch = 76.1705538778317
+I1204 19:49:02.148047 137274321021824 utils.py:1231] [95300] img/sec/core = 164.2118520306162
+I1204 19:49:02.148113 137274321021824 utils.py:1231] [95300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 165.94665744345056
+I1204 19:49:02.148174 137274321021824 utils.py:1231] [95300] core_hours = 165.94665744345056
+I1204 19:49:02.148245 137274321021824 train.py:125] NOTE: Steps:95300/112603 [84.6%]
+Walltime:6d21h58m (0s eval)
+ETA:1d6h7m
+Total train time:8d4h4m
+I1204 19:54:13.932281 137274321021824 utils.py:1231] [95350] l2_params = 242.55969905649044
+I1204 19:54:13.932541 137274321021824 utils.py:1231] [95350] train/loss = 2.2248603254556656
+I1204 19:54:13.932684 137274321021824 utils.py:1231] [95350] l2_grads = 2.559271812438965
+I1204 19:54:13.932790 137274321021824 utils.py:1231] [95350] lr = 6.816707731289177e-05
+I1204 19:54:13.932868 137274321021824 utils.py:1231] [95350] uptime = 597843.295225625
+I1204 19:54:13.932975 137274321021824 utils.py:1231] [95350] examples_seen = 97638400.0
+I1204 19:54:13.933033 137274321021824 utils.py:1231] [95350] progress = 0.8467802811648002
+I1204 19:54:13.933093 137274321021824 utils.py:1231] [95350] epoch = 76.21051744230066
+I1204 19:54:13.933148 137274321021824 utils.py:1231] [95350] img/sec/core = 164.21569359971298
+I1204 19:54:13.933227 137274321021824 utils.py:1231] [95350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.03326440476138
+I1204 19:54:13.933289 137274321021824 utils.py:1231] [95350] core_hours = 166.03326440476138
+I1204 19:54:13.933359 137274321021824 train.py:125] NOTE: Steps:95350/112603 [84.7%]
+Walltime:6d22h4m (0s eval)
+ETA:1d6h2m
+Total train time:8d4h4m
+I1204 19:59:25.713911 137274321021824 utils.py:1231] [95400] l2_params = 242.52211499370466
+I1204 19:59:25.714113 137274321021824 utils.py:1231] [95400] train/loss = 1.6176043301820755
+I1204 19:59:25.714220 137274321021824 utils.py:1231] [95400] l2_grads = 2.586024761199951
+I1204 19:59:25.714296 137274321021824 utils.py:1231] [95400] lr = 6.778173559277724e-05
+I1204 19:59:25.714365 137274321021824 utils.py:1231] [95400] uptime = 598155.076720806
+I1204 19:59:25.714425 137274321021824 utils.py:1231] [95400] examples_seen = 97689600.0
+I1204 19:59:25.714485 137274321021824 utils.py:1231] [95400] progress = 0.8472243190678757
+I1204 19:59:25.714542 137274321021824 utils.py:1231] [95400] epoch = 76.2504810067696
+I1204 19:59:25.714600 137274321021824 utils.py:1231] [95400] img/sec/core = 164.217571572973
+I1204 19:59:25.714668 137274321021824 utils.py:1231] [95400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.11987037564498
+I1204 19:59:25.714725 137274321021824 utils.py:1231] [95400] core_hours = 166.11987037564498
+I1204 19:59:25.714795 137274321021824 train.py:125] NOTE: Steps:95400/112603 [84.7%]
+Walltime:6d22h9m (0s eval)
+ETA:1d5h57m
+Total train time:8d4h4m
+I1204 20:04:37.501544 137274321021824 utils.py:1231] [95450] l2_params = 242.4892239017287
+I1204 20:04:37.501790 137274321021824 utils.py:1231] [95450] train/loss = 2.172017514705658
+I1204 20:04:37.501926 137274321021824 utils.py:1231] [95450] l2_grads = 2.456605911254883
+I1204 20:04:37.502019 137274321021824 utils.py:1231] [95450] lr = 6.739740690350006e-05
+I1204 20:04:37.502083 137274321021824 utils.py:1231] [95450] uptime = 598466.864444704
+I1204 20:04:37.502162 137274321021824 utils.py:1231] [95450] examples_seen = 97740800.0
+I1204 20:04:37.502216 137274321021824 utils.py:1231] [95450] progress = 0.847668356970951
+I1204 20:04:37.502284 137274321021824 utils.py:1231] [95450] epoch = 76.29044457123857
+I1204 20:04:37.502345 137274321021824 utils.py:1231] [95450] img/sec/core = 164.2142909280736
+I1204 20:04:37.502417 137274321021824 utils.py:1231] [95450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.2064780767278
+I1204 20:04:37.502471 137274321021824 utils.py:1231] [95450] core_hours = 166.2064780767278
+I1204 20:04:37.502536 137274321021824 train.py:125] NOTE: Steps:95450/112603 [84.8%]
+Walltime:6d22h14m (0s eval)
+ETA:1d5h52m
+Total train time:8d4h4m
+I1204 20:09:49.287414 137274321021824 utils.py:1231] [95500] l2_params = 242.45397513748753
+I1204 20:09:49.287606 137274321021824 utils.py:1231] [95500] train/loss = 3.251505672931671
+I1204 20:09:49.287708 137274321021824 utils.py:1231] [95500] l2_grads = 2.489492177963257
+I1204 20:09:49.287770 137274321021824 utils.py:1231] [95500] lr = 6.701409214584782e-05
+I1204 20:09:49.287822 137274321021824 utils.py:1231] [95500] uptime = 598778.650184627
+I1204 20:09:49.287874 137274321021824 utils.py:1231] [95500] examples_seen = 97792000.0
+I1204 20:09:49.287929 137274321021824 utils.py:1231] [95500] progress = 0.8481123948740265
+I1204 20:09:49.287977 137274321021824 utils.py:1231] [95500] epoch = 76.33040813570753
+I1204 20:09:49.288027 137274321021824 utils.py:1231] [95500] img/sec/core = 164.21533586700448
+I1204 20:09:49.288084 137274321021824 utils.py:1231] [95500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.29308522670638
+I1204 20:09:49.288135 137274321021824 utils.py:1231] [95500] core_hours = 166.29308522670638
+I1204 20:09:49.288195 137274321021824 train.py:125] NOTE: Steps:95500/112603 [84.8%]
+Walltime:6d22h19m (0s eval)
+ETA:1d5h46m
+Total train time:8d4h4m
+I1204 20:15:01.081869 137274321021824 utils.py:1231] [95550] l2_params = 242.41594029955448
+I1204 20:15:01.082142 137274321021824 utils.py:1231] [95550] train/loss = 2.625431716442108
+I1204 20:15:01.082262 137274321021824 utils.py:1231] [95550] l2_grads = 2.670912742614746
+I1204 20:15:01.082353 137274321021824 utils.py:1231] [95550] lr = 6.663179221823136e-05
+I1204 20:15:01.082421 137274321021824 utils.py:1231] [95550] uptime = 599090.444783218
+I1204 20:15:01.082489 137274321021824 utils.py:1231] [95550] examples_seen = 97843200.0
+I1204 20:15:01.082548 137274321021824 utils.py:1231] [95550] progress = 0.8485564327771018
+I1204 20:15:01.082599 137274321021824 utils.py:1231] [95550] epoch = 76.37037170017648
+I1204 20:15:01.082650 137274321021824 utils.py:1231] [95550] img/sec/core = 164.2106702020626
+I1204 20:15:01.082706 137274321021824 utils.py:1231] [95550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.3796948374261
+I1204 20:15:01.082761 137274321021824 utils.py:1231] [95550] core_hours = 166.3796948374261
+I1204 20:15:01.082834 137274321021824 train.py:125] NOTE: Steps:95550/112603 [84.9%]
+Walltime:6d22h24m (0s eval)
+ETA:1d5h41m
+Total train time:8d4h4m
+I1204 20:20:12.876782 137274321021824 utils.py:1231] [95600] l2_params = 242.38140590680712
+I1204 20:20:12.877059 137274321021824 utils.py:1231] [95600] train/loss = 1.7535921186208725
+I1204 20:20:12.877249 137274321021824 utils.py:1231] [95600] l2_grads = 2.5432190895080566
+I1204 20:20:12.877369 137274321021824 utils.py:1231] [95600] lr = 6.625050801668354e-05
+I1204 20:20:12.877459 137274321021824 utils.py:1231] [95600] uptime = 599402.239814297
+I1204 20:20:12.877544 137274321021824 utils.py:1231] [95600] examples_seen = 97894400.0
+I1204 20:20:12.877657 137274321021824 utils.py:1231] [95600] progress = 0.8490004706801773
+I1204 20:20:12.877735 137274321021824 utils.py:1231] [95600] epoch = 76.41033526464544
+I1204 20:20:12.877809 137274321021824 utils.py:1231] [95600] img/sec/core = 164.21044242694018
+I1204 20:20:12.877907 137274321021824 utils.py:1231] [95600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.46630456828137
+I1204 20:20:12.877966 137274321021824 utils.py:1231] [95600] core_hours = 166.46630456828137
+I1204 20:20:12.878033 137274321021824 train.py:125] NOTE: Steps:95600/112603 [84.9%]
+Walltime:6d22h30m (0s eval)
+ETA:1d5h36m
+Total train time:8d4h4m
+I1204 20:25:24.684361 137274321021824 utils.py:1231] [95650] l2_params = 242.34587631215095
+I1204 20:25:24.684648 137274321021824 utils.py:1231] [95650] train/loss = 2.6209208369255066
+I1204 20:25:24.684761 137274321021824 utils.py:1231] [95650] l2_grads = 2.4790663719177246
+I1204 20:25:24.684844 137274321021824 utils.py:1231] [95650] lr = 6.587024043485599e-05
+I1204 20:25:24.684917 137274321021824 utils.py:1231] [95650] uptime = 599714.047278064
+I1204 20:25:24.684971 137274321021824 utils.py:1231] [95650] examples_seen = 97945600.0
+I1204 20:25:24.685023 137274321021824 utils.py:1231] [95650] progress = 0.8494445085832527
+I1204 20:25:24.685072 137274321021824 utils.py:1231] [95650] epoch = 76.45029882911439
+I1204 20:25:24.685125 137274321021824 utils.py:1231] [95650] img/sec/core = 164.2038948697468
+I1204 20:25:24.685182 137274321021824 utils.py:1231] [95650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.55291775266107
+I1204 20:25:24.685235 137274321021824 utils.py:1231] [95650] core_hours = 166.55291775266107
+I1204 20:25:24.685299 137274321021824 train.py:125] NOTE: Steps:95650/112603 [84.9%]
+Walltime:6d22h35m (0s eval)
+ETA:1d5h31m
+Total train time:8d4h4m
+I1204 20:30:36.493147 137274321021824 utils.py:1231] [95700] l2_params = 242.31184516812846
+I1204 20:30:36.493413 137274321021824 utils.py:1231] [95700] train/loss = 1.6681329309940338
+I1204 20:30:36.493550 137274321021824 utils.py:1231] [95700] l2_grads = 2.503349542617798
+I1204 20:30:36.493653 137274321021824 utils.py:1231] [95700] lr = 6.549099036401783e-05
+I1204 20:30:36.493729 137274321021824 utils.py:1231] [95700] uptime = 600025.8560815629
+I1204 20:30:36.493812 137274321021824 utils.py:1231] [95700] examples_seen = 97996800.0
+I1204 20:30:36.493879 137274321021824 utils.py:1231] [95700] progress = 0.8498885464863281
+I1204 20:30:36.493946 137274321021824 utils.py:1231] [95700] epoch = 76.49026239358335
+I1204 20:30:36.494005 137274321021824 utils.py:1231] [95700] img/sec/core = 164.20318934377954
+I1204 20:30:36.494072 137274321021824 utils.py:1231] [95700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.63953130918858
+I1204 20:30:36.494130 137274321021824 utils.py:1231] [95700] core_hours = 166.63953130918858
+I1204 20:30:36.494211 137274321021824 train.py:125] NOTE: Steps:95700/112603 [85.0%]
+Walltime:6d22h40m (0s eval)
+ETA:1d5h25m
+Total train time:8d4h4m
+I1204 20:35:48.286723 137274321021824 utils.py:1231] [95750] l2_params = 242.2770944140796
+I1204 20:35:48.286972 137274321021824 utils.py:1231] [95750] train/loss = 1.5453812628984451
+I1204 20:35:48.287081 137274321021824 utils.py:1231] [95750] l2_grads = 2.6899924278259277
+I1204 20:35:48.287157 137274321021824 utils.py:1231] [95750] lr = 6.511275869305346e-05
+I1204 20:35:48.287221 137274321021824 utils.py:1231] [95750] uptime = 600337.649581424
+I1204 20:35:48.287303 137274321021824 utils.py:1231] [95750] examples_seen = 98048000.0
+I1204 20:35:48.287371 137274321021824 utils.py:1231] [95750] progress = 0.8503325843894035
+I1204 20:35:48.287440 137274321021824 utils.py:1231] [95750] epoch = 76.53022595805231
+I1204 20:35:48.287505 137274321021824 utils.py:1231] [95750] img/sec/core = 164.2112488644587
+I1204 20:35:48.287580 137274321021824 utils.py:1231] [95750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.72614061470554
+I1204 20:35:48.287652 137274321021824 utils.py:1231] [95750] core_hours = 166.72614061470554
+I1204 20:35:48.287730 137274321021824 train.py:125] NOTE: Steps:95750/112603 [85.0%]
+Walltime:6d22h45m (0s eval)
+ETA:1d5h20m
+Total train time:8d4h4m
+I1204 20:41:00.084508 137274321021824 utils.py:1231] [95800] l2_params = 242.2419645974183
+I1204 20:41:00.084722 137274321021824 utils.py:1231] [95800] train/loss = 1.9646898061037064
+I1204 20:41:00.084821 137274321021824 utils.py:1231] [95800] l2_grads = 2.5712924003601074
+I1204 20:41:00.084893 137274321021824 utils.py:1231] [95800] lr = 6.473554630845992e-05
+I1204 20:41:00.084948 137274321021824 utils.py:1231] [95800] uptime = 600649.4473096849
+I1204 20:41:00.085002 137274321021824 utils.py:1231] [95800] examples_seen = 98099200.0
+I1204 20:41:00.085054 137274321021824 utils.py:1231] [95800] progress = 0.8507766222924789
+I1204 20:41:00.085104 137274321021824 utils.py:1231] [95800] epoch = 76.57018952252126
+I1204 20:41:00.085157 137274321021824 utils.py:1231] [95800] img/sec/core = 164.20902193727622
+I1204 20:41:00.085216 137274321021824 utils.py:1231] [95800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.81275109477804
+I1204 20:41:00.085267 137274321021824 utils.py:1231] [95800] core_hours = 166.81275109477804
+I1204 20:41:00.085331 137274321021824 train.py:125] NOTE: Steps:95800/112603 [85.1%]
+Walltime:6d22h50m (0s eval)
+ETA:1d5h15m
+Total train time:8d4h4m
+I1204 20:46:11.820898 137274321021824 utils.py:1231] [95850] l2_params = 242.20775086147032
+I1204 20:46:11.821132 137274321021824 utils.py:1231] [95850] train/loss = 1.614662989974022
+I1204 20:46:11.821232 137274321021824 utils.py:1231] [95850] l2_grads = 2.72377347946167
+I1204 20:46:11.821305 137274321021824 utils.py:1231] [95850] lr = 6.435935409434594e-05
+I1204 20:46:11.821365 137274321021824 utils.py:1231] [95850] uptime = 600961.183726042
+I1204 20:46:11.821426 137274321021824 utils.py:1231] [95850] examples_seen = 98150400.0
+I1204 20:46:11.821482 137274321021824 utils.py:1231] [95850] progress = 0.8512206601955543
+I1204 20:46:11.821541 137274321021824 utils.py:1231] [95850] epoch = 76.61015308699022
+I1204 20:46:11.821598 137274321021824 utils.py:1231] [95850] img/sec/core = 164.24131834937376
+I1204 20:46:11.821659 137274321021824 utils.py:1231] [95850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.89934454376615
+I1204 20:46:11.821715 137274321021824 utils.py:1231] [95850] core_hours = 166.89934454376615
+I1204 20:46:11.821784 137274321021824 train.py:125] NOTE: Steps:95850/112603 [85.1%]
+Walltime:6d22h56m (0s eval)
+ETA:1d5h10m
+Total train time:8d4h4m
+I1204 20:51:23.619884 137274321021824 utils.py:1231] [95900] l2_params = 242.17406777067794
+I1204 20:51:23.620150 137274321021824 utils.py:1231] [95900] train/loss = 1.6551679521799088
+I1204 20:51:23.620268 137274321021824 utils.py:1231] [95900] l2_grads = 2.780578374862671
+I1204 20:51:23.620351 137274321021824 utils.py:1231] [95900] lr = 6.39841829324287e-05
+I1204 20:51:23.620409 137274321021824 utils.py:1231] [95900] uptime = 601272.982771092
+I1204 20:51:23.620474 137274321021824 utils.py:1231] [95900] examples_seen = 98201600.0
+I1204 20:51:23.620525 137274321021824 utils.py:1231] [95900] progress = 0.8516646980986297
+I1204 20:51:23.620586 137274321021824 utils.py:1231] [95900] epoch = 76.65011665145917
+I1204 20:51:23.620643 137274321021824 utils.py:1231] [95900] img/sec/core = 164.20832845014107
+I1204 20:51:23.620707 137274321021824 utils.py:1231] [95900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 166.98595538961334
+I1204 20:51:23.620771 137274321021824 utils.py:1231] [95900] core_hours = 166.98595538961334
+I1204 20:51:23.620841 137274321021824 train.py:125] NOTE: Steps:95900/112603 [85.2%]
+Walltime:6d23h1m (0s eval)
+ETA:1d5h5m
+Total train time:8d4h4m
+I1204 20:56:35.409415 137274321021824 utils.py:1231] [95950] l2_params = 242.1383505620438
+I1204 20:56:35.409692 137274321021824 utils.py:1231] [95950] train/loss = 1.4526079595088959
+I1204 20:56:35.409813 137274321021824 utils.py:1231] [95950] l2_grads = 2.5655150413513184
+I1204 20:56:35.409900 137274321021824 utils.py:1231] [95950] lr = 6.36100337020321e-05
+I1204 20:56:35.409968 137274321021824 utils.py:1231] [95950] uptime = 601584.772329028
+I1204 20:56:35.410032 137274321021824 utils.py:1231] [95950] examples_seen = 98252800.0
+I1204 20:56:35.410090 137274321021824 utils.py:1231] [95950] progress = 0.8521087360017051
+I1204 20:56:35.410151 137274321021824 utils.py:1231] [95950] epoch = 76.69008021592813
+I1204 20:56:35.410209 137274321021824 utils.py:1231] [95950] img/sec/core = 164.21332497128236
+I1204 20:56:35.410274 137274321021824 utils.py:1231] [95950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.0725636001511
+I1204 20:56:35.410334 137274321021824 utils.py:1231] [95950] core_hours = 167.0725636001511
+I1204 20:56:35.410402 137274321021824 train.py:125] NOTE: Steps:95950/112603 [85.2%]
+Walltime:6d23h6m (0s eval)
+ETA:1d4h59m
+Total train time:8d4h4m
+I1204 21:01:47.132034 137274321021824 utils.py:1231] [96000] l2_params = 242.105060513346
+I1204 21:01:47.132253 137274321021824 utils.py:1231] [96000] train/loss = 3.6450426280498505
+I1204 21:01:47.132353 137274321021824 utils.py:1231] [96000] l2_grads = 2.4930834770202637
+I1204 21:01:47.132428 137274321021824 utils.py:1231] [96000] lr = 6.323690728008544e-05
+I1204 21:01:47.132489 137274321021824 utils.py:1231] [96000] uptime = 601896.494850487
+I1204 21:01:47.132549 137274321021824 utils.py:1231] [96000] examples_seen = 98304000.0
+I1204 21:01:47.132606 137274321021824 utils.py:1231] [96000] progress = 0.8525527739047805
+I1204 21:01:47.132665 137274321021824 utils.py:1231] [96000] epoch = 76.7300437803971
+I1204 21:01:47.132722 137274321021824 utils.py:1231] [96000] img/sec/core = 164.24863933591854
+I1204 21:01:47.132785 137274321021824 utils.py:1231] [96000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.15915318944525
+I1204 21:01:47.132842 137274321021824 utils.py:1231] [96000] core_hours = 167.15915318944525
+I1204 21:01:47.132921 137274321021824 train.py:125] NOTE: Steps:96000/112603 [85.3%]
+Walltime:6d23h11m (0s eval)
+ETA:1d4h54m
+Total train time:8d4h4m
+I1204 21:06:59.218490 137274321021824 utils.py:1231] [96050] l2_params = 242.0729310583978
+I1204 21:06:59.218707 137274321021824 utils.py:1231] [96050] train/loss = 1.5632643550634384
+I1204 21:06:59.218802 137274321021824 utils.py:1231] [96050] l2_grads = 2.7915894985198975
+I1204 21:06:59.218865 137274321021824 utils.py:1231] [96050] lr = 6.286480454112006e-05
+I1204 21:06:59.218922 137274321021824 utils.py:1231] [96050] uptime = 602208.581284493
+I1204 21:06:59.218974 137274321021824 utils.py:1231] [96050] examples_seen = 98355200.0
+I1204 21:06:59.219023 137274321021824 utils.py:1231] [96050] progress = 0.8529968118078559
+I1204 21:06:59.219071 137274321021824 utils.py:1231] [96050] epoch = 76.77000734486604
+I1204 21:06:59.219123 137274321021824 utils.py:1231] [96050] img/sec/core = 164.0571150203116
+I1204 21:06:59.219178 137274321021824 utils.py:1231] [96050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.24584386555804
+I1204 21:06:59.219228 137274321021824 utils.py:1231] [96050] core_hours = 167.24584386555804
+I1204 21:06:59.219289 137274321021824 train.py:125] NOTE: Steps:96050/112603 [85.3%]
+Walltime:6d23h16m (0s eval)
+ETA:1d4h49m
+Total train time:8d4h4m
+I1204 21:12:10.973873 137274321021824 utils.py:1231] [96100] l2_params = 242.0398411082448
+I1204 21:12:10.974086 137274321021824 utils.py:1231] [96100] train/loss = 3.833412855863571
+I1204 21:12:10.974188 137274321021824 utils.py:1231] [96100] l2_grads = 2.7772622108459473
+I1204 21:12:10.974263 137274321021824 utils.py:1231] [96100] lr = 6.249372635726878e-05
+I1204 21:12:10.974351 137274321021824 utils.py:1231] [96100] uptime = 602520.33671047
+I1204 21:12:10.974416 137274321021824 utils.py:1231] [96100] examples_seen = 98406400.0
+I1204 21:12:10.974476 137274321021824 utils.py:1231] [96100] progress = 0.8534408497109314
+I1204 21:12:10.974535 137274321021824 utils.py:1231] [96100] epoch = 76.80997090933501
+I1204 21:12:10.974597 137274321021824 utils.py:1231] [96100] img/sec/core = 164.23130355963167
+I1204 21:12:10.974663 137274321021824 utils.py:1231] [96100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.33244259499614
+I1204 21:12:10.974720 137274321021824 utils.py:1231] [96100] core_hours = 167.33244259499614
+I1204 21:12:10.974791 137274321021824 train.py:125] NOTE: Steps:96100/112603 [85.3%]
+Walltime:6d23h22m (0s eval)
+ETA:1d4h44m
+Total train time:8d4h4m
+I1204 21:17:22.755454 137274321021824 utils.py:1231] [96150] l2_params = 242.00720343910442
+I1204 21:17:22.755695 137274321021824 utils.py:1231] [96150] train/loss = 1.5873979181051254
+I1204 21:17:22.755848 137274321021824 utils.py:1231] [96150] l2_grads = 2.5221962928771973
+I1204 21:17:22.755930 137274321021824 utils.py:1231] [96150] lr = 6.21236735982623e-05
+I1204 21:17:22.755993 137274321021824 utils.py:1231] [96150] uptime = 602832.118353386
+I1204 21:17:22.756053 137274321021824 utils.py:1231] [96150] examples_seen = 98457600.0
+I1204 21:17:22.756104 137274321021824 utils.py:1231] [96150] progress = 0.8538848876140067
+I1204 21:17:22.756160 137274321021824 utils.py:1231] [96150] epoch = 76.84993447380396
+I1204 21:17:22.756217 137274321021824 utils.py:1231] [96150] img/sec/core = 164.2174937599146
+I1204 21:17:22.756277 137274321021824 utils.py:1231] [96150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.41904860691722
+I1204 21:17:22.756333 137274321021824 utils.py:1231] [96150] core_hours = 167.41904860691722
+I1204 21:17:22.756401 137274321021824 train.py:125] NOTE: Steps:96150/112603 [85.4%]
+Walltime:6d23h27m (0s eval)
+ETA:1d4h38m
+Total train time:8d4h4m
+I1204 21:22:34.564454 137274321021824 utils.py:1231] [96200] l2_params = 241.97213450021553
+I1204 21:22:34.564657 137274321021824 utils.py:1231] [96200] train/loss = 1.6179668009281158
+I1204 21:22:34.564754 137274321021824 utils.py:1231] [96200] l2_grads = 2.6015241146087646
+I1204 21:22:34.564816 137274321021824 utils.py:1231] [96200] lr = 6.175464713142834e-05
+I1204 21:22:34.564867 137274321021824 utils.py:1231] [96200] uptime = 603143.92722849
+I1204 21:22:34.564924 137274321021824 utils.py:1231] [96200] examples_seen = 98508800.0
+I1204 21:22:34.564972 137274321021824 utils.py:1231] [96200] progress = 0.8543289255170822
+I1204 21:22:34.565020 137274321021824 utils.py:1231] [96200] epoch = 76.88989803827292
+I1204 21:22:34.565068 137274321021824 utils.py:1231] [96200] img/sec/core = 164.203151635477
+I1204 21:22:34.565124 137274321021824 utils.py:1231] [96200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.505662183335
+I1204 21:22:34.565172 137274321021824 utils.py:1231] [96200] core_hours = 167.505662183335
+I1204 21:22:34.565233 137274321021824 train.py:125] NOTE: Steps:96200/112603 [85.4%]
+Walltime:6d23h32m (0s eval)
+ETA:1d4h33m
+Total train time:8d4h4m
+I1204 21:27:46.314807 137274321021824 utils.py:1231] [96250] l2_params = 241.94003685079701
+I1204 21:27:46.315020 137274321021824 utils.py:1231] [96250] train/loss = 2.912402182817459
+I1204 21:27:46.315118 137274321021824 utils.py:1231] [96250] l2_grads = 2.4144680500030518
+I1204 21:27:46.315188 137274321021824 utils.py:1231] [96250] lr = 6.13866478216894e-05
+I1204 21:27:46.315247 137274321021824 utils.py:1231] [96250] uptime = 603455.677608877
+I1204 21:27:46.315308 137274321021824 utils.py:1231] [96250] examples_seen = 98560000.0
+I1204 21:27:46.315364 137274321021824 utils.py:1231] [96250] progress = 0.8547729634201575
+I1204 21:27:46.315425 137274321021824 utils.py:1231] [96250] epoch = 76.92986160274188
+I1204 21:27:46.315480 137274321021824 utils.py:1231] [96250] img/sec/core = 164.2339615959732
+I1204 21:27:46.315541 137274321021824 utils.py:1231] [96250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.59225951122025
+I1204 21:27:46.315615 137274321021824 utils.py:1231] [96250] core_hours = 167.59225951122025
+I1204 21:27:46.315695 137274321021824 train.py:125] NOTE: Steps:96250/112603 [85.5%]
+Walltime:6d23h37m (0s eval)
+ETA:1d4h28m
+Total train time:8d4h4m
+I1204 21:32:58.112021 137274321021824 utils.py:1231] [96300] l2_params = 241.9067054084401
+I1204 21:32:58.112229 137274321021824 utils.py:1231] [96300] train/loss = 2.976105809211731
+I1204 21:32:58.112329 137274321021824 utils.py:1231] [96300] l2_grads = 2.4312100410461426
+I1204 21:32:58.112399 137274321021824 utils.py:1231] [96300] lr = 6.101967653155976e-05
+I1204 21:32:58.112464 137274321021824 utils.py:1231] [96300] uptime = 603767.474820921
+I1204 21:32:58.112535 137274321021824 utils.py:1231] [96300] examples_seen = 98611200.0
+I1204 21:32:58.112590 137274321021824 utils.py:1231] [96300] progress = 0.855217001323233
+I1204 21:32:58.112645 137274321021824 utils.py:1231] [96300] epoch = 76.96982516721083
+I1204 21:32:58.112702 137274321021824 utils.py:1231] [96300] img/sec/core = 164.2092938046321
+I1204 21:32:58.112765 137274321021824 utils.py:1231] [96300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.67886984789914
+I1204 21:32:58.112817 137274321021824 utils.py:1231] [96300] core_hours = 167.67886984789914
+I1204 21:32:58.112889 137274321021824 train.py:125] NOTE: Steps:96300/112603 [85.5%]
+Walltime:6d23h42m (0s eval)
+ETA:1d4h23m
+Total train time:8d4h4m
+I1204 21:38:09.909094 137274321021824 utils.py:1231] [96350] l2_params = 241.8749683235946
+I1204 21:38:09.909461 137274321021824 utils.py:1231] [96350] train/loss = 2.98815456032753
+I1204 21:38:09.909646 137274321021824 utils.py:1231] [96350] l2_grads = 2.460334539413452
+I1204 21:38:09.909737 137274321021824 utils.py:1231] [96350] lr = 6.065373412114527e-05
+I1204 21:38:09.909805 137274321021824 utils.py:1231] [96350] uptime = 604079.272167052
+I1204 21:38:09.909863 137274321021824 utils.py:1231] [96350] examples_seen = 98662400.0
+I1204 21:38:09.909927 137274321021824 utils.py:1231] [96350] progress = 0.8556610392263083
+I1204 21:38:09.909981 137274321021824 utils.py:1231] [96350] epoch = 77.00978873167979
+I1204 21:38:09.910036 137274321021824 utils.py:1231] [96350] img/sec/core = 164.20922318716126
+I1204 21:38:09.910097 137274321021824 utils.py:1231] [96350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.76548022182445
+I1204 21:38:09.910151 137274321021824 utils.py:1231] [96350] core_hours = 167.76548022182445
+I1204 21:38:09.910215 137274321021824 train.py:125] NOTE: Steps:96350/112603 [85.6%]
+Walltime:6d23h47m (0s eval)
+ETA:1d4h18m
+Total train time:8d4h4m
+I1204 21:43:21.748837 137274321021824 utils.py:1231] [96400] l2_params = 241.8419219014603
+I1204 21:43:21.749084 137274321021824 utils.py:1231] [96400] train/loss = 1.5970513075590134
+I1204 21:43:21.749204 137274321021824 utils.py:1231] [96400] l2_grads = 2.576744794845581
+I1204 21:43:21.749280 137274321021824 utils.py:1231] [96400] lr = 6.028882144813961e-05
+I1204 21:43:21.749340 137274321021824 utils.py:1231] [96400] uptime = 604391.111698933
+I1204 21:43:21.749400 137274321021824 utils.py:1231] [96400] examples_seen = 98713600.0
+I1204 21:43:21.749464 137274321021824 utils.py:1231] [96400] progress = 0.8561050771293838
+I1204 21:43:21.749517 137274321021824 utils.py:1231] [96400] epoch = 77.04975229614874
+I1204 21:43:21.749571 137274321021824 utils.py:1231] [96400] img/sec/core = 164.1870089118509
+I1204 21:43:21.749638 137274321021824 utils.py:1231] [96400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.8521023140136
+I1204 21:43:21.749689 137274321021824 utils.py:1231] [96400] core_hours = 167.8521023140136
+I1204 21:43:21.749752 137274321021824 train.py:125] NOTE: Steps:96400/112603 [85.6%]
+Walltime:6d23h53m (0s eval)
+ETA:1d4h12m
+Total train time:8d4h4m
+I1204 21:48:33.550258 137274321021824 utils.py:1231] [96450] l2_params = 241.81048052396326
+I1204 21:48:33.550482 137274321021824 utils.py:1231] [96450] train/loss = 3.4038892686367035
+I1204 21:48:33.550578 137274321021824 utils.py:1231] [96450] l2_grads = 2.6155848503112793
+I1204 21:48:33.550654 137274321021824 utils.py:1231] [96450] lr = 5.992493936782296e-05
+I1204 21:48:33.550708 137274321021824 utils.py:1231] [96450] uptime = 604702.913069494
+I1204 21:48:33.550766 137274321021824 utils.py:1231] [96450] examples_seen = 98764800.0
+I1204 21:48:33.550817 137274321021824 utils.py:1231] [96450] progress = 0.8565491150324591
+I1204 21:48:33.550867 137274321021824 utils.py:1231] [96450] epoch = 77.0897158606177
+I1204 21:48:33.550925 137274321021824 utils.py:1231] [96450] img/sec/core = 164.20710373363258
+I1204 21:48:33.550984 137274321021824 utils.py:1231] [96450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 167.93871380583607
+I1204 21:48:33.551036 137274321021824 utils.py:1231] [96450] core_hours = 167.93871380583607
+I1204 21:48:33.551098 137274321021824 train.py:125] NOTE: Steps:96450/112603 [85.7%]
+Walltime:6d23h58m (0s eval)
+ETA:1d4h7m
+Total train time:8d4h4m
+I1204 21:53:45.340225 137274321021824 utils.py:1231] [96500] l2_params = 241.77738267636465
+I1204 21:53:45.340450 137274321021824 utils.py:1231] [96500] train/loss = 2.0641332119703293
+I1204 21:53:45.340569 137274321021824 utils.py:1231] [96500] l2_grads = 2.459277391433716
+I1204 21:53:45.340659 137274321021824 utils.py:1231] [96500] lr = 5.956208873306049e-05
+I1204 21:53:45.340738 137274321021824 utils.py:1231] [96500] uptime = 605014.7030935109
+I1204 21:53:45.340813 137274321021824 utils.py:1231] [96500] examples_seen = 98816000.0
+I1204 21:53:45.340874 137274321021824 utils.py:1231] [96500] progress = 0.8569931529355346
+I1204 21:53:45.340946 137274321021824 utils.py:1231] [96500] epoch = 77.12967942508666
+I1204 21:53:45.341006 137274321021824 utils.py:1231] [96500] img/sec/core = 164.213079496126
+I1204 21:53:45.341070 137274321021824 utils.py:1231] [96500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.0253221458408
+I1204 21:53:45.341123 137274321021824 utils.py:1231] [96500] core_hours = 168.0253221458408
+I1204 21:53:45.341187 137274321021824 train.py:125] NOTE: Steps:96500/112603 [85.7%]
+Walltime:7d0h3m (0s eval)
+ETA:1d4h2m
+Total train time:8d4h4m
+I1204 21:58:57.134434 137274321021824 utils.py:1231] [96550] l2_params = 241.74450837690975
+I1204 21:58:57.134651 137274321021824 utils.py:1231] [96550] train/loss = 1.5351468324661255
+I1204 21:58:57.134793 137274321021824 utils.py:1231] [96550] l2_grads = 2.6726291179656982
+I1204 21:58:57.134907 137274321021824 utils.py:1231] [96550] lr = 5.9200270394299225e-05
+I1204 21:58:57.134982 137274321021824 utils.py:1231] [96550] uptime = 605326.49734165
+I1204 21:58:57.135048 137274321021824 utils.py:1231] [96550] examples_seen = 98867200.0
+I1204 21:58:57.135126 137274321021824 utils.py:1231] [96550] progress = 0.85743719083861
+I1204 21:58:57.135254 137274321021824 utils.py:1231] [96550] epoch = 77.16964298955561
+I1204 21:58:57.135364 137274321021824 utils.py:1231] [96550] img/sec/core = 164.21085477231574
+I1204 21:58:57.135447 137274321021824 utils.py:1231] [96550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.11193165921276
+I1204 21:58:57.135505 137274321021824 utils.py:1231] [96550] core_hours = 168.11193165921276
+I1204 21:58:57.135602 137274321021824 train.py:125] NOTE: Steps:96550/112603 [85.7%]
+Walltime:7d0h8m (0s eval)
+ETA:1d3h57m
+Total train time:8d4h4m
+I1204 22:04:08.909019 137274321021824 utils.py:1231] [96600] l2_params = 241.7099593422722
+I1204 22:04:08.909251 137274321021824 utils.py:1231] [96600] train/loss = 3.6358506083488464
+I1204 22:04:08.909357 137274321021824 utils.py:1231] [96600] l2_grads = 2.6892411708831787
+I1204 22:04:08.909431 137274321021824 utils.py:1231] [96600] lr = 5.883948519956733e-05
+I1204 22:04:08.909492 137274321021824 utils.py:1231] [96600] uptime = 605638.271854102
+I1204 22:04:08.909583 137274321021824 utils.py:1231] [96600] examples_seen = 98918400.0
+I1204 22:04:08.909651 137274321021824 utils.py:1231] [96600] progress = 0.8578812287416854
+I1204 22:04:08.909706 137274321021824 utils.py:1231] [96600] epoch = 77.20960655402457
+I1204 22:04:08.909763 137274321021824 utils.py:1231] [96600] img/sec/core = 164.22124950919638
+I1204 22:04:08.909836 137274321021824 utils.py:1231] [96600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.19853569044943
+I1204 22:04:08.909901 137274321021824 utils.py:1231] [96600] core_hours = 168.19853569044943
+I1204 22:04:08.909992 137274321021824 train.py:125] NOTE: Steps:96600/112603 [85.8%]
+Walltime:7d0h13m (0s eval)
+ETA:1d3h51m
+Total train time:8d4h4m
+I1204 22:09:20.684108 137274321021824 utils.py:1231] [96650] l2_params = 241.6805910098826
+I1204 22:09:20.684313 137274321021824 utils.py:1231] [96650] train/loss = 1.5690958946943283
+I1204 22:09:20.684407 137274321021824 utils.py:1231] [96650] l2_grads = 2.525724172592163
+I1204 22:09:20.684468 137274321021824 utils.py:1231] [96650] lr = 5.847973399447081e-05
+I1204 22:09:20.684521 137274321021824 utils.py:1231] [96650] uptime = 605950.046882476
+I1204 22:09:20.684575 137274321021824 utils.py:1231] [96650] examples_seen = 98969600.0
+I1204 22:09:20.684625 137274321021824 utils.py:1231] [96650] progress = 0.8583252666447608
+I1204 22:09:20.684674 137274321021824 utils.py:1231] [96650] epoch = 77.24957011849352
+I1204 22:09:20.684726 137274321021824 utils.py:1231] [96650] img/sec/core = 164.22097775769555
+I1204 22:09:20.684784 137274321021824 utils.py:1231] [96650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.28513986499777
+I1204 22:09:20.684835 137274321021824 utils.py:1231] [96650] core_hours = 168.28513986499777
+I1204 22:09:20.684906 137274321021824 train.py:125] NOTE: Steps:96650/112603 [85.8%]
+Walltime:7d0h19m (0s eval)
+ETA:1d3h46m
+Total train time:8d4h3m
+I1204 22:14:32.477408 137274321021824 utils.py:1231] [96700] l2_params = 241.6488494424317
+I1204 22:14:32.477677 137274321021824 utils.py:1231] [96700] train/loss = 3.9038797318935394
+I1204 22:14:32.477788 137274321021824 utils.py:1231] [96700] l2_grads = 2.791175603866577
+I1204 22:14:32.477871 137274321021824 utils.py:1231] [96700] lr = 5.8121017622193114e-05
+I1204 22:14:32.477935 137274321021824 utils.py:1231] [96700] uptime = 606261.840296754
+I1204 22:14:32.477987 137274321021824 utils.py:1231] [96700] examples_seen = 99020800.0
+I1204 22:14:32.478037 137274321021824 utils.py:1231] [96700] progress = 0.8587693045478362
+I1204 22:14:32.478086 137274321021824 utils.py:1231] [96700] epoch = 77.28953368296249
+I1204 22:14:32.478137 137274321021824 utils.py:1231] [96700] img/sec/core = 164.21129393816312
+I1204 22:14:32.478193 137274321021824 utils.py:1231] [96700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.37174914674168
+I1204 22:14:32.478243 137274321021824 utils.py:1231] [96700] core_hours = 168.37174914674168
+I1204 22:14:32.478303 137274321021824 train.py:125] NOTE: Steps:96700/112603 [85.9%]
+Walltime:7d0h24m (0s eval)
+ETA:1d3h41m
+Total train time:8d4h3m
+I1204 22:19:44.263542 137274321021824 utils.py:1231] [96750] l2_params = 241.617005123926
+I1204 22:19:44.263768 137274321021824 utils.py:1231] [96750] train/loss = 1.5957248657941818
+I1204 22:19:44.263928 137274321021824 utils.py:1231] [96750] l2_grads = 2.6044890880584717
+I1204 22:19:44.264043 137274321021824 utils.py:1231] [96750] lr = 5.776333692349144e-05
+I1204 22:19:44.264137 137274321021824 utils.py:1231] [96750] uptime = 606573.626494401
+I1204 22:19:44.264220 137274321021824 utils.py:1231] [96750] examples_seen = 99072000.0
+I1204 22:19:44.264299 137274321021824 utils.py:1231] [96750] progress = 0.8592133424509116
+I1204 22:19:44.264378 137274321021824 utils.py:1231] [96750] epoch = 77.32949724743145
+I1204 22:19:44.264448 137274321021824 utils.py:1231] [96750] img/sec/core = 164.21509478739054
+I1204 22:19:44.264522 137274321021824 utils.py:1231] [96750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.45835642386584
+I1204 22:19:44.264592 137274321021824 utils.py:1231] [96750] core_hours = 168.45835642386584
+I1204 22:19:44.264670 137274321021824 train.py:125] NOTE: Steps:96750/112603 [85.9%]
+Walltime:7d0h29m (0s eval)
+ETA:1d3h36m
+Total train time:8d4h3m
+I1204 22:24:56.293398 137274321021824 utils.py:1231] [96800] l2_params = 241.5857186043978
+I1204 22:24:56.293612 137274321021824 utils.py:1231] [96800] train/loss = 1.837610051035881
+I1204 22:24:56.293720 137274321021824 utils.py:1231] [96800] l2_grads = 2.6105048656463623
+I1204 22:24:56.293795 137274321021824 utils.py:1231] [96800] lr = 5.7406692736695825e-05
+I1204 22:24:56.293858 137274321021824 utils.py:1231] [96800] uptime = 606885.656219365
+I1204 22:24:56.293926 137274321021824 utils.py:1231] [96800] examples_seen = 99123200.0
+I1204 22:24:56.293984 137274321021824 utils.py:1231] [96800] progress = 0.859657380353987
+I1204 22:24:56.294041 137274321021824 utils.py:1231] [96800] epoch = 77.3694608119004
+I1204 22:24:56.294103 137274321021824 utils.py:1231] [96800] img/sec/core = 164.08693115989786
+I1204 22:24:56.294173 137274321021824 utils.py:1231] [96800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.54503134746693
+I1204 22:24:56.294233 137274321021824 utils.py:1231] [96800] core_hours = 168.54503134746693
+I1204 22:24:56.294295 137274321021824 train.py:125] NOTE: Steps:96800/112603 [86.0%]
+Walltime:7d0h34m (0s eval)
+ETA:1d3h30m
+Total train time:8d4h3m
+I1204 22:30:08.075901 137274321021824 utils.py:1231] [96850] l2_params = 241.5547123799826
+I1204 22:30:08.076165 137274321021824 utils.py:1231] [96850] train/loss = 1.6008794456720352
+I1204 22:30:08.076284 137274321021824 utils.py:1231] [96850] l2_grads = 2.552703857421875
+I1204 22:30:08.076376 137274321021824 utils.py:1231] [96850] lr = 5.705108589770723e-05
+I1204 22:30:08.076451 137274321021824 utils.py:1231] [96850] uptime = 607197.438812978
+I1204 22:30:08.076513 137274321021824 utils.py:1231] [96850] examples_seen = 99174400.0
+I1204 22:30:08.076571 137274321021824 utils.py:1231] [96850] progress = 0.8601014182570624
+I1204 22:30:08.076632 137274321021824 utils.py:1231] [96850] epoch = 77.40942437636936
+I1204 22:30:08.076695 137274321021824 utils.py:1231] [96850] img/sec/core = 164.21699302286228
+I1204 22:30:08.076758 137274321021824 utils.py:1231] [96850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.63163762347054
+I1204 22:30:08.076815 137274321021824 utils.py:1231] [96850] core_hours = 168.63163762347054
+I1204 22:30:08.076890 137274321021824 train.py:125] NOTE: Steps:96850/112603 [86.0%]
+Walltime:7d0h39m (0s eval)
+ETA:1d3h25m
+Total train time:8d4h3m
+I1204 22:35:19.872461 137274321021824 utils.py:1231] [96900] l2_params = 241.5241053466349
+I1204 22:35:19.872707 137274321021824 utils.py:1231] [96900] train/loss = 1.590329959988594
+I1204 22:35:19.872826 137274321021824 utils.py:1231] [96900] l2_grads = 2.5485336780548096
+I1204 22:35:19.872908 137274321021824 utils.py:1231] [96900] lr = 5.6696517239994664e-05
+I1204 22:35:19.872962 137274321021824 utils.py:1231] [96900] uptime = 607509.23532355
+I1204 22:35:19.873014 137274321021824 utils.py:1231] [96900] examples_seen = 99225600.0
+I1204 22:35:19.873062 137274321021824 utils.py:1231] [96900] progress = 0.8605454561601378
+I1204 22:35:19.873110 137274321021824 utils.py:1231] [96900] epoch = 77.44938794083832
+I1204 22:35:19.873160 137274321021824 utils.py:1231] [96900] img/sec/core = 164.20966323862356
+I1204 22:35:19.873224 137274321021824 utils.py:1231] [96900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.71824776529607
+I1204 22:35:19.873276 137274321021824 utils.py:1231] [96900] core_hours = 168.71824776529607
+I1204 22:35:19.873340 137274321021824 train.py:125] NOTE: Steps:96900/112603 [86.1%]
+Walltime:7d0h45m (0s eval)
+ETA:1d3h20m
+Total train time:8d4h3m
+I1204 22:40:31.643589 137274321021824 utils.py:1231] [96950] l2_params = 241.4937636283978
+I1204 22:40:31.643822 137274321021824 utils.py:1231] [96950] train/loss = 1.5291985422372818
+I1204 22:40:31.643940 137274321021824 utils.py:1231] [96950] l2_grads = 2.6470983028411865
+I1204 22:40:31.644027 137274321021824 utils.py:1231] [96950] lr = 5.634298759459468e-05
+I1204 22:40:31.644090 137274321021824 utils.py:1231] [96950] uptime = 607821.006451961
+I1204 22:40:31.644140 137274321021824 utils.py:1231] [96950] examples_seen = 99276800.0
+I1204 22:40:31.644187 137274321021824 utils.py:1231] [96950] progress = 0.8609894940632132
+I1204 22:40:31.644236 137274321021824 utils.py:1231] [96950] epoch = 77.48935150530727
+I1204 22:40:31.644285 137274321021824 utils.py:1231] [96950] img/sec/core = 164.2230320072974
+I1204 22:40:31.644337 137274321021824 utils.py:1231] [96950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.80485085652137
+I1204 22:40:31.644386 137274321021824 utils.py:1231] [96950] core_hours = 168.80485085652137
+I1204 22:40:31.644445 137274321021824 train.py:125] NOTE: Steps:96950/112603 [86.1%]
+Walltime:7d0h50m (0s eval)
+ETA:1d3h15m
+Total train time:8d4h3m
+I1204 22:45:43.422038 137274321021824 utils.py:1231] [97000] l2_params = 241.4640657869831
+I1204 22:45:43.422359 137274321021824 utils.py:1231] [97000] train/loss = 1.3458739668130875
+I1204 22:45:43.422595 137274321021824 utils.py:1231] [97000] l2_grads = 2.459265947341919
+I1204 22:45:43.422740 137274321021824 utils.py:1231] [97000] lr = 5.5990497790108014e-05
+I1204 22:45:43.422863 137274321021824 utils.py:1231] [97000] uptime = 608132.785210511
+I1204 22:45:43.422972 137274321021824 utils.py:1231] [97000] examples_seen = 99328000.0
+I1204 22:45:43.423048 137274321021824 utils.py:1231] [97000] progress = 0.8614335319662887
+I1204 22:45:43.423121 137274321021824 utils.py:1231] [97000] epoch = 77.52931506977623
+I1204 22:45:43.423187 137274321021824 utils.py:1231] [97000] img/sec/core = 164.21901298897455
+I1204 22:45:43.423272 137274321021824 utils.py:1231] [97000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.8914560672297
+I1204 22:45:43.423343 137274321021824 utils.py:1231] [97000] core_hours = 168.8914560672297
+I1204 22:45:43.423428 137274321021824 train.py:125] NOTE: Steps:97000/112603 [86.1%]
+Walltime:7d0h55m (0s eval)
+ETA:1d3h10m
+Total train time:8d4h3m
+I1204 22:50:55.596304 137274321021824 utils.py:1231] [97050] l2_params = 241.43236840208863
+I1204 22:50:55.596520 137274321021824 utils.py:1231] [97050] train/loss = 2.9623195230960846
+I1204 22:50:55.596662 137274321021824 utils.py:1231] [97050] l2_grads = 2.4076414108276367
+I1204 22:50:55.596771 137274321021824 utils.py:1231] [97050] lr = 5.563904865269814e-05
+I1204 22:50:55.596857 137274321021824 utils.py:1231] [97050] uptime = 608444.9592134
+I1204 22:50:55.596959 137274321021824 utils.py:1231] [97050] examples_seen = 99379200.0
+I1204 22:50:55.597038 137274321021824 utils.py:1231] [97050] progress = 0.861877569869364
+I1204 22:50:55.597118 137274321021824 utils.py:1231] [97050] epoch = 77.56927863424518
+I1204 22:50:55.597192 137274321021824 utils.py:1231] [97050] img/sec/core = 164.01109485786495
+I1204 22:50:55.597269 137274321021824 utils.py:1231] [97050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 168.97817106803222
+I1204 22:50:55.597332 137274321021824 utils.py:1231] [97050] core_hours = 168.97817106803222
+I1204 22:50:55.597410 137274321021824 train.py:125] NOTE: Steps:97050/112603 [86.2%]
+Walltime:7d1h0m (0s eval)
+ETA:1d3h4m
+Total train time:8d4h3m
+I1204 22:56:07.388708 137274321021824 utils.py:1231] [97100] l2_params = 241.40046493018144
+I1204 22:56:07.388976 137274321021824 utils.py:1231] [97100] train/loss = 3.804466098546982
+I1204 22:56:07.389109 137274321021824 utils.py:1231] [97100] l2_grads = 2.734043836593628
+I1204 22:56:07.389216 137274321021824 utils.py:1231] [97100] lr = 5.528864100608995e-05
+I1204 22:56:07.389332 137274321021824 utils.py:1231] [97100] uptime = 608756.75168578
+I1204 22:56:07.389433 137274321021824 utils.py:1231] [97100] examples_seen = 99430400.0
+I1204 22:56:07.389506 137274321021824 utils.py:1231] [97100] progress = 0.8623216077724395
+I1204 22:56:07.389585 137274321021824 utils.py:1231] [97100] epoch = 77.60924219871414
+I1204 22:56:07.389674 137274321021824 utils.py:1231] [97100] img/sec/core = 164.21179000627924
+I1204 22:56:07.389743 137274321021824 utils.py:1231] [97100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.06478008813778
+I1204 22:56:07.389802 137274321021824 utils.py:1231] [97100] core_hours = 169.06478008813778
+I1204 22:56:07.389871 137274321021824 train.py:125] NOTE: Steps:97100/112603 [86.2%]
+Walltime:7d1h5m (0s eval)
+ETA:1d2h59m
+Total train time:8d4h3m
+I1204 23:01:19.177545 137274321021824 utils.py:1231] [97150] l2_params = 241.3705118819944
+I1204 23:01:19.177767 137274321021824 utils.py:1231] [97150] train/loss = 1.5192686468362808
+I1204 23:01:19.177871 137274321021824 utils.py:1231] [97150] l2_grads = 2.7215349674224854
+I1204 23:01:19.177968 137274321021824 utils.py:1231] [97150] lr = 5.493927567156663e-05
+I1204 23:01:19.178036 137274321021824 utils.py:1231] [97150] uptime = 609068.540396785
+I1204 23:01:19.178098 137274321021824 utils.py:1231] [97150] examples_seen = 99481600.0
+I1204 23:01:19.178157 137274321021824 utils.py:1231] [97150] progress = 0.8627656456755148
+I1204 23:01:19.178215 137274321021824 utils.py:1231] [97150] epoch = 77.6492057631831
+I1204 23:01:19.178275 137274321021824 utils.py:1231] [97150] img/sec/core = 164.21377103412124
+I1204 23:01:19.178337 137274321021824 utils.py:1231] [97150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.15138806341696
+I1204 23:01:19.178405 137274321021824 utils.py:1231] [97150] core_hours = 169.15138806341696
+I1204 23:01:19.178488 137274321021824 train.py:125] NOTE: Steps:97150/112603 [86.3%]
+Walltime:7d1h11m (0s eval)
+ETA:1d2h54m
+Total train time:8d4h3m
+I1204 23:06:30.828223 137274321021824 utils.py:1231] [97200] l2_params = 241.33974746979533
+I1204 23:06:30.828461 137274321021824 utils.py:1231] [97200] train/loss = 1.5070694237947464
+I1204 23:06:30.828557 137274321021824 utils.py:1231] [97200] l2_grads = 2.7455804347991943
+I1204 23:06:30.828623 137274321021824 utils.py:1231] [97200] lr = 5.4590953467969194e-05
+I1204 23:06:30.828675 137274321021824 utils.py:1231] [97200] uptime = 609380.1910373659
+I1204 23:06:30.828727 137274321021824 utils.py:1231] [97200] examples_seen = 99532800.0
+I1204 23:06:30.828777 137274321021824 utils.py:1231] [97200] progress = 0.8632096835785903
+I1204 23:06:30.828827 137274321021824 utils.py:1231] [97200] epoch = 77.68916932765205
+I1204 23:06:30.828879 137274321021824 utils.py:1231] [97200] img/sec/core = 164.28652257721257
+I1204 23:06:30.828941 137274321021824 utils.py:1231] [97200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.23795768580055
+I1204 23:06:30.828992 137274321021824 utils.py:1231] [97200] core_hours = 169.23795768580055
+I1204 23:06:30.829054 137274321021824 train.py:125] NOTE: Steps:97200/112603 [86.3%]
+Walltime:7d1h16m (0s eval)
+ETA:1d2h49m
+Total train time:8d4h3m
+I1204 23:11:42.628992 137274321021824 utils.py:1231] [97250] l2_params = 241.31221660850798
+I1204 23:11:42.629294 137274321021824 utils.py:1231] [97250] train/loss = 3.808673322200775
+I1204 23:11:42.629539 137274321021824 utils.py:1231] [97250] l2_grads = 2.7302443981170654
+I1204 23:11:42.629677 137274321021824 utils.py:1231] [97250] lr = 5.424367521169318e-05
+I1204 23:11:42.629795 137274321021824 utils.py:1231] [97250] uptime = 609691.992151479
+I1204 23:11:42.629910 137274321021824 utils.py:1231] [97250] examples_seen = 99584000.0
+I1204 23:11:42.630011 137274321021824 utils.py:1231] [97250] progress = 0.8636537214816656
+I1204 23:11:42.630083 137274321021824 utils.py:1231] [97250] epoch = 77.72913289212102
+I1204 23:11:42.630154 137274321021824 utils.py:1231] [97250] img/sec/core = 164.20723878950952
+I1204 23:11:42.630229 137274321021824 utils.py:1231] [97250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.32456910638751
+I1204 23:11:42.630296 137274321021824 utils.py:1231] [97250] core_hours = 169.32456910638751
+I1204 23:11:42.630381 137274321021824 train.py:125] NOTE: Steps:97250/112603 [86.4%]
+Walltime:7d1h21m (0s eval)
+ETA:1d2h43m
+Total train time:8d4h3m
+I1204 23:16:54.435362 137274321021824 utils.py:1231] [97300] l2_params = 241.28371837886448
+I1204 23:16:54.435569 137274321021824 utils.py:1231] [97300] train/loss = 1.4503920078277588
+I1204 23:16:54.435657 137274321021824 utils.py:1231] [97300] l2_grads = 2.576587200164795
+I1204 23:16:54.435717 137274321021824 utils.py:1231] [97300] lr = 5.389744171668734e-05
+I1204 23:16:54.435768 137274321021824 utils.py:1231] [97300] uptime = 610003.798129697
+I1204 23:16:54.435819 137274321021824 utils.py:1231] [97300] examples_seen = 99635200.0
+I1204 23:16:54.435866 137274321021824 utils.py:1231] [97300] progress = 0.8640977593847411
+I1204 23:16:54.435919 137274321021824 utils.py:1231] [97300] epoch = 77.76909645658996
+I1204 23:16:54.435969 137274321021824 utils.py:1231] [97300] img/sec/core = 164.2046771925814
+I1204 23:16:54.436022 137274321021824 utils.py:1231] [97300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.41118187811472
+I1204 23:16:54.436069 137274321021824 utils.py:1231] [97300] core_hours = 169.41118187811472
+I1204 23:16:54.436126 137274321021824 train.py:125] NOTE: Steps:97300/112603 [86.4%]
+Walltime:7d1h26m (0s eval)
+ETA:1d2h38m
+Total train time:8d4h3m
+I1204 23:22:06.231711 137274321021824 utils.py:1231] [97350] l2_params = 241.25412788948557
+I1204 23:22:06.231935 137274321021824 utils.py:1231] [97350] train/loss = 2.712117314338684
+I1204 23:22:06.232042 137274321021824 utils.py:1231] [97350] l2_grads = 2.3909404277801514
+I1204 23:22:06.232120 137274321021824 utils.py:1231] [97350] lr = 5.3552253794452134e-05
+I1204 23:22:06.232188 137274321021824 utils.py:1231] [97350] uptime = 610315.594548969
+I1204 23:22:06.232248 137274321021824 utils.py:1231] [97350] examples_seen = 99686400.0
+I1204 23:22:06.232305 137274321021824 utils.py:1231] [97350] progress = 0.8645417972878164
+I1204 23:22:06.232359 137274321021824 utils.py:1231] [97350] epoch = 77.80906002105893
+I1204 23:22:06.232425 137274321021824 utils.py:1231] [97350] img/sec/core = 164.2097113223756
+I1204 23:22:06.232494 137274321021824 utils.py:1231] [97350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.49779199457913
+I1204 23:22:06.232550 137274321021824 utils.py:1231] [97350] core_hours = 169.49779199457913
+I1204 23:22:06.232631 137274321021824 train.py:125] NOTE: Steps:97350/112603 [86.5%]
+Walltime:7d1h31m (0s eval)
+ETA:1d2h33m
+Total train time:8d4h3m
+I1204 23:27:17.960925 137274321021824 utils.py:1231] [97400] l2_params = 241.22481268595178
+I1204 23:27:17.961137 137274321021824 utils.py:1231] [97400] train/loss = 3.460695803165436
+I1204 23:27:17.961237 137274321021824 utils.py:1231] [97400] l2_grads = 2.5499308109283447
+I1204 23:27:17.961313 137274321021824 utils.py:1231] [97400] lr = 5.320811225403677e-05
+I1204 23:27:17.961383 137274321021824 utils.py:1231] [97400] uptime = 610627.323743677
+I1204 23:27:17.961447 137274321021824 utils.py:1231] [97400] examples_seen = 99737600.0
+I1204 23:27:17.961507 137274321021824 utils.py:1231] [97400] progress = 0.8649858351908919
+I1204 23:27:17.961565 137274321021824 utils.py:1231] [97400] epoch = 77.84902358552789
+I1204 23:27:17.961624 137274321021824 utils.py:1231] [97400] img/sec/core = 164.24512323252566
+I1204 23:27:17.961694 137274321021824 utils.py:1231] [97400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.5843834375536
+I1204 23:27:17.961752 137274321021824 utils.py:1231] [97400] core_hours = 169.5843834375536
+I1204 23:27:17.961822 137274321021824 train.py:125] NOTE: Steps:97400/112603 [86.5%]
+Walltime:7d1h37m (0s eval)
+ETA:1d2h28m
+Total train time:8d4h3m
+I1204 23:32:29.763837 137274321021824 utils.py:1231] [97450] l2_params = 241.19650126043084
+I1204 23:32:29.764044 137274321021824 utils.py:1231] [97450] train/loss = 2.1983682215213776
+I1204 23:32:29.764138 137274321021824 utils.py:1231] [97450] l2_grads = 2.476973533630371
+I1204 23:32:29.764199 137274321021824 utils.py:1231] [97450] lr = 5.2865017902038755e-05
+I1204 23:32:29.764255 137274321021824 utils.py:1231] [97450] uptime = 610939.126616818
+I1204 23:32:29.764308 137274321021824 utils.py:1231] [97450] examples_seen = 99788800.0
+I1204 23:32:29.764360 137274321021824 utils.py:1231] [97450] progress = 0.8654298730939674
+I1204 23:32:29.764409 137274321021824 utils.py:1231] [97450] epoch = 77.88898714999684
+I1204 23:32:29.764461 137274321021824 utils.py:1231] [97450] img/sec/core = 164.20631241856552
+I1204 23:32:29.764522 137274321021824 utils.py:1231] [97450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.67099534675944
+I1204 23:32:29.764575 137274321021824 utils.py:1231] [97450] core_hours = 169.67099534675944
+I1204 23:32:29.764654 137274321021824 train.py:125] NOTE: Steps:97450/112603 [86.5%]
+Walltime:7d1h42m (0s eval)
+ETA:1d2h23m
+Total train time:8d4h3m
+I1204 23:37:41.561523 137274321021824 utils.py:1231] [97500] l2_params = 241.16702833245478
+I1204 23:37:41.561769 137274321021824 utils.py:1231] [97500] train/loss = 1.7059118747711182
+I1204 23:37:41.561914 137274321021824 utils.py:1231] [97500] l2_grads = 2.518988609313965
+I1204 23:37:41.561995 137274321021824 utils.py:1231] [97500] lr = 5.252297154260067e-05
+I1204 23:37:41.562058 137274321021824 utils.py:1231] [97500] uptime = 611250.924419216
+I1204 23:37:41.562133 137274321021824 utils.py:1231] [97500] examples_seen = 99840000.0
+I1204 23:37:41.562190 137274321021824 utils.py:1231] [97500] progress = 0.8658739109970427
+I1204 23:37:41.562248 137274321021824 utils.py:1231] [97500] epoch = 77.9289507144658
+I1204 23:37:41.562308 137274321021824 utils.py:1231] [97500] img/sec/core = 164.20898289285535
+I1204 23:37:41.562371 137274321021824 utils.py:1231] [97500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.75760584742557
+I1204 23:37:41.562437 137274321021824 utils.py:1231] [97500] core_hours = 169.75760584742557
+I1204 23:37:41.562508 137274321021824 train.py:125] NOTE: Steps:97500/112603 [86.6%]
+Walltime:7d1h47m (0s eval)
+ETA:1d2h17m
+Total train time:8d4h3m
+I1204 23:37:41.562607 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:97500/112603 [86.6%]
+Walltime:7d1h47m (0s eval)
+ETA:1d2h17m
+Total train time:8d4h3m
+I1204 23:39:19.956631 137274321021824 utils.py:1231] [97500] val/acc@1 = 0.7555404974489796
+I1204 23:39:19.956871 137274321021824 utils.py:1231] [97500] val/loss = 0.9566464282724322
+I1204 23:39:19.957063 137274321021824 utils.py:1231] [97500] z/secs/eval/val = 98.39437384810299
+I1204 23:39:19.957145 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.39437384810299
+I1204 23:44:31.388785 137274321021824 utils.py:1231] [97550] l2_params = 241.13926929250235
+I1204 23:44:31.389008 137274321021824 utils.py:1231] [97550] train/loss = 1.565350666642189
+I1204 23:44:31.389117 137274321021824 utils.py:1231] [97550] l2_grads = 2.676225423812866
+I1204 23:44:31.389185 137274321021824 utils.py:1231] [97550] lr = 5.218197397740876e-05
+I1204 23:44:31.389239 137274321021824 utils.py:1231] [97550] uptime = 611660.751600582
+I1204 23:44:31.389291 137274321021824 utils.py:1231] [97550] examples_seen = 99891200.0
+I1204 23:44:31.389346 137274321021824 utils.py:1231] [97550] progress = 0.8663179489001182
+I1204 23:44:31.389396 137274321021824 utils.py:1231] [97550] epoch = 77.96891427893475
+I1204 23:44:31.389448 137274321021824 utils.py:1231] [97550] img/sec/core = 124.93070818128686
+I1204 23:44:31.389504 137274321021824 utils.py:1231] [97550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.87144673113835
+I1204 23:44:31.389555 137274321021824 utils.py:1231] [97550] core_hours = 169.87144673113835
+I1204 23:44:31.389622 137274321021824 train.py:125] NOTE: Steps:97550/112603 [86.6%]
+Walltime:7d1h54m (0s eval)
+ETA:1d2h12m
+Total train time:8d4h5m
+I1204 23:49:43.163155 137274321021824 utils.py:1231] [97600] l2_params = 241.11091537680667
+I1204 23:49:43.163402 137274321021824 utils.py:1231] [97600] train/loss = 2.3090375661849976
+I1204 23:49:43.163503 137274321021824 utils.py:1231] [97600] l2_grads = 2.5186214447021484
+I1204 23:49:43.163573 137274321021824 utils.py:1231] [97600] lr = 5.1842026005691466e-05
+I1204 23:49:43.163635 137274321021824 utils.py:1231] [97600] uptime = 611972.525996706
+I1204 23:49:43.163696 137274321021824 utils.py:1231] [97600] examples_seen = 99942400.0
+I1204 23:49:43.163752 137274321021824 utils.py:1231] [97600] progress = 0.8667619868031935
+I1204 23:49:43.163807 137274321021824 utils.py:1231] [97600] epoch = 78.00887784340371
+I1204 23:49:43.163866 137274321021824 utils.py:1231] [97600] img/sec/core = 164.22131078283067
+I1204 23:49:43.163938 137274321021824 utils.py:1231] [97600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 169.95805073006164
+I1204 23:49:43.163993 137274321021824 utils.py:1231] [97600] core_hours = 169.95805073006164
+I1204 23:49:43.164057 137274321021824 train.py:125] NOTE: Steps:97600/112603 [86.7%]
+Walltime:7d1h59m (0s eval)
+ETA:1d2h7m
+Total train time:8d4h5m
+I1204 23:54:54.954284 137274321021824 utils.py:1231] [97650] l2_params = 241.08429927572806
+I1204 23:54:54.954480 137274321021824 utils.py:1231] [97650] train/loss = 1.4414406418800354
+I1204 23:54:54.954584 137274321021824 utils.py:1231] [97650] l2_grads = 2.7333943843841553
+I1204 23:54:54.954658 137274321021824 utils.py:1231] [97650] lr = 5.15031284242168e-05
+I1204 23:54:54.954720 137274321021824 utils.py:1231] [97650] uptime = 612284.317081482
+I1204 23:54:54.954783 137274321021824 utils.py:1231] [97650] examples_seen = 99993600.0
+I1204 23:54:54.954842 137274321021824 utils.py:1231] [97650] progress = 0.867206024706269
+I1204 23:54:54.954910 137274321021824 utils.py:1231] [97650] epoch = 78.04884140787267
+I1204 23:54:54.954970 137274321021824 utils.py:1231] [97650] img/sec/core = 164.21252081910978
+I1204 23:54:54.955029 137274321021824 utils.py:1231] [97650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.04465936472167
+I1204 23:54:54.955087 137274321021824 utils.py:1231] [97650] core_hours = 170.04465936472167
+I1204 23:54:54.955155 137274321021824 train.py:125] NOTE: Steps:97650/112603 [86.7%]
+Walltime:7d2h4m (0s eval)
+ETA:1d2h2m
+Total train time:8d4h5m
+I1205 00:00:06.599008 137274321021824 utils.py:1231] [97700] l2_params = 241.0556230791772
+I1205 00:00:06.599272 137274321021824 utils.py:1231] [97700] train/loss = 1.6259848773479462
+I1205 00:00:06.599398 137274321021824 utils.py:1231] [97700] l2_grads = 2.563596487045288
+I1205 00:00:06.599487 137274321021824 utils.py:1231] [97700] lr = 5.116528202729123e-05
+I1205 00:00:06.599568 137274321021824 utils.py:1231] [97700] uptime = 612595.961930363
+I1205 00:00:06.599641 137274321021824 utils.py:1231] [97700] examples_seen = 100044800.0
+I1205 00:00:06.599695 137274321021824 utils.py:1231] [97700] progress = 0.8676500626093443
+I1205 00:00:06.599751 137274321021824 utils.py:1231] [97700] epoch = 78.08880497234162
+I1205 00:00:06.599806 137274321021824 utils.py:1231] [97700] img/sec/core = 164.28957572650367
+I1205 00:00:06.599863 137274321021824 utils.py:1231] [97700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.1312273782997
+I1205 00:00:06.599923 137274321021824 utils.py:1231] [97700] core_hours = 170.1312273782997
+I1205 00:00:06.599990 137274321021824 train.py:125] NOTE: Steps:97700/112603 [86.8%]
+Walltime:7d2h9m (0s eval)
+ETA:1d1h57m
+Total train time:8d4h5m
+I1205 00:05:18.372337 137274321021824 utils.py:1231] [97750] l2_params = 241.02788301675915
+I1205 00:05:18.372595 137274321021824 utils.py:1231] [97750] train/loss = 3.78684601187706
+I1205 00:05:18.372700 137274321021824 utils.py:1231] [97750] l2_grads = 2.7925467491149902
+I1205 00:05:18.372778 137274321021824 utils.py:1231] [97750] lr = 5.0828487606757436e-05
+I1205 00:05:18.372841 137274321021824 utils.py:1231] [97750] uptime = 612907.735202533
+I1205 00:05:18.372913 137274321021824 utils.py:1231] [97750] examples_seen = 100096000.0
+I1205 00:05:18.372972 137274321021824 utils.py:1231] [97750] progress = 0.8680941005124198
+I1205 00:05:18.373028 137274321021824 utils.py:1231] [97750] epoch = 78.12876853681058
+I1205 00:05:18.373086 137274321021824 utils.py:1231] [97750] img/sec/core = 164.22190280661536
+I1205 00:05:18.373159 137274321021824 utils.py:1231] [97750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.2178310650136
+I1205 00:05:18.373218 137274321021824 utils.py:1231] [97750] core_hours = 170.2178310650136
+I1205 00:05:18.373295 137274321021824 train.py:125] NOTE: Steps:97750/112603 [86.8%]
+Walltime:7d2h15m (0s eval)
+ETA:1d1h51m
+Total train time:8d4h5m
+I1205 00:10:30.159758 137274321021824 utils.py:1231] [97800] l2_params = 240.99940178008444
+I1205 00:10:30.159989 137274321021824 utils.py:1231] [97800] train/loss = 1.582230731844902
+I1205 00:10:30.160104 137274321021824 utils.py:1231] [97800] l2_grads = 2.6704142093658447
+I1205 00:10:30.160182 137274321021824 utils.py:1231] [97800] lr = 5.049274595199214e-05
+I1205 00:10:30.160246 137274321021824 utils.py:1231] [97800] uptime = 613219.522602876
+I1205 00:10:30.160299 137274321021824 utils.py:1231] [97800] examples_seen = 100147200.0
+I1205 00:10:30.160350 137274321021824 utils.py:1231] [97800] progress = 0.8685381384154951
+I1205 00:10:30.160401 137274321021824 utils.py:1231] [97800] epoch = 78.16873210127953
+I1205 00:10:30.160454 137274321021824 utils.py:1231] [97800] img/sec/core = 164.214461340247
+I1205 00:10:30.160509 137274321021824 utils.py:1231] [97800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.30443867621997
+I1205 00:10:30.160561 137274321021824 utils.py:1231] [97800] core_hours = 170.30443867621997
+I1205 00:10:30.160623 137274321021824 train.py:125] NOTE: Steps:97800/112603 [86.9%]
+Walltime:7d2h20m (0s eval)
+ETA:1d1h46m
+Total train time:8d4h5m
+I1205 00:15:41.948343 137274321021824 utils.py:1231] [97850] l2_params = 240.97136093229565
+I1205 00:15:41.948618 137274321021824 utils.py:1231] [97850] train/loss = 1.485938161611557
+I1205 00:15:41.948754 137274321021824 utils.py:1231] [97850] l2_grads = 2.6461124420166016
+I1205 00:15:41.948857 137274321021824 utils.py:1231] [97850] lr = 5.0158057849904917e-05
+I1205 00:15:41.948925 137274321021824 utils.py:1231] [97850] uptime = 613531.31128602
+I1205 00:15:41.948989 137274321021824 utils.py:1231] [97850] examples_seen = 100198400.0
+I1205 00:15:41.949048 137274321021824 utils.py:1231] [97850] progress = 0.8689821763185706
+I1205 00:15:41.949105 137274321021824 utils.py:1231] [97850] epoch = 78.2086956657485
+I1205 00:15:41.949173 137274321021824 utils.py:1231] [97850] img/sec/core = 164.21378570802352
+I1205 00:15:41.949244 137274321021824 utils.py:1231] [97850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.39104664376
+I1205 00:15:41.949309 137274321021824 utils.py:1231] [97850] core_hours = 170.39104664376
+I1205 00:15:41.949381 137274321021824 train.py:125] NOTE: Steps:97850/112603 [86.9%]
+Walltime:7d2h25m (0s eval)
+ETA:1d1h41m
+Total train time:8d4h5m
+I1205 00:20:53.737188 137274321021824 utils.py:1231] [97900] l2_params = 240.94196765483156
+I1205 00:20:53.737400 137274321021824 utils.py:1231] [97900] train/loss = 3.7841348350048065
+I1205 00:20:53.737495 137274321021824 utils.py:1231] [97900] l2_grads = 2.638601064682007
+I1205 00:20:53.737564 137274321021824 utils.py:1231] [97900] lr = 4.982442408493595e-05
+I1205 00:20:53.737617 137274321021824 utils.py:1231] [97900] uptime = 613843.099978698
+I1205 00:20:53.737672 137274321021824 utils.py:1231] [97900] examples_seen = 100249600.0
+I1205 00:20:53.737721 137274321021824 utils.py:1231] [97900] progress = 0.869426214221646
+I1205 00:20:53.737788 137274321021824 utils.py:1231] [97900] epoch = 78.24865923021746
+I1205 00:20:53.737840 137274321021824 utils.py:1231] [97900] img/sec/core = 164.2137806866551
+I1205 00:20:53.737905 137274321021824 utils.py:1231] [97900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.47765461394832
+I1205 00:20:53.737957 137274321021824 utils.py:1231] [97900] core_hours = 170.47765461394832
+I1205 00:20:53.738018 137274321021824 train.py:125] NOTE: Steps:97900/112603 [86.9%]
+Walltime:7d2h30m (0s eval)
+ETA:1d1h36m
+Total train time:8d4h5m
+I1205 00:26:05.605240 137274321021824 utils.py:1231] [97950] l2_params = 240.91701244175272
+I1205 00:26:05.605446 137274321021824 utils.py:1231] [97950] train/loss = 1.744197055697441
+I1205 00:26:05.605542 137274321021824 utils.py:1231] [97950] l2_grads = 2.626321792602539
+I1205 00:26:05.605626 137274321021824 utils.py:1231] [97950] lr = 4.949184543905423e-05
+I1205 00:26:05.605691 137274321021824 utils.py:1231] [97950] uptime = 614154.968049266
+I1205 00:26:05.605776 137274321021824 utils.py:1231] [97950] examples_seen = 100300800.0
+I1205 00:26:05.605851 137274321021824 utils.py:1231] [97950] progress = 0.8698702521247214
+I1205 00:26:05.605920 137274321021824 utils.py:1231] [97950] epoch = 78.2886227946864
+I1205 00:26:05.605998 137274321021824 utils.py:1231] [97950] img/sec/core = 164.17198434818314
+I1205 00:26:05.606073 137274321021824 utils.py:1231] [97950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.56428463355056
+I1205 00:26:05.606140 137274321021824 utils.py:1231] [97950] core_hours = 170.56428463355056
+I1205 00:26:05.606238 137274321021824 train.py:125] NOTE: Steps:97950/112603 [87.0%]
+Walltime:7d2h35m (0s eval)
+ETA:1d1h30m
+Total train time:8d4h5m
+I1205 00:31:17.374115 137274321021824 utils.py:1231] [98000] l2_params = 240.88810345731156
+I1205 00:31:17.374393 137274321021824 utils.py:1231] [98000] train/loss = 1.5280642211437225
+I1205 00:31:17.374554 137274321021824 utils.py:1231] [98000] l2_grads = 2.6372745037078857
+I1205 00:31:17.374631 137274321021824 utils.py:1231] [98000] lr = 4.916032269175589e-05
+I1205 00:31:17.374692 137274321021824 utils.py:1231] [98000] uptime = 614466.737053355
+I1205 00:31:17.374754 137274321021824 utils.py:1231] [98000] examples_seen = 100352000.0
+I1205 00:31:17.374819 137274321021824 utils.py:1231] [98000] progress = 0.8703142900277968
+I1205 00:31:17.374879 137274321021824 utils.py:1231] [98000] epoch = 78.32858635915537
+I1205 00:31:17.374947 137274321021824 utils.py:1231] [98000] img/sec/core = 164.22415098517436
+I1205 00:31:17.375012 137274321021824 utils.py:1231] [98000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.65088713468637
+I1205 00:31:17.375068 137274321021824 utils.py:1231] [98000] core_hours = 170.65088713468637
+I1205 00:31:17.375137 137274321021824 train.py:125] NOTE: Steps:98000/112603 [87.0%]
+Walltime:7d2h41m (0s eval)
+ETA:1d1h25m
+Total train time:8d4h5m
+I1205 00:36:29.372884 137274321021824 utils.py:1231] [98050] l2_params = 240.8614330989524
+I1205 00:36:29.373100 137274321021824 utils.py:1231] [98050] train/loss = 1.5824202597141266
+I1205 00:36:29.373198 137274321021824 utils.py:1231] [98050] l2_grads = 2.5363540649414062
+I1205 00:36:29.373269 137274321021824 utils.py:1231] [98050] lr = 4.882985662006188e-05
+I1205 00:36:29.373333 137274321021824 utils.py:1231] [98050] uptime = 614778.735694953
+I1205 00:36:29.373403 137274321021824 utils.py:1231] [98050] examples_seen = 100403200.0
+I1205 00:36:29.373460 137274321021824 utils.py:1231] [98050] progress = 0.8707583279308722
+I1205 00:36:29.373517 137274321021824 utils.py:1231] [98050] epoch = 78.36854992362431
+I1205 00:36:29.373575 137274321021824 utils.py:1231] [98050] img/sec/core = 164.10327858402417
+I1205 00:36:29.373635 137274321021824 utils.py:1231] [98050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.73755342401915
+I1205 00:36:29.373687 137274321021824 utils.py:1231] [98050] core_hours = 170.73755342401915
+I1205 00:36:29.373755 137274321021824 train.py:125] NOTE: Steps:98050/112603 [87.1%]
+Walltime:7d2h46m (0s eval)
+ETA:1d1h20m
+Total train time:8d4h4m
+I1205 00:41:41.167041 137274321021824 utils.py:1231] [98100] l2_params = 240.83360093913396
+I1205 00:41:41.167244 137274321021824 utils.py:1231] [98100] train/loss = 1.4182530790567398
+I1205 00:41:41.167361 137274321021824 utils.py:1231] [98100] l2_grads = 2.5839545726776123
+I1205 00:41:41.167437 137274321021824 utils.py:1231] [98100] lr = 4.850044799851716e-05
+I1205 00:41:41.167498 137274321021824 utils.py:1231] [98100] uptime = 615090.52985993
+I1205 00:41:41.167559 137274321021824 utils.py:1231] [98100] examples_seen = 100454400.0
+I1205 00:41:41.167618 137274321021824 utils.py:1231] [98100] progress = 0.8712023658339476
+I1205 00:41:41.167676 137274321021824 utils.py:1231] [98100] epoch = 78.40851348809328
+I1205 00:41:41.167733 137274321021824 utils.py:1231] [98100] img/sec/core = 164.21089857076342
+I1205 00:41:41.167795 137274321021824 utils.py:1231] [98100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.82416291429055
+I1205 00:41:41.167853 137274321021824 utils.py:1231] [98100] core_hours = 170.82416291429055
+I1205 00:41:41.167927 137274321021824 train.py:125] NOTE: Steps:98100/112603 [87.1%]
+Walltime:7d2h51m (0s eval)
+ETA:1d1h15m
+Total train time:8d4h4m
+I1205 00:46:52.885750 137274321021824 utils.py:1231] [98150] l2_params = 240.80448379324818
+I1205 00:46:52.885972 137274321021824 utils.py:1231] [98150] train/loss = 2.3792938590049744
+I1205 00:46:52.886069 137274321021824 utils.py:1231] [98150] l2_grads = 2.569509744644165
+I1205 00:46:52.886140 137274321021824 utils.py:1231] [98150] lr = 4.817209759918763e-05
+I1205 00:46:52.886193 137274321021824 utils.py:1231] [98150] uptime = 615402.248555277
+I1205 00:46:52.886248 137274321021824 utils.py:1231] [98150] examples_seen = 100505600.0
+I1205 00:46:52.886298 137274321021824 utils.py:1231] [98150] progress = 0.871646403737023
+I1205 00:46:52.886349 137274321021824 utils.py:1231] [98150] epoch = 78.44847705256224
+I1205 00:46:52.886402 137274321021824 utils.py:1231] [98150] img/sec/core = 164.25065536416045
+I1205 00:46:52.886458 137274321021824 utils.py:1231] [98150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.91075144077584
+I1205 00:46:52.886510 137274321021824 utils.py:1231] [98150] core_hours = 170.91075144077584
+I1205 00:46:52.886572 137274321021824 train.py:125] NOTE: Steps:98150/112603 [87.2%]
+Walltime:7d2h56m (0s eval)
+ETA:1d1h10m
+Total train time:8d4h4m
+I1205 00:52:04.602396 137274321021824 utils.py:1231] [98200] l2_params = 240.776877951366
+I1205 00:52:04.602618 137274321021824 utils.py:1231] [98200] train/loss = 1.498205378651619
+I1205 00:52:04.602725 137274321021824 utils.py:1231] [98200] l2_grads = 2.5123989582061768
+I1205 00:52:04.602799 137274321021824 utils.py:1231] [98200] lr = 4.7844806191659334e-05
+I1205 00:52:04.602862 137274321021824 utils.py:1231] [98200] uptime = 615713.965223055
+I1205 00:52:04.602931 137274321021824 utils.py:1231] [98200] examples_seen = 100556800.0
+I1205 00:52:04.602995 137274321021824 utils.py:1231] [98200] progress = 0.8720904416400984
+I1205 00:52:04.603052 137274321021824 utils.py:1231] [98200] epoch = 78.48844061703119
+I1205 00:52:04.603111 137274321021824 utils.py:1231] [98200] img/sec/core = 164.25172373672356
+I1205 00:52:04.603172 137274321021824 utils.py:1231] [98200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 170.9973394040475
+I1205 00:52:04.603228 137274321021824 utils.py:1231] [98200] core_hours = 170.9973394040475
+I1205 00:52:04.603299 137274321021824 train.py:125] NOTE: Steps:98200/112603 [87.2%]
+Walltime:7d3h1m (0s eval)
+ETA:1d1h4m
+Total train time:8d4h4m
+I1205 00:57:16.377519 137274321021824 utils.py:1231] [98250] l2_params = 240.7494314778479
+I1205 00:57:16.377741 137274321021824 utils.py:1231] [98250] train/loss = 2.1024137139320374
+I1205 00:57:16.377855 137274321021824 utils.py:1231] [98250] l2_grads = 2.5246472358703613
+I1205 00:57:16.377966 137274321021824 utils.py:1231] [98250] lr = 4.751857454303605e-05
+I1205 00:57:16.378038 137274321021824 utils.py:1231] [98250] uptime = 616025.740398678
+I1205 00:57:16.378107 137274321021824 utils.py:1231] [98250] examples_seen = 100608000.0
+I1205 00:57:16.378179 137274321021824 utils.py:1231] [98250] progress = 0.8725344795431738
+I1205 00:57:16.378242 137274321021824 utils.py:1231] [98250] epoch = 78.52840418150015
+I1205 00:57:16.378306 137274321021824 utils.py:1231] [98250] img/sec/core = 164.2209001973558
+I1205 00:57:16.378375 137274321021824 utils.py:1231] [98250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.08394361949834
+I1205 00:57:16.378447 137274321021824 utils.py:1231] [98250] core_hours = 171.08394361949834
+I1205 00:57:16.378532 137274321021824 train.py:125] NOTE: Steps:98250/112603 [87.3%]
+Walltime:7d3h7m (0s eval)
+ETA:24h59m
+Total train time:8d4h4m
+I1205 01:02:28.151732 137274321021824 utils.py:1231] [98300] l2_params = 240.72418296023417
+I1205 01:02:28.151994 137274321021824 utils.py:1231] [98300] train/loss = 2.1125037223100662
+I1205 01:02:28.152124 137274321021824 utils.py:1231] [98300] l2_grads = 2.524362325668335
+I1205 01:02:28.152220 137274321021824 utils.py:1231] [98300] lr = 4.719340341793751e-05
+I1205 01:02:28.152285 137274321021824 utils.py:1231] [98300] uptime = 616337.514646626
+I1205 01:02:28.152369 137274321021824 utils.py:1231] [98300] examples_seen = 100659200.0
+I1205 01:02:28.152440 137274321021824 utils.py:1231] [98300] progress = 0.8729785174462492
+I1205 01:02:28.152499 137274321021824 utils.py:1231] [98300] epoch = 78.5683677459691
+I1205 01:02:28.152574 137274321021824 utils.py:1231] [98300] img/sec/core = 164.22138883178525
+I1205 01:02:28.152648 137274321021824 utils.py:1231] [98300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.17054757726163
+I1205 01:02:28.152710 137274321021824 utils.py:1231] [98300] core_hours = 171.17054757726163
+I1205 01:02:28.152780 137274321021824 train.py:125] NOTE: Steps:98300/112603 [87.3%]
+Walltime:7d3h12m (0s eval)
+ETA:24h54m
+Total train time:8d4h4m
+I1205 01:07:39.916862 137274321021824 utils.py:1231] [98350] l2_params = 240.69801440906744
+I1205 01:07:39.917112 137274321021824 utils.py:1231] [98350] train/loss = 1.5330588519573212
+I1205 01:07:39.917244 137274321021824 utils.py:1231] [98350] l2_grads = 2.5504539012908936
+I1205 01:07:39.917331 137274321021824 utils.py:1231] [98350] lr = 4.6869293578498535e-05
+I1205 01:07:39.917387 137274321021824 utils.py:1231] [98350] uptime = 616649.279748986
+I1205 01:07:39.917440 137274321021824 utils.py:1231] [98350] examples_seen = 100710400.0
+I1205 01:07:39.917490 137274321021824 utils.py:1231] [98350] progress = 0.8734225553493247
+I1205 01:07:39.917539 137274321021824 utils.py:1231] [98350] epoch = 78.60833131043806
+I1205 01:07:39.917590 137274321021824 utils.py:1231] [98350] img/sec/core = 164.22620624443923
+I1205 01:07:39.917647 137274321021824 utils.py:1231] [98350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.2571489945839
+I1205 01:07:39.917696 137274321021824 utils.py:1231] [98350] core_hours = 171.2571489945839
+I1205 01:07:39.917757 137274321021824 train.py:125] NOTE: Steps:98350/112603 [87.3%]
+Walltime:7d3h17m (0s eval)
+ETA:24h49m
+Total train time:8d4h4m
+I1205 01:12:51.693507 137274321021824 utils.py:1231] [98400] l2_params = 240.67356494631457
+I1205 01:12:51.693726 137274321021824 utils.py:1231] [98400] train/loss = 1.5579630434513092
+I1205 01:12:51.693865 137274321021824 utils.py:1231] [98400] l2_grads = 2.727341413497925
+I1205 01:12:51.693962 137274321021824 utils.py:1231] [98400] lr = 4.654624578436562e-05
+I1205 01:12:51.694037 137274321021824 utils.py:1231] [98400] uptime = 616961.0563945239
+I1205 01:12:51.694113 137274321021824 utils.py:1231] [98400] examples_seen = 100761600.0
+I1205 01:12:51.694184 137274321021824 utils.py:1231] [98400] progress = 0.8738665932524
+I1205 01:12:51.694255 137274321021824 utils.py:1231] [98400] epoch = 78.64829487490702
+I1205 01:12:51.694326 137274321021824 utils.py:1231] [98400] img/sec/core = 164.2201259547867
+I1205 01:12:51.694400 137274321021824 utils.py:1231] [98400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.34375361834444
+I1205 01:12:51.694463 137274321021824 utils.py:1231] [98400] core_hours = 171.34375361834444
+I1205 01:12:51.694535 137274321021824 train.py:125] NOTE: Steps:98400/112603 [87.4%]
+Walltime:7d3h22m (0s eval)
+ETA:24h43m
+Total train time:8d4h4m
+I1205 01:18:03.468730 137274321021824 utils.py:1231] [98450] l2_params = 240.64664411488025
+I1205 01:18:03.469007 137274321021824 utils.py:1231] [98450] train/loss = 3.201193869113922
+I1205 01:18:03.469143 137274321021824 utils.py:1231] [98450] l2_grads = 2.6028780937194824
+I1205 01:18:03.469232 137274321021824 utils.py:1231] [98450] lr = 4.6224260792696706e-05
+I1205 01:18:03.469289 137274321021824 utils.py:1231] [98450] uptime = 617272.831650005
+I1205 01:18:03.469365 137274321021824 utils.py:1231] [98450] examples_seen = 100812800.0
+I1205 01:18:03.469416 137274321021824 utils.py:1231] [98450] progress = 0.8743106311554755
+I1205 01:18:03.469468 137274321021824 utils.py:1231] [98450] epoch = 78.68825843937597
+I1205 01:18:03.469521 137274321021824 utils.py:1231] [98450] img/sec/core = 164.22085813387423
+I1205 01:18:03.469600 137274321021824 utils.py:1231] [98450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.43035785597803
+I1205 01:18:03.469659 137274321021824 utils.py:1231] [98450] core_hours = 171.43035785597803
+I1205 01:18:03.469721 137274321021824 train.py:125] NOTE: Steps:98450/112603 [87.4%]
+Walltime:7d3h27m (0s eval)
+ETA:24h38m
+Total train time:8d4h4m
+I1205 01:23:15.238396 137274321021824 utils.py:1231] [98500] l2_params = 240.61899134911837
+I1205 01:23:15.238631 137274321021824 utils.py:1231] [98500] train/loss = 2.645490199327469
+I1205 01:23:15.238746 137274321021824 utils.py:1231] [98500] l2_grads = 2.549408197402954
+I1205 01:23:15.238826 137274321021824 utils.py:1231] [98500] lr = 4.590333935815835e-05
+I1205 01:23:15.238901 137274321021824 utils.py:1231] [98500] uptime = 617584.6012579959
+I1205 01:23:15.238974 137274321021824 utils.py:1231] [98500] examples_seen = 100864000.0
+I1205 01:23:15.239039 137274321021824 utils.py:1231] [98500] progress = 0.8747546690585508
+I1205 01:23:15.239103 137274321021824 utils.py:1231] [98500] epoch = 78.72822200384493
+I1205 01:23:15.239170 137274321021824 utils.py:1231] [98500] img/sec/core = 164.2238328807254
+I1205 01:23:15.239241 137274321021824 utils.py:1231] [98500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.51696052486443
+I1205 01:23:15.239303 137274321021824 utils.py:1231] [98500] core_hours = 171.51696052486443
+I1205 01:23:15.239384 137274321021824 train.py:125] NOTE: Steps:98500/112603 [87.5%]
+Walltime:7d3h33m (0s eval)
+ETA:24h33m
+Total train time:8d4h4m
+I1205 01:28:27.027919 137274321021824 utils.py:1231] [98550] l2_params = 240.5938566477438
+I1205 01:28:27.028143 137274321021824 utils.py:1231] [98550] train/loss = 2.958561420440674
+I1205 01:28:27.028287 137274321021824 utils.py:1231] [98550] l2_grads = 2.52911114692688
+I1205 01:28:27.028397 137274321021824 utils.py:1231] [98550] lr = 4.5583482232924336e-05
+I1205 01:28:27.028481 137274321021824 utils.py:1231] [98550] uptime = 617896.390838857
+I1205 01:28:27.028573 137274321021824 utils.py:1231] [98550] examples_seen = 100915200.0
+I1205 01:28:27.028658 137274321021824 utils.py:1231] [98550] progress = 0.8751987069616263
+I1205 01:28:27.028747 137274321021824 utils.py:1231] [98550] epoch = 78.76818556831388
+I1205 01:28:27.028834 137274321021824 utils.py:1231] [98550] img/sec/core = 164.21331289709826
+I1205 01:28:27.028930 137274321021824 utils.py:1231] [98550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.6035687417703
+I1205 01:28:27.029021 137274321021824 utils.py:1231] [98550] core_hours = 171.6035687417703
+I1205 01:28:27.029129 137274321021824 train.py:125] NOTE: Steps:98550/112603 [87.5%]
+Walltime:7d3h38m (0s eval)
+ETA:24h28m
+Total train time:8d4h4m
+I1205 01:33:38.820443 137274321021824 utils.py:1231] [98600] l2_params = 240.56705217496952
+I1205 01:33:38.820706 137274321021824 utils.py:1231] [98600] train/loss = 1.928623616695404
+I1205 01:33:38.820838 137274321021824 utils.py:1231] [98600] l2_grads = 2.5639185905456543
+I1205 01:33:38.820937 137274321021824 utils.py:1231] [98600] lr = 4.526469016667435e-05
+I1205 01:33:38.821000 137274321021824 utils.py:1231] [98600] uptime = 618208.183361558
+I1205 01:33:38.821062 137274321021824 utils.py:1231] [98600] examples_seen = 100966400.0
+I1205 01:33:38.821120 137274321021824 utils.py:1231] [98600] progress = 0.8756427448647016
+I1205 01:33:38.821178 137274321021824 utils.py:1231] [98600] epoch = 78.80814913278284
+I1205 01:33:38.821250 137274321021824 utils.py:1231] [98600] img/sec/core = 164.21176350372068
+I1205 01:33:38.821318 137274321021824 utils.py:1231] [98600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.6901777758539
+I1205 01:33:38.821377 137274321021824 utils.py:1231] [98600] core_hours = 171.6901777758539
+I1205 01:33:38.821453 137274321021824 train.py:125] NOTE: Steps:98600/112603 [87.6%]
+Walltime:7d3h43m (0s eval)
+ETA:24h23m
+Total train time:8d4h4m
+I1205 01:38:50.436353 137274321021824 utils.py:1231] [98650] l2_params = 240.5399552082494
+I1205 01:38:50.436551 137274321021824 utils.py:1231] [98650] train/loss = 1.5734595507383347
+I1205 01:38:50.436669 137274321021824 utils.py:1231] [98650] l2_grads = 2.65651535987854
+I1205 01:38:50.436735 137274321021824 utils.py:1231] [98650] lr = 4.494696390659133e-05
+I1205 01:38:50.436786 137274321021824 utils.py:1231] [98650] uptime = 618519.799148524
+I1205 01:38:50.436840 137274321021824 utils.py:1231] [98650] examples_seen = 101017600.0
+I1205 01:38:50.436895 137274321021824 utils.py:1231] [98650] progress = 0.8760867827677771
+I1205 01:38:50.436945 137274321021824 utils.py:1231] [98650] epoch = 78.8481126972518
+I1205 01:38:50.436996 137274321021824 utils.py:1231] [98650] img/sec/core = 164.3048977027496
+I1205 01:38:50.437051 137274321021824 utils.py:1231] [98650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.77673771667776
+I1205 01:38:50.437099 137274321021824 utils.py:1231] [98650] core_hours = 171.77673771667776
+I1205 01:38:50.437163 137274321021824 train.py:125] NOTE: Steps:98650/112603 [87.6%]
+Walltime:7d3h48m (0s eval)
+ETA:24h17m
+Total train time:8d4h4m
+I1205 01:44:02.214933 137274321021824 utils.py:1231] [98700] l2_params = 240.51569504229732
+I1205 01:44:02.215161 137274321021824 utils.py:1231] [98700] train/loss = 1.390695184469223
+I1205 01:44:02.215283 137274321021824 utils.py:1231] [98700] l2_grads = 2.4617154598236084
+I1205 01:44:02.215354 137274321021824 utils.py:1231] [98700] lr = 4.463030419736049e-05
+I1205 01:44:02.215420 137274321021824 utils.py:1231] [98700] uptime = 618831.577774302
+I1205 01:44:02.215476 137274321021824 utils.py:1231] [98700] examples_seen = 101068800.0
+I1205 01:44:02.215526 137274321021824 utils.py:1231] [98700] progress = 0.8765308206708524
+I1205 01:44:02.215584 137274321021824 utils.py:1231] [98700] epoch = 78.88807626172076
+I1205 01:44:02.215635 137274321021824 utils.py:1231] [98700] img/sec/core = 164.21908292215718
+I1205 01:44:02.215692 137274321021824 utils.py:1231] [98700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.86334289050498
+I1205 01:44:02.215744 137274321021824 utils.py:1231] [98700] core_hours = 171.86334289050498
+I1205 01:44:02.215802 137274321021824 train.py:125] NOTE: Steps:98700/112603 [87.7%]
+Walltime:7d3h53m (0s eval)
+ETA:24h12m
+Total train time:8d4h4m
+I1205 01:49:14.004925 137274321021824 utils.py:1231] [98750] l2_params = 240.49081799413034
+I1205 01:49:14.005187 137274321021824 utils.py:1231] [98750] train/loss = 1.4551428109407425
+I1205 01:49:14.005327 137274321021824 utils.py:1231] [98750] l2_grads = 2.561697006225586
+I1205 01:49:14.005426 137274321021824 utils.py:1231] [98750] lr = 4.4314711781167084e-05
+I1205 01:49:14.005484 137274321021824 utils.py:1231] [98750] uptime = 619143.3678460179
+I1205 01:49:14.005537 137274321021824 utils.py:1231] [98750] examples_seen = 101120000.0
+I1205 01:49:14.005595 137274321021824 utils.py:1231] [98750] progress = 0.8769748585739279
+I1205 01:49:14.005642 137274321021824 utils.py:1231] [98750] epoch = 78.92803982618972
+I1205 01:49:14.005697 137274321021824 utils.py:1231] [98750] img/sec/core = 164.21305437411334
+I1205 01:49:14.005755 137274321021824 utils.py:1231] [98750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 171.94995124375944
+I1205 01:49:14.005806 137274321021824 utils.py:1231] [98750] core_hours = 171.94995124375944
+I1205 01:49:14.005867 137274321021824 train.py:125] NOTE: Steps:98750/112603 [87.7%]
+Walltime:7d3h59m (0s eval)
+ETA:24h7m
+Total train time:8d4h4m
+I1205 01:54:25.779242 137274321021824 utils.py:1231] [98800] l2_params = 240.46499525228487
+I1205 01:54:25.779450 137274321021824 utils.py:1231] [98800] train/loss = 1.580012857913971
+I1205 01:54:25.779543 137274321021824 utils.py:1231] [98800] l2_grads = 2.4758973121643066
+I1205 01:54:25.779621 137274321021824 utils.py:1231] [98800] lr = 4.4000187397694786e-05
+I1205 01:54:25.779680 137274321021824 utils.py:1231] [98800] uptime = 619455.142042102
+I1205 01:54:25.779739 137274321021824 utils.py:1231] [98800] examples_seen = 101171200.0
+I1205 01:54:25.779793 137274321021824 utils.py:1231] [98800] progress = 0.8774188964770033
+I1205 01:54:25.779847 137274321021824 utils.py:1231] [98800] epoch = 78.96800339065867
+I1205 01:54:25.779908 137274321021824 utils.py:1231] [98800] img/sec/core = 164.22141615018165
+I1205 01:54:25.779968 137274321021824 utils.py:1231] [98800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.03655518711608
+I1205 01:54:25.780018 137274321021824 utils.py:1231] [98800] core_hours = 172.03655518711608
+I1205 01:54:25.780076 137274321021824 train.py:125] NOTE: Steps:98800/112603 [87.7%]
+Walltime:7d4h4m (0s eval)
+ETA:24h2m
+Total train time:8d4h4m
+I1205 01:59:37.555850 137274321021824 utils.py:1231] [98850] l2_params = 240.4393393018808
+I1205 01:59:37.556192 137274321021824 utils.py:1231] [98850] train/loss = 1.395536795258522
+I1205 01:59:37.556388 137274321021824 utils.py:1231] [98850] l2_grads = 2.5161588191986084
+I1205 01:59:37.556481 137274321021824 utils.py:1231] [98850] lr = 4.368673178412445e-05
+I1205 01:59:37.556551 137274321021824 utils.py:1231] [98850] uptime = 619766.918911346
+I1205 01:59:37.556638 137274321021824 utils.py:1231] [98850] examples_seen = 101222400.0
+I1205 01:59:37.556705 137274321021824 utils.py:1231] [98850] progress = 0.8778629343800787
+I1205 01:59:37.556777 137274321021824 utils.py:1231] [98850] epoch = 79.00796695512763
+I1205 01:59:37.556843 137274321021824 utils.py:1231] [98850] img/sec/core = 164.2200081235683
+I1205 01:59:37.556917 137274321021824 utils.py:1231] [98850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.12315987301722
+I1205 01:59:37.556989 137274321021824 utils.py:1231] [98850] core_hours = 172.12315987301722
+I1205 01:59:37.557071 137274321021824 train.py:125] NOTE: Steps:98850/112603 [87.8%]
+Walltime:7d4h9m (0s eval)
+ETA:23h56m
+Total train time:8d4h4m
+I1205 02:04:49.307181 137274321021824 utils.py:1231] [98900] l2_params = 240.41462624993596
+I1205 02:04:49.307404 137274321021824 utils.py:1231] [98900] train/loss = 2.930893450975418
+I1205 02:04:49.307508 137274321021824 utils.py:1231] [98900] l2_grads = 2.502784490585327
+I1205 02:04:49.307585 137274321021824 utils.py:1231] [98900] lr = 4.3374345675131424e-05
+I1205 02:04:49.307647 137274321021824 utils.py:1231] [98900] uptime = 620078.670008323
+I1205 02:04:49.307708 137274321021824 utils.py:1231] [98900] examples_seen = 101273600.0
+I1205 02:04:49.307769 137274321021824 utils.py:1231] [98900] progress = 0.8783069722831541
+I1205 02:04:49.307828 137274321021824 utils.py:1231] [98900] epoch = 79.04793051959659
+I1205 02:04:49.307891 137274321021824 utils.py:1231] [98900] img/sec/core = 164.2335840883063
+I1205 02:04:49.307954 137274321021824 utils.py:1231] [98900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.2097573999553
+I1205 02:04:49.308009 137274321021824 utils.py:1231] [98900] core_hours = 172.2097573999553
+I1205 02:04:49.308072 137274321021824 train.py:125] NOTE: Steps:98900/112603 [87.8%]
+Walltime:7d4h14m (0s eval)
+ETA:23h51m
+Total train time:8d4h4m
+I1205 02:10:01.079198 137274321021824 utils.py:1231] [98950] l2_params = 240.393752487785
+I1205 02:10:01.079439 137274321021824 utils.py:1231] [98950] train/loss = 2.67720627784729
+I1205 02:10:01.079561 137274321021824 utils.py:1231] [98950] l2_grads = 2.5337891578674316
+I1205 02:10:01.079639 137274321021824 utils.py:1231] [98950] lr = 4.306302980288472e-05
+I1205 02:10:01.079692 137274321021824 utils.py:1231] [98950] uptime = 620390.4420544329
+I1205 02:10:01.079744 137274321021824 utils.py:1231] [98950] examples_seen = 101324800.0
+I1205 02:10:01.079791 137274321021824 utils.py:1231] [98950] progress = 0.8787510101862295
+I1205 02:10:01.079838 137274321021824 utils.py:1231] [98950] epoch = 79.08789408406554
+I1205 02:10:01.079907 137274321021824 utils.py:1231] [98950] img/sec/core = 164.22254861795042
+I1205 02:10:01.079969 137274321021824 utils.py:1231] [98950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.29636074609692
+I1205 02:10:01.080028 137274321021824 utils.py:1231] [98950] core_hours = 172.29636074609692
+I1205 02:10:01.080098 137274321021824 train.py:125] NOTE: Steps:98950/112603 [87.9%]
+Walltime:7d4h19m (0s eval)
+ETA:23h46m
+Total train time:8d4h4m
+I1205 02:15:12.859438 137274321021824 utils.py:1231] [99000] l2_params = 240.3683209180666
+I1205 02:15:12.859656 137274321021824 utils.py:1231] [99000] train/loss = 2.4759214520454407
+I1205 02:15:12.859770 137274321021824 utils.py:1231] [99000] l2_grads = 2.3896126747131348
+I1205 02:15:12.859848 137274321021824 utils.py:1231] [99000] lr = 4.2752784897044704e-05
+I1205 02:15:12.859936 137274321021824 utils.py:1231] [99000] uptime = 620702.2222932699
+I1205 02:15:12.860007 137274321021824 utils.py:1231] [99000] examples_seen = 101376000.0
+I1205 02:15:12.860067 137274321021824 utils.py:1231] [99000] progress = 0.8791950480893049
+I1205 02:15:12.860125 137274321021824 utils.py:1231] [99000] epoch = 79.1278576485345
+I1205 02:15:12.860184 137274321021824 utils.py:1231] [99000] img/sec/core = 164.21823330107512
+I1205 02:15:12.860249 137274321021824 utils.py:1231] [99000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.3829663679961
+I1205 02:15:12.860337 137274321021824 utils.py:1231] [99000] core_hours = 172.3829663679961
+I1205 02:15:12.860428 137274321021824 train.py:125] NOTE: Steps:99000/112603 [87.9%]
+Walltime:7d4h25m (0s eval)
+ETA:23h41m
+Total train time:8d4h4m
+I1205 02:20:24.998959 137274321021824 utils.py:1231] [99050] l2_params = 240.34325047749704
+I1205 02:20:24.999236 137274321021824 utils.py:1231] [99050] train/loss = 1.64731165766716
+I1205 02:20:24.999365 137274321021824 utils.py:1231] [99050] l2_grads = 2.8692681789398193
+I1205 02:20:24.999445 137274321021824 utils.py:1231] [99050] lr = 4.244361168476169e-05
+I1205 02:20:24.999505 137274321021824 utils.py:1231] [99050] uptime = 621014.361867082
+I1205 02:20:24.999563 137274321021824 utils.py:1231] [99050] examples_seen = 101427200.0
+I1205 02:20:24.999618 137274321021824 utils.py:1231] [99050] progress = 0.8796390859923803
+I1205 02:20:24.999669 137274321021824 utils.py:1231] [99050] epoch = 79.16782121300346
+I1205 02:20:24.999722 137274321021824 utils.py:1231] [99050] img/sec/core = 164.02918532474948
+I1205 02:20:24.999780 137274321021824 utils.py:1231] [99050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.4696718051661
+I1205 02:20:24.999836 137274321021824 utils.py:1231] [99050] core_hours = 172.4696718051661
+I1205 02:20:24.999906 137274321021824 train.py:125] NOTE: Steps:99050/112603 [88.0%]
+Walltime:7d4h30m (0s eval)
+ETA:23h35m
+Total train time:8d4h4m
+I1205 02:25:36.765748 137274321021824 utils.py:1231] [99100] l2_params = 240.3194300664275
+I1205 02:25:36.766009 137274321021824 utils.py:1231] [99100] train/loss = 1.7619285583496094
+I1205 02:25:36.766146 137274321021824 utils.py:1231] [99100] l2_grads = 2.71574330329895
+I1205 02:25:36.766242 137274321021824 utils.py:1231] [99100] lr = 4.213551089067431e-05
+I1205 02:25:36.766313 137274321021824 utils.py:1231] [99100] uptime = 621326.128669535
+I1205 02:25:36.766391 137274321021824 utils.py:1231] [99100] examples_seen = 101478400.0
+I1205 02:25:36.766455 137274321021824 utils.py:1231] [99100] progress = 0.8800831238954557
+I1205 02:25:36.766526 137274321021824 utils.py:1231] [99100] epoch = 79.20778477747241
+I1205 02:25:36.766584 137274321021824 utils.py:1231] [99100] img/sec/core = 164.2253107038372
+I1205 02:25:36.766668 137274321021824 utils.py:1231] [99100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.5562736947364
+I1205 02:25:36.766730 137274321021824 utils.py:1231] [99100] core_hours = 172.5562736947364
+I1205 02:25:36.766803 137274321021824 train.py:125] NOTE: Steps:99100/112603 [88.0%]
+Walltime:7d4h35m (0s eval)
+ETA:23h30m
+Total train time:8d4h4m
+I1205 02:30:48.546438 137274321021824 utils.py:1231] [99150] l2_params = 240.2953676234395
+I1205 02:30:48.546682 137274321021824 utils.py:1231] [99150] train/loss = 3.32333567738533
+I1205 02:30:48.546811 137274321021824 utils.py:1231] [99150] l2_grads = 2.5595226287841797
+I1205 02:30:48.546910 137274321021824 utils.py:1231] [99150] lr = 4.1828483236907324e-05
+I1205 02:30:48.546970 137274321021824 utils.py:1231] [99150] uptime = 621637.90933165
+I1205 02:30:48.547029 137274321021824 utils.py:1231] [99150] examples_seen = 101529600.0
+I1205 02:30:48.547083 137274321021824 utils.py:1231] [99150] progress = 0.8805271617985311
+I1205 02:30:48.547139 137274321021824 utils.py:1231] [99150] epoch = 79.24774834194137
+I1205 02:30:48.547193 137274321021824 utils.py:1231] [99150] img/sec/core = 164.2180103560134
+I1205 02:30:48.547254 137274321021824 utils.py:1231] [99150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.64287943421277
+I1205 02:30:48.547310 137274321021824 utils.py:1231] [99150] core_hours = 172.64287943421277
+I1205 02:30:48.547408 137274321021824 train.py:125] NOTE: Steps:99150/112603 [88.1%]
+Walltime:7d4h40m (0s eval)
+ETA:23h25m
+Total train time:8d4h4m
+I1205 02:35:59.190674 137274321021824 utils.py:1231] [99200] l2_params = 240.27197758642427
+I1205 02:35:59.190937 137274321021824 utils.py:1231] [99200] train/loss = 2.9759602546691895
+I1205 02:35:59.191079 137274321021824 utils.py:1231] [99200] l2_grads = 2.535890579223633
+I1205 02:35:59.191173 137274321021824 utils.py:1231] [99200] lr = 4.1522529443070805e-05
+I1205 02:35:59.191252 137274321021824 utils.py:1231] [99200] uptime = 621948.553613522
+I1205 02:35:59.191315 137274321021824 utils.py:1231] [99200] examples_seen = 101580800.0
+I1205 02:35:59.191374 137274321021824 utils.py:1231] [99200] progress = 0.8809711997016065
+I1205 02:35:59.191433 137274321021824 utils.py:1231] [99200] epoch = 79.28771190641032
+I1205 02:35:59.191494 137274321021824 utils.py:1231] [99200] img/sec/core = 164.81874281240556
+I1205 02:35:59.191560 137274321021824 utils.py:1231] [99200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.72916951251054
+I1205 02:35:59.191623 137274321021824 utils.py:1231] [99200] core_hours = 172.72916951251054
+I1205 02:35:59.191693 137274321021824 train.py:125] NOTE: Steps:99200/112603 [88.1%]
+Walltime:7d4h45m (0s eval)
+ETA:23h20m
+Total train time:8d4h4m
+I1205 02:41:10.970006 137274321021824 utils.py:1231] [99250] l2_params = 240.2481707268599
+I1205 02:41:10.970245 137274321021824 utils.py:1231] [99250] train/loss = 3.222594439983368
+I1205 02:41:10.970342 137274321021824 utils.py:1231] [99250] l2_grads = 2.6025094985961914
+I1205 02:41:10.970421 137274321021824 utils.py:1231] [99250] lr = 4.121765022625736e-05
+I1205 02:41:10.970487 137274321021824 utils.py:1231] [99250] uptime = 622260.3328482549
+I1205 02:41:10.970539 137274321021824 utils.py:1231] [99250] examples_seen = 101632000.0
+I1205 02:41:10.970587 137274321021824 utils.py:1231] [99250] progress = 0.881415237604682
+I1205 02:41:10.970633 137274321021824 utils.py:1231] [99250] epoch = 79.32767547087929
+I1205 02:41:10.970683 137274321021824 utils.py:1231] [99250] img/sec/core = 164.21876217590903
+I1205 02:41:10.970739 137274321021824 utils.py:1231] [99250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.81577485549192
+I1205 02:41:10.970787 137274321021824 utils.py:1231] [99250] core_hours = 172.81577485549192
+I1205 02:41:10.970846 137274321021824 train.py:125] NOTE: Steps:99250/112603 [88.1%]
+Walltime:7d4h51m (0s eval)
+ETA:23h15m
+Total train time:8d4h4m
+I1205 02:46:22.743469 137274321021824 utils.py:1231] [99300] l2_params = 240.22386051137087
+I1205 02:46:22.743696 137274321021824 utils.py:1231] [99300] train/loss = 3.517188251018524
+I1205 02:46:22.743798 137274321021824 utils.py:1231] [99300] l2_grads = 2.6640853881835938
+I1205 02:46:22.743868 137274321021824 utils.py:1231] [99300] lr = 4.0913846301041304e-05
+I1205 02:46:22.743930 137274321021824 utils.py:1231] [99300] uptime = 622572.106291092
+I1205 02:46:22.743987 137274321021824 utils.py:1231] [99300] examples_seen = 101683200.0
+I1205 02:46:22.744040 137274321021824 utils.py:1231] [99300] progress = 0.8818592755077573
+I1205 02:46:22.744092 137274321021824 utils.py:1231] [99300] epoch = 79.36763903534825
+I1205 02:46:22.744149 137274321021824 utils.py:1231] [99300] img/sec/core = 164.22181291033897
+I1205 02:46:22.744208 137274321021824 utils.py:1231] [99300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.90237858961333
+I1205 02:46:22.744262 137274321021824 utils.py:1231] [99300] core_hours = 172.90237858961333
+I1205 02:46:22.744326 137274321021824 train.py:125] NOTE: Steps:99300/112603 [88.2%]
+Walltime:7d4h56m (0s eval)
+ETA:23h9m
+Total train time:8d4h4m
+I1205 02:51:34.520059 137274321021824 utils.py:1231] [99350] l2_params = 240.2002225188204
+I1205 02:51:34.520317 137274321021824 utils.py:1231] [99350] train/loss = 1.3951845467090607
+I1205 02:51:34.520443 137274321021824 utils.py:1231] [99350] l2_grads = 2.627734422683716
+I1205 02:51:34.520527 137274321021824 utils.py:1231] [99350] lr = 4.0611118379476755e-05
+I1205 02:51:34.520596 137274321021824 utils.py:1231] [99350] uptime = 622883.882957733
+I1205 02:51:34.520651 137274321021824 utils.py:1231] [99350] examples_seen = 101734400.0
+I1205 02:51:34.520701 137274321021824 utils.py:1231] [99350] progress = 0.8823033134108328
+I1205 02:51:34.520750 137274321021824 utils.py:1231] [99350] epoch = 79.4076025998172
+I1205 02:51:34.520802 137274321021824 utils.py:1231] [99350] img/sec/core = 164.22011483930132
+I1205 02:51:34.520858 137274321021824 utils.py:1231] [99350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 172.98898321923582
+I1205 02:51:34.520915 137274321021824 utils.py:1231] [99350] core_hours = 172.98898321923582
+I1205 02:51:34.520975 137274321021824 train.py:125] NOTE: Steps:99350/112603 [88.2%]
+Walltime:7d5h1m (0s eval)
+ETA:23h4m
+Total train time:8d4h4m
+I1205 02:56:46.293419 137274321021824 utils.py:1231] [99400] l2_params = 240.1760053982433
+I1205 02:56:46.293665 137274321021824 utils.py:1231] [99400] train/loss = 1.5148408561944962
+I1205 02:56:46.293766 137274321021824 utils.py:1231] [99400] l2_grads = 2.7632083892822266
+I1205 02:56:46.293906 137274321021824 utils.py:1231] [99400] lr = 4.030946717109555e-05
+I1205 02:56:46.294006 137274321021824 utils.py:1231] [99400] uptime = 623195.656361531
+I1205 02:56:46.294095 137274321021824 utils.py:1231] [99400] examples_seen = 101785600.0
+I1205 02:56:46.294177 137274321021824 utils.py:1231] [99400] progress = 0.8827473513139081
+I1205 02:56:46.294258 137274321021824 utils.py:1231] [99400] epoch = 79.44756616428616
+I1205 02:56:46.294342 137274321021824 utils.py:1231] [99400] img/sec/core = 164.22183347356105
+I1205 02:56:46.294420 137274321021824 utils.py:1231] [99400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.07558694251304
+I1205 02:56:46.294477 137274321021824 utils.py:1231] [99400] core_hours = 173.07558694251304
+I1205 02:56:46.294541 137274321021824 train.py:125] NOTE: Steps:99400/112603 [88.3%]
+Walltime:7d5h6m (0s eval)
+ETA:22h59m
+Total train time:8d4h4m
+I1205 03:01:58.069572 137274321021824 utils.py:1231] [99450] l2_params = 240.1537692752847
+I1205 03:01:58.069785 137274321021824 utils.py:1231] [99450] train/loss = 1.6145429909229279
+I1205 03:01:58.069890 137274321021824 utils.py:1231] [99450] l2_grads = 2.9081532955169678
+I1205 03:01:58.069987 137274321021824 utils.py:1231] [99450] lr = 4.000889338290647e-05
+I1205 03:01:58.070053 137274321021824 utils.py:1231] [99450] uptime = 623507.432414275
+I1205 03:01:58.070129 137274321021824 utils.py:1231] [99450] examples_seen = 101836800.0
+I1205 03:01:58.070198 137274321021824 utils.py:1231] [99450] progress = 0.8831913892169836
+I1205 03:01:58.070276 137274321021824 utils.py:1231] [99450] epoch = 79.4875297287551
+I1205 03:01:58.070347 137274321021824 utils.py:1231] [99450] img/sec/core = 164.2204381939448
+I1205 03:01:58.070441 137274321021824 utils.py:1231] [99450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.16219140160862
+I1205 03:01:58.070529 137274321021824 utils.py:1231] [99450] core_hours = 173.16219140160862
+I1205 03:01:58.070617 137274321021824 train.py:125] NOTE: Steps:99450/112603 [88.3%]
+Walltime:7d5h11m (0s eval)
+ETA:22h54m
+Total train time:8d4h4m
+I1205 03:07:09.835477 137274321021824 utils.py:1231] [99500] l2_params = 240.1306213906678
+I1205 03:07:09.835670 137274321021824 utils.py:1231] [99500] train/loss = 2.487773150205612
+I1205 03:07:09.835768 137274321021824 utils.py:1231] [99500] l2_grads = 2.51572585105896
+I1205 03:07:09.835836 137274321021824 utils.py:1231] [99500] lr = 3.970939771939241e-05
+I1205 03:07:09.835895 137274321021824 utils.py:1231] [99500] uptime = 623819.1982503859
+I1205 03:07:09.835950 137274321021824 utils.py:1231] [99500] examples_seen = 101888000.0
+I1205 03:07:09.836004 137274321021824 utils.py:1231] [99500] progress = 0.8836354271200589
+I1205 03:07:09.836056 137274321021824 utils.py:1231] [99500] epoch = 79.52749329322407
+I1205 03:07:09.836107 137274321021824 utils.py:1231] [99500] img/sec/core = 164.2258197328073
+I1205 03:07:09.836176 137274321021824 utils.py:1231] [99500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.24879302275056
+I1205 03:07:09.836230 137274321021824 utils.py:1231] [99500] core_hours = 173.24879302275056
+I1205 03:07:09.836294 137274321021824 train.py:125] NOTE: Steps:99500/112603 [88.4%]
+Walltime:7d5h16m (0s eval)
+ETA:22h48m
+Total train time:8d4h4m
+I1205 03:12:21.602023 137274321021824 utils.py:1231] [99550] l2_params = 240.10893534438065
+I1205 03:12:21.602227 137274321021824 utils.py:1231] [99550] train/loss = 2.0138094276189804
+I1205 03:12:21.602321 137274321021824 utils.py:1231] [99550] l2_grads = 2.4817023277282715
+I1205 03:12:21.602382 137274321021824 utils.py:1231] [99550] lr = 3.941098088251004e-05
+I1205 03:12:21.602432 137274321021824 utils.py:1231] [99550] uptime = 624130.96479451
+I1205 03:12:21.602489 137274321021824 utils.py:1231] [99550] examples_seen = 101939200.0
+I1205 03:12:21.602540 137274321021824 utils.py:1231] [99550] progress = 0.8840794650231344
+I1205 03:12:21.602594 137274321021824 utils.py:1231] [99550] epoch = 79.56745685769303
+I1205 03:12:21.602651 137274321021824 utils.py:1231] [99550] img/sec/core = 164.22544678051705
+I1205 03:12:21.602706 137274321021824 utils.py:1231] [99550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.3353948405628
+I1205 03:12:21.602759 137274321021824 utils.py:1231] [99550] core_hours = 173.3353948405628
+I1205 03:12:21.602818 137274321021824 train.py:125] NOTE: Steps:99550/112603 [88.4%]
+Walltime:7d5h22m (0s eval)
+ETA:22h43m
+Total train time:8d4h4m
+I1205 03:17:30.689288 137274321021824 utils.py:1231] [99600] l2_params = 240.0861025555173
+I1205 03:17:30.689551 137274321021824 utils.py:1231] [99600] train/loss = 3.6022534668445587
+I1205 03:17:30.689672 137274321021824 utils.py:1231] [99600] l2_grads = 2.6588168144226074
+I1205 03:17:30.689755 137274321021824 utils.py:1231] [99600] lr = 3.911364357168688e-05
+I1205 03:17:30.689814 137274321021824 utils.py:1231] [99600] uptime = 624440.052176173
+I1205 03:17:30.689871 137274321021824 utils.py:1231] [99600] examples_seen = 101990400.0
+I1205 03:17:30.689931 137274321021824 utils.py:1231] [99600] progress = 0.8845235029262097
+I1205 03:17:30.689982 137274321021824 utils.py:1231] [99600] epoch = 79.60742042216198
+I1205 03:17:30.690033 137274321021824 utils.py:1231] [99600] img/sec/core = 165.64894925357189
+I1205 03:17:30.690126 137274321021824 utils.py:1231] [99600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.42125244658027
+I1205 03:17:30.690179 137274321021824 utils.py:1231] [99600] core_hours = 173.42125244658027
+I1205 03:17:30.690239 137274321021824 train.py:125] NOTE: Steps:99600/112603 [88.5%]
+Walltime:7d5h27m (0s eval)
+ETA:22h38m
+Total train time:8d4h3m
+I1205 03:22:42.458744 137274321021824 utils.py:1231] [99650] l2_params = 240.06405278594787
+I1205 03:22:42.458991 137274321021824 utils.py:1231] [99650] train/loss = 1.7522230446338654
+I1205 03:22:42.459122 137274321021824 utils.py:1231] [99650] l2_grads = 2.5889394283294678
+I1205 03:22:42.459213 137274321021824 utils.py:1231] [99650] lr = 3.8817386483820416e-05
+I1205 03:22:42.459291 137274321021824 utils.py:1231] [99650] uptime = 624751.821652151
+I1205 03:22:42.459364 137274321021824 utils.py:1231] [99650] examples_seen = 102041600.0
+I1205 03:22:42.459433 137274321021824 utils.py:1231] [99650] progress = 0.8849675408292852
+I1205 03:22:42.459494 137274321021824 utils.py:1231] [99650] epoch = 79.64738398663094
+I1205 03:22:42.459565 137274321021824 utils.py:1231] [99650] img/sec/core = 164.22390241825826
+I1205 03:22:42.459632 137274321021824 utils.py:1231] [99650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.50785507879635
+I1205 03:22:42.459700 137274321021824 utils.py:1231] [99650] core_hours = 173.50785507879635
+I1205 03:22:42.459782 137274321021824 train.py:125] NOTE: Steps:99650/112603 [88.5%]
+Walltime:7d5h32m (0s eval)
+ETA:22h33m
+Total train time:8d4h3m
+I1205 03:27:52.075212 137274321021824 utils.py:1231] [99700] l2_params = 240.040829286087
+I1205 03:27:52.075481 137274321021824 utils.py:1231] [99700] train/loss = 1.4852754324674606
+I1205 03:27:52.075601 137274321021824 utils.py:1231] [99700] l2_grads = 2.8043017387390137
+I1205 03:27:52.075690 137274321021824 utils.py:1231] [99700] lr = 3.8522210313276785e-05
+I1205 03:27:52.075753 137274321021824 utils.py:1231] [99700] uptime = 625061.4381142859
+I1205 03:27:52.075812 137274321021824 utils.py:1231] [99700] examples_seen = 102092800.0
+I1205 03:27:52.075892 137274321021824 utils.py:1231] [99700] progress = 0.8854115787323606
+I1205 03:27:52.075952 137274321021824 utils.py:1231] [99700] epoch = 79.68734755109989
+I1205 03:27:52.076008 137274321021824 utils.py:1231] [99700] img/sec/core = 165.3658841230394
+I1205 03:27:52.076071 137274321021824 utils.py:1231] [99700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.59385965161164
+I1205 03:27:52.076127 137274321021824 utils.py:1231] [99700] core_hours = 173.59385965161164
+I1205 03:27:52.076191 137274321021824 train.py:125] NOTE: Steps:99700/112603 [88.5%]
+Walltime:7d5h37m (0s eval)
+ETA:22h27m
+Total train time:8d4h3m
+I1205 03:33:03.854621 137274321021824 utils.py:1231] [99750] l2_params = 240.018300801638
+I1205 03:33:03.854915 137274321021824 utils.py:1231] [99750] train/loss = 1.4709405899047852
+I1205 03:33:03.855131 137274321021824 utils.py:1231] [99750] l2_grads = 2.6433300971984863
+I1205 03:33:03.855221 137274321021824 utils.py:1231] [99750] lr = 3.822811575188805e-05
+I1205 03:33:03.855293 137274321021824 utils.py:1231] [99750] uptime = 625373.217652219
+I1205 03:33:03.855365 137274321021824 utils.py:1231] [99750] examples_seen = 102144000.0
+I1205 03:33:03.855432 137274321021824 utils.py:1231] [99750] progress = 0.885855616635436
+I1205 03:33:03.855497 137274321021824 utils.py:1231] [99750] epoch = 79.72731111556885
+I1205 03:33:03.855559 137274321021824 utils.py:1231] [99750] img/sec/core = 164.21860247606173
+I1205 03:33:03.855624 137274321021824 utils.py:1231] [99750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.68046507881527
+I1205 03:33:03.855683 137274321021824 utils.py:1231] [99750] core_hours = 173.68046507881527
+I1205 03:33:03.855753 137274321021824 train.py:125] NOTE: Steps:99750/112603 [88.6%]
+Walltime:7d5h42m (0s eval)
+ETA:22h22m
+Total train time:8d4h3m
+I1205 03:38:15.623914 137274321021824 utils.py:1231] [99800] l2_params = 239.99634420570922
+I1205 03:38:15.624176 137274321021824 utils.py:1231] [99800] train/loss = 1.7672650068998337
+I1205 03:38:15.624324 137274321021824 utils.py:1231] [99800] l2_grads = 2.687685251235962
+I1205 03:38:15.624418 137274321021824 utils.py:1231] [99800] lr = 3.7935103488951534e-05
+I1205 03:38:15.624489 137274321021824 utils.py:1231] [99800] uptime = 625684.986849972
+I1205 03:38:15.624567 137274321021824 utils.py:1231] [99800] examples_seen = 102195200.0
+I1205 03:38:15.624624 137274321021824 utils.py:1231] [99800] progress = 0.8862996545385114
+I1205 03:38:15.624679 137274321021824 utils.py:1231] [99800] epoch = 79.76727468003781
+I1205 03:38:15.624735 137274321021824 utils.py:1231] [99800] img/sec/core = 164.2240489727769
+I1205 03:38:15.624813 137274321021824 utils.py:1231] [99800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.76706763374668
+I1205 03:38:15.624879 137274321021824 utils.py:1231] [99800] core_hours = 173.76706763374668
+I1205 03:38:15.624952 137274321021824 train.py:125] NOTE: Steps:99800/112603 [88.6%]
+Walltime:7d5h48m (0s eval)
+ETA:22h17m
+Total train time:8d4h3m
+I1205 03:43:26.150534 137274321021824 utils.py:1231] [99850] l2_params = 239.9748867408703
+I1205 03:43:26.150953 137274321021824 utils.py:1231] [99850] train/loss = 1.5328982025384903
+I1205 03:43:26.151208 137274321021824 utils.py:1231] [99850] l2_grads = 2.688520669937134
+I1205 03:43:26.151355 137274321021824 utils.py:1231] [99850] lr = 3.764317421122778e-05
+I1205 03:43:26.151457 137274321021824 utils.py:1231] [99850] uptime = 625995.513810555
+I1205 03:43:26.151588 137274321021824 utils.py:1231] [99850] examples_seen = 102246400.0
+I1205 03:43:26.151677 137274321021824 utils.py:1231] [99850] progress = 0.8867436924415868
+I1205 03:43:26.151758 137274321021824 utils.py:1231] [99850] epoch = 79.80723824450676
+I1205 03:43:26.151837 137274321021824 utils.py:1231] [99850] img/sec/core = 164.88101356444085
+I1205 03:43:26.151933 137274321021824 utils.py:1231] [99850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.8533251227975
+I1205 03:43:26.152003 137274321021824 utils.py:1231] [99850] core_hours = 173.8533251227975
+I1205 03:43:26.152083 137274321021824 train.py:125] NOTE: Steps:99850/112603 [88.7%]
+Walltime:7d5h53m (0s eval)
+ETA:22h12m
+Total train time:8d4h3m
+I1205 03:48:37.917189 137274321021824 utils.py:1231] [99900] l2_params = 239.95316347299615
+I1205 03:48:37.917439 137274321021824 utils.py:1231] [99900] train/loss = 2.327133923768997
+I1205 03:48:37.917558 137274321021824 utils.py:1231] [99900] l2_grads = 2.5471601486206055
+I1205 03:48:37.917661 137274321021824 utils.py:1231] [99900] lr = 3.7352328602938904e-05
+I1205 03:48:37.917743 137274321021824 utils.py:1231] [99900] uptime = 626307.280096902
+I1205 03:48:37.917840 137274321021824 utils.py:1231] [99900] examples_seen = 102297600.0
+I1205 03:48:37.917929 137274321021824 utils.py:1231] [99900] progress = 0.8871877303446623
+I1205 03:48:37.918003 137274321021824 utils.py:1231] [99900] epoch = 79.84720180897573
+I1205 03:48:37.918095 137274321021824 utils.py:1231] [99900] img/sec/core = 164.22558256669112
+I1205 03:48:37.918195 137274321021824 utils.py:1231] [99900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 173.939926869005
+I1205 03:48:37.918269 137274321021824 utils.py:1231] [99900] core_hours = 173.939926869005
+I1205 03:48:37.918364 137274321021824 train.py:125] NOTE: Steps:99900/112603 [88.7%]
+Walltime:7d5h58m (0s eval)
+ETA:22h7m
+Total train time:8d4h3m
+I1205 03:53:49.689227 137274321021824 utils.py:1231] [99950] l2_params = 239.9308860300217
+I1205 03:53:49.689504 137274321021824 utils.py:1231] [99950] train/loss = 1.5480139702558517
+I1205 03:53:49.689694 137274321021824 utils.py:1231] [99950] l2_grads = 2.665008544921875
+I1205 03:53:49.689810 137274321021824 utils.py:1231] [99950] lr = 3.706256734576766e-05
+I1205 03:53:49.689895 137274321021824 utils.py:1231] [99950] uptime = 626619.05224895
+I1205 03:53:49.690002 137274321021824 utils.py:1231] [99950] examples_seen = 102348800.0
+I1205 03:53:49.690085 137274321021824 utils.py:1231] [99950] progress = 0.8876317682477376
+I1205 03:53:49.690170 137274321021824 utils.py:1231] [99950] epoch = 79.88716537344467
+I1205 03:53:49.690246 137274321021824 utils.py:1231] [99950] img/sec/core = 164.22249281624403
+I1205 03:53:49.690330 137274321021824 utils.py:1231] [99950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.02653024457388
+I1205 03:53:49.690411 137274321021824 utils.py:1231] [99950] core_hours = 174.02653024457388
+I1205 03:53:49.690499 137274321021824 train.py:125] NOTE: Steps:99950/112603 [88.8%]
+Walltime:7d6h3m (0s eval)
+ETA:22h1m
+Total train time:8d4h3m
+I1205 03:58:59.088798 137274321021824 utils.py:1231] [100000] l2_params = 239.908734014073
+I1205 03:58:59.089138 137274321021824 utils.py:1231] [100000] train/loss = 2.101550877094269
+I1205 03:58:59.089382 137274321021824 utils.py:1231] [100000] l2_grads = 2.5637810230255127
+I1205 03:58:59.089464 137274321021824 utils.py:1231] [100000] lr = 3.67738911188545e-05
+I1205 03:58:59.089525 137274321021824 utils.py:1231] [100000] uptime = 626928.451886925
+I1205 03:58:59.089587 137274321021824 utils.py:1231] [100000] examples_seen = 102400000.0
+I1205 03:58:59.089645 137274321021824 utils.py:1231] [100000] progress = 0.888075806150813
+I1205 03:58:59.089703 137274321021824 utils.py:1231] [100000] epoch = 79.92712893791364
+I1205 03:58:59.089763 137274321021824 utils.py:1231] [100000] img/sec/core = 165.48177087437907
+I1205 03:58:59.089826 137274321021824 utils.py:1231] [100000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.11247458845583
+I1205 03:58:59.089897 137274321021824 utils.py:1231] [100000] core_hours = 174.11247458845583
+I1205 03:58:59.089976 137274321021824 train.py:125] NOTE: Steps:100000/112603 [88.8%]
+Walltime:7d6h8m (0s eval)
+ETA:21h56m
+Total train time:8d4h3m
+I1205 03:58:59.455220 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:100000/112603 [88.8%]
+Walltime:7d6h8m (0s eval)
+ETA:21h56m
+Total train time:8d4h3m
+I1205 04:00:32.176896 137274321021824 utils.py:1231] [100000] val/acc@1 = 0.7581513073979592
+I1205 04:00:32.177133 137274321021824 utils.py:1231] [100000] val/loss = 0.9489771287356105
+I1205 04:00:32.177280 137274321021824 utils.py:1231] [100000] z/secs/eval/val = 92.72181525395717
+I1205 04:00:32.177341 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 92.72181525395717
+I1205 04:05:42.552453 137274321021824 utils.py:1231] [100050] l2_params = 239.88776218343725
+I1205 04:05:42.552688 137274321021824 utils.py:1231] [100050] train/loss = 1.4942973107099533
+I1205 04:05:42.552838 137274321021824 utils.py:1231] [100050] l2_grads = 2.767738103866577
+I1205 04:05:42.552968 137274321021824 utils.py:1231] [100050] lr = 3.648630059879748e-05
+I1205 04:05:42.553073 137274321021824 utils.py:1231] [100050] uptime = 627331.915430011
+I1205 04:05:42.553181 137274321021824 utils.py:1231] [100050] examples_seen = 102451200.0
+I1205 04:05:42.553285 137274321021824 utils.py:1231] [100050] progress = 0.8885198440538884
+I1205 04:05:42.553370 137274321021824 utils.py:1231] [100050] epoch = 79.9670925023826
+I1205 04:05:42.553445 137274321021824 utils.py:1231] [100050] img/sec/core = 126.90118073218609
+I1205 04:05:42.553555 137274321021824 utils.py:1231] [100050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.2245477948686
+I1205 04:05:42.553608 137274321021824 utils.py:1231] [100050] core_hours = 174.2245477948686
+I1205 04:05:42.553682 137274321021824 train.py:125] NOTE: Steps:100050/112603 [88.9%]
+Walltime:7d6h15m (0s eval)
+ETA:21h51m
+Total train time:8d4h5m
+I1205 04:10:54.326077 137274321021824 utils.py:1231] [100100] l2_params = 239.86632900806893
+I1205 04:10:54.326340 137274321021824 utils.py:1231] [100100] train/loss = 2.0640313029289246
+I1205 04:10:54.326469 137274321021824 utils.py:1231] [100100] l2_grads = 2.4072065353393555
+I1205 04:10:54.326549 137274321021824 utils.py:1231] [100100] lr = 3.619979645964959e-05
+I1205 04:10:54.326637 137274321021824 utils.py:1231] [100100] uptime = 627643.68899385
+I1205 04:10:54.326708 137274321021824 utils.py:1231] [100100] examples_seen = 102502400.0
+I1205 04:10:54.326759 137274321021824 utils.py:1231] [100100] progress = 0.8889638819569639
+I1205 04:10:54.326818 137274321021824 utils.py:1231] [100100] epoch = 80.00705606685155
+I1205 04:10:54.326869 137274321021824 utils.py:1231] [100100] img/sec/core = 164.22174917445807
+I1205 04:10:54.326935 137274321021824 utils.py:1231] [100100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.31115156260165
+I1205 04:10:54.326984 137274321021824 utils.py:1231] [100100] core_hours = 174.31115156260165
+I1205 04:10:54.327045 137274321021824 train.py:125] NOTE: Steps:100100/112603 [88.9%]
+Walltime:7d6h20m (0s eval)
+ETA:21h46m
+Total train time:8d4h5m
+I1205 04:16:04.257896 137274321021824 utils.py:1231] [100150] l2_params = 239.84485235737512
+I1205 04:16:04.258190 137274321021824 utils.py:1231] [100150] train/loss = 3.3986693918704987
+I1205 04:16:04.258363 137274321021824 utils.py:1231] [100150] l2_grads = 2.6418519020080566
+I1205 04:16:04.258451 137274321021824 utils.py:1231] [100150] lr = 3.591437937291769e-05
+I1205 04:16:04.258513 137274321021824 utils.py:1231] [100150] uptime = 627953.620871966
+I1205 04:16:04.258576 137274321021824 utils.py:1231] [100150] examples_seen = 102553600.0
+I1205 04:16:04.258625 137274321021824 utils.py:1231] [100150] progress = 0.8894079198600392
+I1205 04:16:04.258684 137274321021824 utils.py:1231] [100150] epoch = 80.04701963132051
+I1205 04:16:04.258734 137274321021824 utils.py:1231] [100150] img/sec/core = 165.19759216520504
+I1205 04:16:04.258788 137274321021824 utils.py:1231] [100150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.39724375096722
+I1205 04:16:04.258839 137274321021824 utils.py:1231] [100150] core_hours = 174.39724375096722
+I1205 04:16:04.258924 137274321021824 train.py:125] NOTE: Steps:100150/112603 [88.9%]
+Walltime:7d6h25m (0s eval)
+ETA:21h41m
+Total train time:8d4h5m
+I1205 04:21:16.050436 137274321021824 utils.py:1231] [100200] l2_params = 239.8248832971167
+I1205 04:21:16.050704 137274321021824 utils.py:1231] [100200] train/loss = 3.1072280406951904
+I1205 04:21:16.050831 137274321021824 utils.py:1231] [100200] l2_grads = 2.5966176986694336
+I1205 04:21:16.050924 137274321021824 utils.py:1231] [100200] lr = 3.56300500075608e-05
+I1205 04:21:16.050986 137274321021824 utils.py:1231] [100200] uptime = 628265.413347279
+I1205 04:21:16.051047 137274321021824 utils.py:1231] [100200] examples_seen = 102604800.0
+I1205 04:21:16.051101 137274321021824 utils.py:1231] [100200] progress = 0.8898519577631147
+I1205 04:21:16.051159 137274321021824 utils.py:1231] [100200] epoch = 80.08698319578946
+I1205 04:21:16.051216 137274321021824 utils.py:1231] [100200] img/sec/core = 164.21178846157275
+I1205 04:21:16.051277 137274321021824 utils.py:1231] [100200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.48385277188748
+I1205 04:21:16.051333 137274321021824 utils.py:1231] [100200] core_hours = 174.48385277188748
+I1205 04:21:16.051399 137274321021824 train.py:125] NOTE: Steps:100200/112603 [89.0%]
+Walltime:7d6h31m (0s eval)
+ETA:21h35m
+Total train time:8d4h5m
+I1205 04:26:26.473506 137274321021824 utils.py:1231] [100250] l2_params = 239.8040217004939
+I1205 04:26:26.473726 137274321021824 utils.py:1231] [100250] train/loss = 1.5054250359535217
+I1205 04:26:26.473870 137274321021824 utils.py:1231] [100250] l2_grads = 2.763947010040283
+I1205 04:26:26.473986 137274321021824 utils.py:1231] [100250] lr = 3.53468090299886e-05
+I1205 04:26:26.474071 137274321021824 utils.py:1231] [100250] uptime = 628575.836427148
+I1205 04:26:26.474156 137274321021824 utils.py:1231] [100250] examples_seen = 102656000.0
+I1205 04:26:26.474243 137274321021824 utils.py:1231] [100250] progress = 0.8902959956661901
+I1205 04:26:26.474324 137274321021824 utils.py:1231] [100250] epoch = 80.12694676025842
+I1205 04:26:26.474398 137274321021824 utils.py:1231] [100250] img/sec/core = 164.93618973693825
+I1205 04:26:26.474474 137274321021824 utils.py:1231] [100250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.57008140518442
+I1205 04:26:26.474543 137274321021824 utils.py:1231] [100250] core_hours = 174.57008140518442
+I1205 04:26:26.474636 137274321021824 train.py:125] NOTE: Steps:100250/112603 [89.0%]
+Walltime:7d6h36m (0s eval)
+ETA:21h30m
+Total train time:8d4h5m
+I1205 04:31:38.119005 137274321021824 utils.py:1231] [100300] l2_params = 239.78209879076286
+I1205 04:31:38.119246 137274321021824 utils.py:1231] [100300] train/loss = 1.4835462123155594
+I1205 04:31:38.119352 137274321021824 utils.py:1231] [100300] l2_grads = 2.66337513923645
+I1205 04:31:38.119422 137274321021824 utils.py:1231] [100300] lr = 3.506465710405989e-05
+I1205 04:31:38.119482 137274321021824 utils.py:1231] [100300] uptime = 628887.481842509
+I1205 04:31:38.119548 137274321021824 utils.py:1231] [100300] examples_seen = 102707200.0
+I1205 04:31:38.119602 137274321021824 utils.py:1231] [100300] progress = 0.8907400335692655
+I1205 04:31:38.119656 137274321021824 utils.py:1231] [100300] epoch = 80.16691032472738
+I1205 04:31:38.119710 137274321021824 utils.py:1231] [100300] img/sec/core = 164.28927709616747
+I1205 04:31:38.119770 137274321021824 utils.py:1231] [100300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.65664957611804
+I1205 04:31:38.119822 137274321021824 utils.py:1231] [100300] core_hours = 174.65664957611804
+I1205 04:31:38.119888 137274321021824 train.py:125] NOTE: Steps:100300/112603 [89.1%]
+Walltime:7d6h41m (0s eval)
+ETA:21h25m
+Total train time:8d4h5m
+I1205 04:36:48.195231 137274321021824 utils.py:1231] [100350] l2_params = 239.7603060434727
+I1205 04:36:48.195476 137274321021824 utils.py:1231] [100350] train/loss = 1.5535383969545364
+I1205 04:36:48.195595 137274321021824 utils.py:1231] [100350] l2_grads = 2.797224283218384
+I1205 04:36:48.195682 137274321021824 utils.py:1231] [100350] lr = 3.478359489108062e-05
+I1205 04:36:48.195750 137274321021824 utils.py:1231] [100350] uptime = 629197.558112396
+I1205 04:36:48.195816 137274321021824 utils.py:1231] [100350] examples_seen = 102758400.0
+I1205 04:36:48.195869 137274321021824 utils.py:1231] [100350] progress = 0.8911840714723409
+I1205 04:36:48.195934 137274321021824 utils.py:1231] [100350] epoch = 80.20687388919633
+I1205 04:36:48.195990 137274321021824 utils.py:1231] [100350] img/sec/core = 165.12066537259454
+I1205 04:36:48.196068 137274321021824 utils.py:1231] [100350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.7427818733089
+I1205 04:36:48.196118 137274321021824 utils.py:1231] [100350] core_hours = 174.7427818733089
+I1205 04:36:48.196204 137274321021824 train.py:125] NOTE: Steps:100350/112603 [89.1%]
+Walltime:7d6h46m (0s eval)
+ETA:21h20m
+Total train time:8d4h4m
+I1205 04:41:58.984086 137274321021824 utils.py:1231] [100400] l2_params = 239.74168138024532
+I1205 04:41:58.984443 137274321021824 utils.py:1231] [100400] train/loss = 1.4808622151613235
+I1205 04:41:58.984675 137274321021824 utils.py:1231] [100400] l2_grads = 2.8384854793548584
+I1205 04:41:58.984760 137274321021824 utils.py:1231] [100400] lr = 3.450362304980286e-05
+I1205 04:41:58.984826 137274321021824 utils.py:1231] [100400] uptime = 629508.347185294
+I1205 04:41:58.984911 137274321021824 utils.py:1231] [100400] examples_seen = 102809600.0
+I1205 04:41:58.984996 137274321021824 utils.py:1231] [100400] progress = 0.8916281093754163
+I1205 04:41:58.985062 137274321021824 utils.py:1231] [100400] epoch = 80.2468374536653
+I1205 04:41:58.985130 137274321021824 utils.py:1231] [100400] img/sec/core = 164.7419567315559
+I1205 04:41:58.985252 137274321021824 utils.py:1231] [100400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.8291121713361
+I1205 04:41:58.985332 137274321021824 utils.py:1231] [100400] core_hours = 174.8291121713361
+I1205 04:41:58.985428 137274321021824 train.py:125] NOTE: Steps:100400/112603 [89.2%]
+Walltime:7d6h51m (0s eval)
+ETA:21h14m
+Total train time:8d4h4m
+I1205 04:47:09.275243 137274321021824 utils.py:1231] [100450] l2_params = 239.7219005553474
+I1205 04:47:09.275499 137274321021824 utils.py:1231] [100450] train/loss = 1.6047794818878174
+I1205 04:47:09.275649 137274321021824 utils.py:1231] [100450] l2_grads = 2.6733341217041016
+I1205 04:47:09.275751 137274321021824 utils.py:1231] [100450] lr = 3.4224742236423354e-05
+I1205 04:47:09.275846 137274321021824 utils.py:1231] [100450] uptime = 629818.638202515
+I1205 04:47:09.275939 137274321021824 utils.py:1231] [100450] examples_seen = 102860800.0
+I1205 04:47:09.276032 137274321021824 utils.py:1231] [100450] progress = 0.8920721472784917
+I1205 04:47:09.276105 137274321021824 utils.py:1231] [100450] epoch = 80.28680101813424
+I1205 04:47:09.276173 137274321021824 utils.py:1231] [100450] img/sec/core = 165.00638806293216
+I1205 04:47:09.276255 137274321021824 utils.py:1231] [100450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 174.91530412056414
+I1205 04:47:09.276332 137274321021824 utils.py:1231] [100450] core_hours = 174.91530412056414
+I1205 04:47:09.276412 137274321021824 train.py:125] NOTE: Steps:100450/112603 [89.2%]
+Walltime:7d6h56m (0s eval)
+ETA:21h9m
+Total train time:8d4h4m
+I1205 04:52:18.557202 137274321021824 utils.py:1231] [100500] l2_params = 239.70145134167177
+I1205 04:52:18.557404 137274321021824 utils.py:1231] [100500] train/loss = 3.6756739616394043
+I1205 04:52:18.557512 137274321021824 utils.py:1231] [100500] l2_grads = 2.812354803085327
+I1205 04:52:18.557583 137274321021824 utils.py:1231] [100500] lr = 3.394695310458112e-05
+I1205 04:52:18.557642 137274321021824 utils.py:1231] [100500] uptime = 630127.920003992
+I1205 04:52:18.557698 137274321021824 utils.py:1231] [100500] examples_seen = 102912000.0
+I1205 04:52:18.557752 137274321021824 utils.py:1231] [100500] progress = 0.8925161851815671
+I1205 04:52:18.557807 137274321021824 utils.py:1231] [100500] epoch = 80.3267645826032
+I1205 04:52:18.557862 137274321021824 utils.py:1231] [100500] img/sec/core = 165.5448194994919
+I1205 04:52:18.557927 137274321021824 utils.py:1231] [100500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.00121573208557
+I1205 04:52:18.557985 137274321021824 utils.py:1231] [100500] core_hours = 175.00121573208557
+I1205 04:52:18.558047 137274321021824 train.py:125] NOTE: Steps:100500/112603 [89.3%]
+Walltime:7d7h2m (0s eval)
+ETA:21h4m
+Total train time:8d4h4m
+I1205 04:57:26.771123 137274321021824 utils.py:1231] [100550] l2_params = 239.68136509488178
+I1205 04:57:26.771369 137274321021824 utils.py:1231] [100550] train/loss = 1.5543691664934158
+I1205 04:57:26.771507 137274321021824 utils.py:1231] [100550] l2_grads = 2.779207468032837
+I1205 04:57:26.771608 137274321021824 utils.py:1231] [100550] lr = 3.367025630535719e-05
+I1205 04:57:26.771683 137274321021824 utils.py:1231] [100550] uptime = 630436.13404023
+I1205 04:57:26.771763 137274321021824 utils.py:1231] [100550] examples_seen = 102963200.0
+I1205 04:57:26.771828 137274321021824 utils.py:1231] [100550] progress = 0.8929602230846425
+I1205 04:57:26.771906 137274321021824 utils.py:1231] [100550] epoch = 80.36672814707217
+I1205 04:57:26.771966 137274321021824 utils.py:1231] [100550] img/sec/core = 166.11832681259884
+I1205 04:57:26.772026 137274321021824 utils.py:1231] [100550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.08683074215165
+I1205 04:57:26.772083 137274321021824 utils.py:1231] [100550] core_hours = 175.08683074215165
+I1205 04:57:26.772149 137274321021824 train.py:125] NOTE: Steps:100550/112603 [89.3%]
+Walltime:7d7h7m (0s eval)
+ETA:20h59m
+Total train time:8d4h4m
+I1205 05:02:35.618163 137274321021824 utils.py:1231] [100600] l2_params = 239.66220653969404
+I1205 05:02:35.618363 137274321021824 utils.py:1231] [100600] train/loss = 1.4923080205917358
+I1205 05:02:35.618474 137274321021824 utils.py:1231] [100600] l2_grads = 2.6030523777008057
+I1205 05:02:35.618546 137274321021824 utils.py:1231] [100600] lr = 3.3394652487271765e-05
+I1205 05:02:35.618605 137274321021824 utils.py:1231] [100600] uptime = 630744.980966727
+I1205 05:02:35.618666 137274321021824 utils.py:1231] [100600] examples_seen = 103014400.0
+I1205 05:02:35.618728 137274321021824 utils.py:1231] [100600] progress = 0.8934042609877179
+I1205 05:02:35.618781 137274321021824 utils.py:1231] [100600] epoch = 80.40669171154111
+I1205 05:02:35.618835 137274321021824 utils.py:1231] [100600] img/sec/core = 165.77791652554114
+I1205 05:02:35.618907 137274321021824 utils.py:1231] [100600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.1726215550675
+I1205 05:02:35.618964 137274321021824 utils.py:1231] [100600] core_hours = 175.1726215550675
+I1205 05:02:35.619053 137274321021824 train.py:125] NOTE: Steps:100600/112603 [89.3%]
+Walltime:7d7h12m (0s eval)
+ETA:20h54m
+Total train time:8d4h4m
+I1205 05:07:42.518687 137274321021824 utils.py:1231] [100650] l2_params = 239.6436048042489
+I1205 05:07:42.518949 137274321021824 utils.py:1231] [100650] train/loss = 1.4298203587532043
+I1205 05:07:42.519145 137274321021824 utils.py:1231] [100650] l2_grads = 2.8458433151245117
+I1205 05:07:42.519232 137274321021824 utils.py:1231] [100650] lr = 3.3120142296283624e-05
+I1205 05:07:42.519296 137274321021824 utils.py:1231] [100650] uptime = 631051.881656517
+I1205 05:07:42.519364 137274321021824 utils.py:1231] [100650] examples_seen = 103065600.0
+I1205 05:07:42.519422 137274321021824 utils.py:1231] [100650] progress = 0.8938482988907933
+I1205 05:07:42.519479 137274321021824 utils.py:1231] [100650] epoch = 80.44665527601008
+I1205 05:07:42.519540 137274321021824 utils.py:1231] [100650] img/sec/core = 166.82921121822798
+I1205 05:07:42.519610 137274321021824 utils.py:1231] [100650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.25787174667585
+I1205 05:07:42.519668 137274321021824 utils.py:1231] [100650] core_hours = 175.25787174667585
+I1205 05:07:42.519735 137274321021824 train.py:125] NOTE: Steps:100650/112603 [89.4%]
+Walltime:7d7h17m (0s eval)
+ETA:20h48m
+Total train time:8d4h4m
+I1205 05:12:50.350006 137274321021824 utils.py:1231] [100700] l2_params = 239.62363325618745
+I1205 05:12:50.350269 137274321021824 utils.py:1231] [100700] train/loss = 2.754681408405304
+I1205 05:12:50.350507 137274321021824 utils.py:1231] [100700] l2_grads = 2.5543696880340576
+I1205 05:12:50.350591 137274321021824 utils.py:1231] [100700] lr = 3.284672637578841e-05
+I1205 05:12:50.350645 137274321021824 utils.py:1231] [100700] uptime = 631359.71300683
+I1205 05:12:50.350703 137274321021824 utils.py:1231] [100700] examples_seen = 103116800.0
+I1205 05:12:50.350757 137274321021824 utils.py:1231] [100700] progress = 0.8942923367938688
+I1205 05:12:50.350805 137274321021824 utils.py:1231] [100700] epoch = 80.48661884047903
+I1205 05:12:50.350857 137274321021824 utils.py:1231] [100700] img/sec/core = 166.32483971484402
+I1205 05:12:50.350927 137274321021824 utils.py:1231] [100700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.34338045509608
+I1205 05:12:50.350982 137274321021824 utils.py:1231] [100700] core_hours = 175.34338045509608
+I1205 05:12:50.351043 137274321021824 train.py:125] NOTE: Steps:100700/112603 [89.4%]
+Walltime:7d7h22m (0s eval)
+ETA:20h43m
+Total train time:8d4h4m
+I1205 05:17:59.611814 137274321021824 utils.py:1231] [100750] l2_params = 239.6043621327423
+I1205 05:17:59.612086 137274321021824 utils.py:1231] [100750] train/loss = 3.466521054506302
+I1205 05:17:59.612264 137274321021824 utils.py:1231] [100750] l2_grads = 2.6603517532348633
+I1205 05:17:59.612368 137274321021824 utils.py:1231] [100750] lr = 3.257440536661664e-05
+I1205 05:17:59.612449 137274321021824 utils.py:1231] [100750] uptime = 631668.974792995
+I1205 05:17:59.612533 137274321021824 utils.py:1231] [100750] examples_seen = 103168000.0
+I1205 05:17:59.612588 137274321021824 utils.py:1231] [100750] progress = 0.8947363746969441
+I1205 05:17:59.612646 137274321021824 utils.py:1231] [100750] epoch = 80.52658240494799
+I1205 05:17:59.612702 137274321021824 utils.py:1231] [100750] img/sec/core = 165.5555335009045
+I1205 05:17:59.612773 137274321021824 utils.py:1231] [100750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.4292865068086
+I1205 05:17:59.612833 137274321021824 utils.py:1231] [100750] core_hours = 175.4292865068086
+I1205 05:17:59.612899 137274321021824 train.py:125] NOTE: Steps:100750/112603 [89.5%]
+Walltime:7d7h27m (0s eval)
+ETA:20h38m
+Total train time:8d4h4m
+I1205 05:23:10.610641 137274321021824 utils.py:1231] [100800] l2_params = 239.58671414545506
+I1205 05:23:10.610870 137274321021824 utils.py:1231] [100800] train/loss = 1.6582448333501816
+I1205 05:23:10.610969 137274321021824 utils.py:1231] [100800] l2_grads = 2.6551127433776855
+I1205 05:23:10.611031 137274321021824 utils.py:1231] [100800] lr = 3.2303179907033165e-05
+I1205 05:23:10.611082 137274321021824 utils.py:1231] [100800] uptime = 631979.973443432
+I1205 05:23:10.611138 137274321021824 utils.py:1231] [100800] examples_seen = 103219200.0
+I1205 05:23:10.611186 137274321021824 utils.py:1231] [100800] progress = 0.8951804126000196
+I1205 05:23:10.611234 137274321021824 utils.py:1231] [100800] epoch = 80.56654596941695
+I1205 05:23:10.611283 137274321021824 utils.py:1231] [100800] img/sec/core = 164.6309394849728
+I1205 05:23:10.611337 137274321021824 utils.py:1231] [100800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.5156750208189
+I1205 05:23:10.611391 137274321021824 utils.py:1231] [100800] core_hours = 175.5156750208189
+I1205 05:23:10.611449 137274321021824 train.py:125] NOTE: Steps:100800/112603 [89.5%]
+Walltime:7d7h32m (0s eval)
+ETA:20h33m
+Total train time:8d4h4m
+I1205 05:28:22.392180 137274321021824 utils.py:1231] [100850] l2_params = 239.5670587713998
+I1205 05:28:22.392417 137274321021824 utils.py:1231] [100850] train/loss = 1.3648950904607773
+I1205 05:28:22.392510 137274321021824 utils.py:1231] [100850] l2_grads = 2.617945432662964
+I1205 05:28:22.392575 137274321021824 utils.py:1231] [100850] lr = 3.203305063273452e-05
+I1205 05:28:22.392627 137274321021824 utils.py:1231] [100850] uptime = 632291.754989291
+I1205 05:28:22.392681 137274321021824 utils.py:1231] [100850] examples_seen = 103270400.0
+I1205 05:28:22.392730 137274321021824 utils.py:1231] [100850] progress = 0.8956244505030949
+I1205 05:28:22.392780 137274321021824 utils.py:1231] [100850] epoch = 80.6065095338859
+I1205 05:28:22.392831 137274321021824 utils.py:1231] [100850] img/sec/core = 164.2175448804896
+I1205 05:28:22.392895 137274321021824 utils.py:1231] [100850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.6022810057797
+I1205 05:28:22.392949 137274321021824 utils.py:1231] [100850] core_hours = 175.6022810057797
+I1205 05:28:22.393011 137274321021824 train.py:125] NOTE: Steps:100850/112603 [89.6%]
+Walltime:7d7h38m (0s eval)
+ETA:20h27m
+Total train time:8d4h4m
+I1205 05:33:31.455701 137274321021824 utils.py:1231] [100900] l2_params = 239.54725753487892
+I1205 05:33:31.455970 137274321021824 utils.py:1231] [100900] train/loss = 1.4856857657432556
+I1205 05:33:31.456076 137274321021824 utils.py:1231] [100900] l2_grads = 2.7929134368896484
+I1205 05:33:31.456156 137274321021824 utils.py:1231] [100900] lr = 3.176401817684824e-05
+I1205 05:33:31.456219 137274321021824 utils.py:1231] [100900] uptime = 632600.818576359
+I1205 05:33:31.456274 137274321021824 utils.py:1231] [100900] examples_seen = 103321600.0
+I1205 05:33:31.456323 137274321021824 utils.py:1231] [100900] progress = 0.8960684884061704
+I1205 05:33:31.456373 137274321021824 utils.py:1231] [100900] epoch = 80.64647309835486
+I1205 05:33:31.456427 137274321021824 utils.py:1231] [100900] img/sec/core = 165.66170245327325
+I1205 05:33:31.456503 137274321021824 utils.py:1231] [100900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.6881320021875
+I1205 05:33:31.456559 137274321021824 utils.py:1231] [100900] core_hours = 175.6881320021875
+I1205 05:33:31.456619 137274321021824 train.py:125] NOTE: Steps:100900/112603 [89.6%]
+Walltime:7d7h43m (0s eval)
+ETA:20h22m
+Total train time:8d4h4m
+I1205 05:38:38.664324 137274321021824 utils.py:1231] [100950] l2_params = 239.52850128159403
+I1205 05:38:38.664581 137274321021824 utils.py:1231] [100950] train/loss = 1.497110202908516
+I1205 05:38:38.664721 137274321021824 utils.py:1231] [100950] l2_grads = 2.727825880050659
+I1205 05:38:38.664819 137274321021824 utils.py:1231] [100950] lr = 3.149608316993129e-05
+I1205 05:38:38.664887 137274321021824 utils.py:1231] [100950] uptime = 632908.02724349
+I1205 05:38:38.664950 137274321021824 utils.py:1231] [100950] examples_seen = 103372800.0
+I1205 05:38:38.665007 137274321021824 utils.py:1231] [100950] progress = 0.8965125263092457
+I1205 05:38:38.665065 137274321021824 utils.py:1231] [100950] epoch = 80.68643666282382
+I1205 05:38:38.665122 137274321021824 utils.py:1231] [100950] img/sec/core = 166.66196457980533
+I1205 05:38:38.665180 137274321021824 utils.py:1231] [100950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.7734677430572
+I1205 05:38:38.665235 137274321021824 utils.py:1231] [100950] core_hours = 175.7734677430572
+I1205 05:38:38.665301 137274321021824 train.py:125] NOTE: Steps:100950/112603 [89.7%]
+Walltime:7d7h48m (0s eval)
+ETA:20h17m
+Total train time:8d4h4m
+I1205 05:43:46.284888 137274321021824 utils.py:1231] [101000] l2_params = 239.510609819075
+I1205 05:43:46.285160 137274321021824 utils.py:1231] [101000] train/loss = 2.9724868834018707
+I1205 05:43:46.285290 137274321021824 utils.py:1231] [101000] l2_grads = 2.5017528533935547
+I1205 05:43:46.285388 137274321021824 utils.py:1231] [101000] lr = 3.122924623996804e-05
+I1205 05:43:46.285454 137274321021824 utils.py:1231] [101000] uptime = 633215.647815947
+I1205 05:43:46.285526 137274321021824 utils.py:1231] [101000] examples_seen = 103424000.0
+I1205 05:43:46.285589 137274321021824 utils.py:1231] [101000] progress = 0.8969565642123212
+I1205 05:43:46.285648 137274321021824 utils.py:1231] [101000] epoch = 80.72640022729277
+I1205 05:43:46.285718 137274321021824 utils.py:1231] [101000] img/sec/core = 166.43880346188897
+I1205 05:43:46.285794 137274321021824 utils.py:1231] [101000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.85891790207305
+I1205 05:43:46.285867 137274321021824 utils.py:1231] [101000] core_hours = 175.85891790207305
+I1205 05:43:46.285941 137274321021824 train.py:125] NOTE: Steps:101000/112603 [89.7%]
+Walltime:7d7h53m (0s eval)
+ETA:20h12m
+Total train time:8d4h3m
+I1205 05:48:55.708053 137274321021824 utils.py:1231] [101050] l2_params = 239.4933626732516
+I1205 05:48:55.708230 137274321021824 utils.py:1231] [101050] train/loss = 3.117815136909485
+I1205 05:48:55.708321 137274321021824 utils.py:1231] [101050] l2_grads = 2.5768356323242188
+I1205 05:48:55.708384 137274321021824 utils.py:1231] [101050] lr = 3.096350801236979e-05
+I1205 05:48:55.708436 137274321021824 utils.py:1231] [101050] uptime = 633525.0707983259
+I1205 05:48:55.708490 137274321021824 utils.py:1231] [101050] examples_seen = 103475200.0
+I1205 05:48:55.708542 137274321021824 utils.py:1231] [101050] progress = 0.8974006021153965
+I1205 05:48:55.708590 137274321021824 utils.py:1231] [101050] epoch = 80.76636379176173
+I1205 05:48:55.708641 137274321021824 utils.py:1231] [101050] img/sec/core = 165.46928610910544
+I1205 05:48:55.708713 137274321021824 utils.py:1231] [101050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 175.94486873051164
+I1205 05:48:55.708763 137274321021824 utils.py:1231] [101050] core_hours = 175.94486873051164
+I1205 05:48:55.708823 137274321021824 train.py:125] NOTE: Steps:101050/112603 [89.7%]
+Walltime:7d7h58m (0s eval)
+ETA:20h6m
+Total train time:8d4h3m
+I1205 05:54:03.956446 137274321021824 utils.py:1231] [101100] l2_params = 239.4730731943992
+I1205 05:54:03.956692 137274321021824 utils.py:1231] [101100] train/loss = 1.4737872928380966
+I1205 05:54:03.956818 137274321021824 utils.py:1231] [101100] l2_grads = 2.797811508178711
+I1205 05:54:03.956900 137274321021824 utils.py:1231] [101100] lr = 3.0698869109972145e-05
+I1205 05:54:03.956954 137274321021824 utils.py:1231] [101100] uptime = 633833.319315557
+I1205 05:54:03.957007 137274321021824 utils.py:1231] [101100] examples_seen = 103526400.0
+I1205 05:54:03.957057 137274321021824 utils.py:1231] [101100] progress = 0.897844640018472
+I1205 05:54:03.957106 137274321021824 utils.py:1231] [101100] epoch = 80.80632735623068
+I1205 05:54:03.957158 137274321021824 utils.py:1231] [101100] img/sec/core = 166.09974464732605
+I1205 05:54:03.957216 137274321021824 utils.py:1231] [101100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.0304933186314
+I1205 05:54:03.957265 137274321021824 utils.py:1231] [101100] core_hours = 176.0304933186314
+I1205 05:54:03.957324 137274321021824 train.py:125] NOTE: Steps:101100/112603 [89.8%]
+Walltime:7d8h3m (0s eval)
+ETA:20h1m
+Total train time:8d4h3m
+I1205 05:59:11.879067 137274321021824 utils.py:1231] [101150] l2_params = 239.45464748626648
+I1205 05:59:11.879259 137274321021824 utils.py:1231] [101150] train/loss = 2.8709352016448975
+I1205 05:59:11.879356 137274321021824 utils.py:1231] [101150] l2_grads = 2.4336841106414795
+I1205 05:59:11.879428 137274321021824 utils.py:1231] [101150] lr = 3.043533015303427e-05
+I1205 05:59:11.879489 137274321021824 utils.py:1231] [101150] uptime = 634141.241848183
+I1205 05:59:11.879545 137274321021824 utils.py:1231] [101150] examples_seen = 103577600.0
+I1205 05:59:11.879597 137274321021824 utils.py:1231] [101150] progress = 0.8982886779215474
+I1205 05:59:11.879650 137274321021824 utils.py:1231] [101150] epoch = 80.84629092069964
+I1205 05:59:11.879706 137274321021824 utils.py:1231] [101150] img/sec/core = 166.27558744521428
+I1205 05:59:11.879782 137274321021824 utils.py:1231] [101150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.11602735547194
+I1205 05:59:11.879864 137274321021824 utils.py:1231] [101150] core_hours = 176.11602735547194
+I1205 05:59:11.879959 137274321021824 train.py:125] NOTE: Steps:101150/112603 [89.8%]
+Walltime:7d8h9m (0s eval)
+ETA:19h56m
+Total train time:8d4h3m
+I1205 06:04:23.664673 137274321021824 utils.py:1231] [101200] l2_params = 239.43620534030043
+I1205 06:04:23.664874 137274321021824 utils.py:1231] [101200] train/loss = 2.1876838952302933
+I1205 06:04:23.664989 137274321021824 utils.py:1231] [101200] l2_grads = 2.595721483230591
+I1205 06:04:23.665091 137274321021824 utils.py:1231] [101200] lr = 3.0172891759237524e-05
+I1205 06:04:23.665161 137274321021824 utils.py:1231] [101200] uptime = 634453.027520689
+I1205 06:04:23.665228 137274321021824 utils.py:1231] [101200] examples_seen = 103628800.0
+I1205 06:04:23.665289 137274321021824 utils.py:1231] [101200] progress = 0.8987327158246228
+I1205 06:04:23.665350 137274321021824 utils.py:1231] [101200] epoch = 80.8862544851686
+I1205 06:04:23.665409 137274321021824 utils.py:1231] [101200] img/sec/core = 164.21537137509077
+I1205 06:04:23.665484 137274321021824 utils.py:1231] [101200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.20263448672358
+I1205 06:04:23.665548 137274321021824 utils.py:1231] [101200] core_hours = 176.20263448672358
+I1205 06:04:23.665623 137274321021824 train.py:125] NOTE: Steps:101200/112603 [89.9%]
+Walltime:7d8h14m (0s eval)
+ETA:19h51m
+Total train time:8d4h3m
+I1205 06:09:33.066136 137274321021824 utils.py:1231] [101250] l2_params = 239.41720202567922
+I1205 06:09:33.066383 137274321021824 utils.py:1231] [101250] train/loss = 1.408611699938774
+I1205 06:09:33.066488 137274321021824 utils.py:1231] [101250] l2_grads = 2.6846320629119873
+I1205 06:09:33.066573 137274321021824 utils.py:1231] [101250] lr = 2.9911554543683467e-05
+I1205 06:09:33.066630 137274321021824 utils.py:1231] [101250] uptime = 634762.4289919
+I1205 06:09:33.066686 137274321021824 utils.py:1231] [101250] examples_seen = 103680000.0
+I1205 06:09:33.066735 137274321021824 utils.py:1231] [101250] progress = 0.8991767537276982
+I1205 06:09:33.066787 137274321021824 utils.py:1231] [101250] epoch = 80.92621804963755
+I1205 06:09:33.066841 137274321021824 utils.py:1231] [101250] img/sec/core = 165.48079037764086
+I1205 06:09:33.066908 137274321021824 utils.py:1231] [101250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.28857933983775
+I1205 06:09:33.066968 137274321021824 utils.py:1231] [101250] core_hours = 176.28857933983775
+I1205 06:09:33.067044 137274321021824 train.py:125] NOTE: Steps:101250/112603 [89.9%]
+Walltime:7d8h19m (0s eval)
+ETA:19h46m
+Total train time:8d4h3m
+I1205 06:14:44.852086 137274321021824 utils.py:1231] [101300] l2_params = 239.3990585926097
+I1205 06:14:44.852323 137274321021824 utils.py:1231] [101300] train/loss = 3.8587308526039124
+I1205 06:14:44.852490 137274321021824 utils.py:1231] [101300] l2_grads = 2.8137118816375732
+I1205 06:14:44.852601 137274321021824 utils.py:1231] [101300] lr = 2.9651319118892933e-05
+I1205 06:14:44.852696 137274321021824 utils.py:1231] [101300] uptime = 635074.215051848
+I1205 06:14:44.852789 137274321021824 utils.py:1231] [101300] examples_seen = 103731200.0
+I1205 06:14:44.852878 137274321021824 utils.py:1231] [101300] progress = 0.8996207916307736
+I1205 06:14:44.852958 137274321021824 utils.py:1231] [101300] epoch = 80.96618161410652
+I1205 06:14:44.853032 137274321021824 utils.py:1231] [101300] img/sec/core = 164.21516731227067
+I1205 06:14:44.853112 137274321021824 utils.py:1231] [101300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.3751865787122
+I1205 06:14:44.853195 137274321021824 utils.py:1231] [101300] core_hours = 176.3751865787122
+I1205 06:14:44.853294 137274321021824 train.py:125] NOTE: Steps:101300/112603 [90.0%]
+Walltime:7d8h24m (0s eval)
+ETA:19h40m
+Total train time:8d4h3m
+I1205 06:19:54.898419 137274321021824 utils.py:1231] [101350] l2_params = 239.3814047821854
+I1205 06:19:54.898696 137274321021824 utils.py:1231] [101350] train/loss = 1.4451881498098373
+I1205 06:19:54.898839 137274321021824 utils.py:1231] [101350] l2_grads = 2.6924495697021484
+I1205 06:19:54.898943 137274321021824 utils.py:1231] [101350] lr = 2.9392186094804333e-05
+I1205 06:19:54.898998 137274321021824 utils.py:1231] [101350] uptime = 635384.261360269
+I1205 06:19:54.899058 137274321021824 utils.py:1231] [101350] examples_seen = 103782400.0
+I1205 06:19:54.899105 137274321021824 utils.py:1231] [101350] progress = 0.900064829533849
+I1205 06:19:54.899152 137274321021824 utils.py:1231] [101350] epoch = 81.00614517857547
+I1205 06:19:54.899201 137274321021824 utils.py:1231] [101350] img/sec/core = 165.13662188316363
+I1205 06:19:54.899257 137274321021824 utils.py:1231] [101350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.46131055327362
+I1205 06:19:54.899309 137274321021824 utils.py:1231] [101350] core_hours = 176.46131055327362
+I1205 06:19:54.899368 137274321021824 train.py:125] NOTE: Steps:101350/112603 [90.0%]
+Walltime:7d8h29m (0s eval)
+ETA:19h35m
+Total train time:8d4h3m
+I1205 06:25:04.783611 137274321021824 utils.py:1231] [101400] l2_params = 239.3629557923608
+I1205 06:25:04.783823 137274321021824 utils.py:1231] [101400] train/loss = 1.5379250347614288
+I1205 06:25:04.783924 137274321021824 utils.py:1231] [101400] l2_grads = 2.8072850704193115
+I1205 06:25:04.783988 137274321021824 utils.py:1231] [101400] lr = 2.9134156078772193e-05
+I1205 06:25:04.784038 137274321021824 utils.py:1231] [101400] uptime = 635694.146400419
+I1205 06:25:04.784088 137274321021824 utils.py:1231] [101400] examples_seen = 103833600.0
+I1205 06:25:04.784135 137274321021824 utils.py:1231] [101400] progress = 0.9005088674369244
+I1205 06:25:04.784181 137274321021824 utils.py:1231] [101400] epoch = 81.04610874304443
+I1205 06:25:04.784230 137274321021824 utils.py:1231] [101400] img/sec/core = 165.2225611640752
+I1205 06:25:04.784286 137274321021824 utils.py:1231] [101400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.54738973109303
+I1205 06:25:04.784333 137274321021824 utils.py:1231] [101400] core_hours = 176.54738973109303
+I1205 06:25:04.784392 137274321021824 train.py:125] NOTE: Steps:101400/112603 [90.1%]
+Walltime:7d8h34m (0s eval)
+ETA:19h30m
+Total train time:8d4h3m
+I1205 06:30:14.717361 137274321021824 utils.py:1231] [101450] l2_params = 239.34583303728252
+I1205 06:30:14.717644 137274321021824 utils.py:1231] [101450] train/loss = 1.4127670228481293
+I1205 06:30:14.717776 137274321021824 utils.py:1231] [101450] l2_grads = 2.6399893760681152
+I1205 06:30:14.717872 137274321021824 utils.py:1231] [101450] lr = 2.887722967556594e-05
+I1205 06:30:14.717953 137274321021824 utils.py:1231] [101450] uptime = 636004.080314461
+I1205 06:30:14.718014 137274321021824 utils.py:1231] [101450] examples_seen = 103884800.0
+I1205 06:30:14.718070 137274321021824 utils.py:1231] [101450] progress = 0.9009529053399998
+I1205 06:30:14.718138 137274321021824 utils.py:1231] [101450] epoch = 81.08607230751339
+I1205 06:30:14.718195 137274321021824 utils.py:1231] [101450] img/sec/core = 165.19650699809787
+I1205 06:30:14.718258 137274321021824 utils.py:1231] [101450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.6334824849936
+I1205 06:30:14.718315 137274321021824 utils.py:1231] [101450] core_hours = 176.6334824849936
+I1205 06:30:14.718380 137274321021824 train.py:125] NOTE: Steps:101450/112603 [90.1%]
+Walltime:7d8h40m (0s eval)
+ETA:19h25m
+Total train time:8d4h3m
+I1205 06:35:25.498318 137274321021824 utils.py:1231] [101500] l2_params = 239.32827843224513
+I1205 06:35:25.498516 137274321021824 utils.py:1231] [101500] train/loss = 1.4875477105379105
+I1205 06:35:25.498627 137274321021824 utils.py:1231] [101500] l2_grads = 2.6474459171295166
+I1205 06:35:25.498701 137274321021824 utils.py:1231] [101500] lr = 2.862140748736813e-05
+I1205 06:35:25.498763 137274321021824 utils.py:1231] [101500] uptime = 636314.861124314
+I1205 06:35:25.498821 137274321021824 utils.py:1231] [101500] examples_seen = 103936000.0
+I1205 06:35:25.498889 137274321021824 utils.py:1231] [101500] progress = 0.9013969432430752
+I1205 06:35:25.498949 137274321021824 utils.py:1231] [101500] epoch = 81.12603587198234
+I1205 06:35:25.499004 137274321021824 utils.py:1231] [101500] img/sec/core = 164.74633689325555
+I1205 06:35:25.499061 137274321021824 utils.py:1231] [101500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.71981048773057
+I1205 06:35:25.499114 137274321021824 utils.py:1231] [101500] core_hours = 176.71981048773057
+I1205 06:35:25.499177 137274321021824 train.py:125] NOTE: Steps:101500/112603 [90.1%]
+Walltime:7d8h45m (0s eval)
+ETA:19h19m
+Total train time:8d4h3m
+I1205 06:40:35.240922 137274321021824 utils.py:1231] [101550] l2_params = 239.31154712129637
+I1205 06:40:35.241195 137274321021824 utils.py:1231] [101550] train/loss = 1.4441606551408768
+I1205 06:40:35.241335 137274321021824 utils.py:1231] [101550] l2_grads = 2.53222393989563
+I1205 06:40:35.241422 137274321021824 utils.py:1231] [101550] lr = 2.8366690113773692e-05
+I1205 06:40:35.241484 137274321021824 utils.py:1231] [101550] uptime = 636624.603845611
+I1205 06:40:35.241546 137274321021824 utils.py:1231] [101550] examples_seen = 103987200.0
+I1205 06:40:35.241599 137274321021824 utils.py:1231] [101550] progress = 0.9018409811461506
+I1205 06:40:35.241658 137274321021824 utils.py:1231] [101550] epoch = 81.1659994364513
+I1205 06:40:35.241730 137274321021824 utils.py:1231] [101550] img/sec/core = 165.29847670225774
+I1205 06:40:35.241804 137274321021824 utils.py:1231] [101550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.80585013253528
+I1205 06:40:35.241857 137274321021824 utils.py:1231] [101550] core_hours = 176.80585013253528
+I1205 06:40:35.241955 137274321021824 train.py:125] NOTE: Steps:101550/112603 [90.2%]
+Walltime:7d8h50m (0s eval)
+ETA:19h14m
+Total train time:8d4h3m
+I1205 06:45:44.958142 137274321021824 utils.py:1231] [101600] l2_params = 239.2938404237066
+I1205 06:45:44.958403 137274321021824 utils.py:1231] [101600] train/loss = 1.379101499915123
+I1205 06:45:44.958535 137274321021824 utils.py:1231] [101600] l2_grads = 2.6439208984375
+I1205 06:45:44.958629 137274321021824 utils.py:1231] [101600] lr = 2.8113078151787532e-05
+I1205 06:45:44.958698 137274321021824 utils.py:1231] [101600] uptime = 636934.321060168
+I1205 06:45:44.958774 137274321021824 utils.py:1231] [101600] examples_seen = 104038400.0
+I1205 06:45:44.958835 137274321021824 utils.py:1231] [101600] progress = 0.9022850190492261
+I1205 06:45:44.958904 137274321021824 utils.py:1231] [101600] epoch = 81.20596300092025
+I1205 06:45:44.958963 137274321021824 utils.py:1231] [101600] img/sec/core = 165.31208984697352
+I1205 06:45:44.959027 137274321021824 utils.py:1231] [101600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.89188269213443
+I1205 06:45:44.959084 137274321021824 utils.py:1231] [101600] core_hours = 176.89188269213443
+I1205 06:45:44.959150 137274321021824 train.py:125] NOTE: Steps:101600/112603 [90.2%]
+Walltime:7d8h55m (0s eval)
+ETA:19h9m
+Total train time:8d4h3m
+I1205 06:50:56.743396 137274321021824 utils.py:1231] [101650] l2_params = 239.27757201208397
+I1205 06:50:56.743607 137274321021824 utils.py:1231] [101650] train/loss = 2.3299254775047302
+I1205 06:50:56.743708 137274321021824 utils.py:1231] [101650] l2_grads = 2.664726972579956
+I1205 06:50:56.743780 137274321021824 utils.py:1231] [101650] lr = 2.7860572195824153e-05
+I1205 06:50:56.743849 137274321021824 utils.py:1231] [101650] uptime = 637246.106211163
+I1205 06:50:56.743916 137274321021824 utils.py:1231] [101650] examples_seen = 104089600.0
+I1205 06:50:56.743985 137274321021824 utils.py:1231] [101650] progress = 0.9027290569523014
+I1205 06:50:56.744038 137274321021824 utils.py:1231] [101650] epoch = 81.24592656538921
+I1205 06:50:56.744093 137274321021824 utils.py:1231] [101650] img/sec/core = 164.21564605178727
+I1205 06:50:56.744150 137274321021824 utils.py:1231] [101650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 176.97848967852192
+I1205 06:50:56.744202 137274321021824 utils.py:1231] [101650] core_hours = 176.97848967852192
+I1205 06:50:56.744265 137274321021824 train.py:125] NOTE: Steps:101650/112603 [90.3%]
+Walltime:7d9h0m (0s eval)
+ETA:19h4m
+Total train time:8d4h3m
+I1205 06:56:04.247317 137274321021824 utils.py:1231] [101700] l2_params = 239.26189468924636
+I1205 06:56:04.247588 137274321021824 utils.py:1231] [101700] train/loss = 3.3744480311870575
+I1205 06:56:04.247725 137274321021824 utils.py:1231] [101700] l2_grads = 2.6699845790863037
+I1205 06:56:04.247797 137274321021824 utils.py:1231] [101700] lr = 2.7609172837705807e-05
+I1205 06:56:04.247853 137274321021824 utils.py:1231] [101700] uptime = 637553.610215218
+I1205 06:56:04.247911 137274321021824 utils.py:1231] [101700] examples_seen = 104140800.0
+I1205 06:56:04.247960 137274321021824 utils.py:1231] [101700] progress = 0.9031730948553769
+I1205 06:56:04.248009 137274321021824 utils.py:1231] [101700] epoch = 81.28589012985817
+I1205 06:56:04.248060 137274321021824 utils.py:1231] [101700] img/sec/core = 166.50189696662852
+I1205 06:56:04.248116 137274321021824 utils.py:1231] [101700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.06390745742613
+I1205 06:56:04.248165 137274321021824 utils.py:1231] [101700] core_hours = 177.06390745742613
+I1205 06:56:04.248225 137274321021824 train.py:125] NOTE: Steps:101700/112603 [90.3%]
+Walltime:7d9h5m (0s eval)
+ETA:18h58m
+Total train time:8d4h3m
+I1205 07:01:10.918176 137274321021824 utils.py:1231] [101750] l2_params = 239.24634099011396
+I1205 07:01:10.918424 137274321021824 utils.py:1231] [101750] train/loss = 1.5267649292945862
+I1205 07:01:10.918525 137274321021824 utils.py:1231] [101750] l2_grads = 2.625818967819214
+I1205 07:01:10.918604 137274321021824 utils.py:1231] [101750] lr = 2.735888066666064e-05
+I1205 07:01:10.918663 137274321021824 utils.py:1231] [101750] uptime = 637860.2810254979
+I1205 07:01:10.918714 137274321021824 utils.py:1231] [101750] examples_seen = 104192000.0
+I1205 07:01:10.918762 137274321021824 utils.py:1231] [101750] progress = 0.9036171327584522
+I1205 07:01:10.918808 137274321021824 utils.py:1231] [101750] epoch = 81.32585369432712
+I1205 07:01:10.918857 137274321021824 utils.py:1231] [101750] img/sec/core = 166.95426588943747
+I1205 07:01:10.918922 137274321021824 utils.py:1231] [101750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.14909379361498
+I1205 07:01:10.918972 137274321021824 utils.py:1231] [101750] core_hours = 177.14909379361498
+I1205 07:01:10.919031 137274321021824 train.py:125] NOTE: Steps:101750/112603 [90.4%]
+Walltime:7d9h11m (0s eval)
+ETA:18h53m
+Total train time:8d4h2m
+I1205 07:06:18.273093 137274321021824 utils.py:1231] [101800] l2_params = 239.22889522128867
+I1205 07:06:18.273298 137274321021824 utils.py:1231] [101800] train/loss = 3.711828589439392
+I1205 07:06:18.273402 137274321021824 utils.py:1231] [101800] l2_grads = 2.906827926635742
+I1205 07:06:18.273492 137274321021824 utils.py:1231] [101800] lr = 2.7109696269322343e-05
+I1205 07:06:18.273587 137274321021824 utils.py:1231] [101800] uptime = 638167.635945796
+I1205 07:06:18.273671 137274321021824 utils.py:1231] [101800] examples_seen = 104243200.0
+I1205 07:06:18.273739 137274321021824 utils.py:1231] [101800] progress = 0.9040611706615277
+I1205 07:06:18.273809 137274321021824 utils.py:1231] [101800] epoch = 81.36581725879608
+I1205 07:06:18.273871 137274321021824 utils.py:1231] [101800] img/sec/core = 166.58265939046208
+I1205 07:06:18.273957 137274321021824 utils.py:1231] [101800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.23447016036445
+I1205 07:06:18.274043 137274321021824 utils.py:1231] [101800] core_hours = 177.23447016036445
+I1205 07:06:18.274175 137274321021824 train.py:125] NOTE: Steps:101800/112603 [90.4%]
+Walltime:7d9h16m (0s eval)
+ETA:18h48m
+Total train time:8d4h2m
+I1205 07:11:30.044585 137274321021824 utils.py:1231] [101850] l2_params = 239.2120926200351
+I1205 07:11:30.044802 137274321021824 utils.py:1231] [101850] train/loss = 3.7387313842773438
+I1205 07:11:30.044916 137274321021824 utils.py:1231] [101850] l2_grads = 2.7859535217285156
+I1205 07:11:30.044992 137274321021824 utils.py:1231] [101850] lr = 2.6861620229727856e-05
+I1205 07:11:30.045054 137274321021824 utils.py:1231] [101850] uptime = 638479.4074156419
+I1205 07:11:30.045120 137274321021824 utils.py:1231] [101850] examples_seen = 104294400.0
+I1205 07:11:30.045178 137274321021824 utils.py:1231] [101850] progress = 0.904505208564603
+I1205 07:11:30.045235 137274321021824 utils.py:1231] [101850] epoch = 81.40578082326503
+I1205 07:11:30.045294 137274321021824 utils.py:1231] [101850] img/sec/core = 164.22285215932806
+I1205 07:11:30.045360 137274321021824 utils.py:1231] [101850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.32107334643277
+I1205 07:11:30.045420 137274321021824 utils.py:1231] [101850] core_hours = 177.32107334643277
+I1205 07:11:30.045488 137274321021824 train.py:125] NOTE: Steps:101850/112603 [90.5%]
+Walltime:7d9h21m (0s eval)
+ETA:18h43m
+Total train time:8d4h2m
+I1205 07:16:36.208852 137274321021824 utils.py:1231] [101900] l2_params = 239.19597637613785
+I1205 07:16:36.209052 137274321021824 utils.py:1231] [101900] train/loss = 1.5276419520378113
+I1205 07:16:36.209149 137274321021824 utils.py:1231] [101900] l2_grads = 2.6491811275482178
+I1205 07:16:36.209231 137274321021824 utils.py:1231] [101900] lr = 2.661465312931634e-05
+I1205 07:16:36.209304 137274321021824 utils.py:1231] [101900] uptime = 638785.57166623
+I1205 07:16:36.209363 137274321021824 utils.py:1231] [101900] examples_seen = 104345600.0
+I1205 07:16:36.209421 137274321021824 utils.py:1231] [101900] progress = 0.9049492464676785
+I1205 07:16:36.209477 137274321021824 utils.py:1231] [101900] epoch = 81.445744387734
+I1205 07:16:36.209530 137274321021824 utils.py:1231] [101900] img/sec/core = 167.23049768763343
+I1205 07:16:36.209607 137274321021824 utils.py:1231] [101900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.4061189715961
+I1205 07:16:36.209661 137274321021824 utils.py:1231] [101900] core_hours = 177.4061189715961
+I1205 07:16:36.209727 137274321021824 train.py:125] NOTE: Steps:101900/112603 [90.5%]
+Walltime:7d9h26m (0s eval)
+ETA:18h38m
+Total train time:8d4h2m
+I1205 07:21:47.977868 137274321021824 utils.py:1231] [101950] l2_params = 239.18123672035952
+I1205 07:21:47.978076 137274321021824 utils.py:1231] [101950] train/loss = 1.6093259006738663
+I1205 07:21:47.978190 137274321021824 utils.py:1231] [101950] l2_grads = 3.1585533618927
+I1205 07:21:47.978264 137274321021824 utils.py:1231] [101950] lr = 2.636879554692824e-05
+I1205 07:21:47.978323 137274321021824 utils.py:1231] [101950] uptime = 639097.340685033
+I1205 07:21:47.978392 137274321021824 utils.py:1231] [101950] examples_seen = 104396800.0
+I1205 07:21:47.978449 137274321021824 utils.py:1231] [101950] progress = 0.9053932843707538
+I1205 07:21:47.978505 137274321021824 utils.py:1231] [101950] epoch = 81.48570795220296
+I1205 07:21:47.978563 137274321021824 utils.py:1231] [101950] img/sec/core = 164.22414323454026
+I1205 07:21:47.978645 137274321021824 utils.py:1231] [101950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.49272147681916
+I1205 07:21:47.978704 137274321021824 utils.py:1231] [101950] core_hours = 177.49272147681916
+I1205 07:21:47.978770 137274321021824 train.py:125] NOTE: Steps:101950/112603 [90.5%]
+Walltime:7d9h31m (0s eval)
+ETA:18h32m
+Total train time:8d4h2m
+I1205 07:26:52.782518 137274321021824 utils.py:1231] [102000] l2_params = 239.16517349631596
+I1205 07:26:52.782772 137274321021824 utils.py:1231] [102000] train/loss = 3.8166200816631317
+I1205 07:26:52.782896 137274321021824 utils.py:1231] [102000] l2_grads = 2.960641384124756
+I1205 07:26:52.782969 137274321021824 utils.py:1231] [102000] lr = 2.6124048058803047e-05
+I1205 07:26:52.783031 137274321021824 utils.py:1231] [102000] uptime = 639402.145393265
+I1205 07:26:52.783093 137274321021824 utils.py:1231] [102000] examples_seen = 104448000.0
+I1205 07:26:52.783147 137274321021824 utils.py:1231] [102000] progress = 0.9058373222738293
+I1205 07:26:52.783196 137274321021824 utils.py:1231] [102000] epoch = 81.5256715166719
+I1205 07:26:52.783262 137274321021824 utils.py:1231] [102000] img/sec/core = 167.9764079006277
+I1205 07:26:52.783337 137274321021824 utils.py:1231] [102000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.57738945132803
+I1205 07:26:52.783397 137274321021824 utils.py:1231] [102000] core_hours = 177.57738945132803
+I1205 07:26:52.783477 137274321021824 train.py:125] NOTE: Steps:102000/112603 [90.6%]
+Walltime:7d9h36m (0s eval)
+ETA:18h27m
+Total train time:8d4h2m
+I1205 07:32:00.554465 137274321021824 utils.py:1231] [102050] l2_params = 239.14949584093196
+I1205 07:32:00.554794 137274321021824 utils.py:1231] [102050] train/loss = 1.6580608040094376
+I1205 07:32:00.554932 137274321021824 utils.py:1231] [102050] l2_grads = 2.9368770122528076
+I1205 07:32:00.554998 137274321021824 utils.py:1231] [102050] lr = 2.5880411238578685e-05
+I1205 07:32:00.555053 137274321021824 utils.py:1231] [102050] uptime = 639709.917414066
+I1205 07:32:00.555108 137274321021824 utils.py:1231] [102050] examples_seen = 104499200.0
+I1205 07:32:00.555160 137274321021824 utils.py:1231] [102050] progress = 0.9062813601769047
+I1205 07:32:00.555212 137274321021824 utils.py:1231] [102050] epoch = 81.56563508114087
+I1205 07:32:00.555265 137274321021824 utils.py:1231] [102050] img/sec/core = 166.3569023159859
+I1205 07:32:00.555325 137274321021824 utils.py:1231] [102050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.66288167932834
+I1205 07:32:00.555379 137274321021824 utils.py:1231] [102050] core_hours = 177.66288167932834
+I1205 07:32:00.555461 137274321021824 train.py:125] NOTE: Steps:102050/112603 [90.6%]
+Walltime:7d9h41m (0s eval)
+ETA:18h22m
+Total train time:8d4h2m
+I1205 07:37:11.157349 137274321021824 utils.py:1231] [102100] l2_params = 239.1339820440144
+I1205 07:37:11.157601 137274321021824 utils.py:1231] [102100] train/loss = 3.436411887407303
+I1205 07:37:11.157820 137274321021824 utils.py:1231] [102100] l2_grads = 2.7250096797943115
+I1205 07:37:11.157931 137274321021824 utils.py:1231] [102100] lr = 2.5637885657289618e-05
+I1205 07:37:11.157994 137274321021824 utils.py:1231] [102100] uptime = 640020.520356383
+I1205 07:37:11.158053 137274321021824 utils.py:1231] [102100] examples_seen = 104550400.0
+I1205 07:37:11.158104 137274321021824 utils.py:1231] [102100] progress = 0.9067253980799801
+I1205 07:37:11.158163 137274321021824 utils.py:1231] [102100] epoch = 81.60559864560982
+I1205 07:37:11.158218 137274321021824 utils.py:1231] [102100] img/sec/core = 164.84067928680616
+I1205 07:37:11.158287 137274321021824 utils.py:1231] [102100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.74916027441637
+I1205 07:37:11.158347 137274321021824 utils.py:1231] [102100] core_hours = 177.74916027441637
+I1205 07:37:11.158425 137274321021824 train.py:125] NOTE: Steps:102100/112603 [90.7%]
+Walltime:7d9h47m (0s eval)
+ETA:18h17m
+Total train time:8d4h2m
+I1205 07:42:21.106534 137274321021824 utils.py:1231] [102150] l2_params = 239.1183499828865
+I1205 07:42:21.106739 137274321021824 utils.py:1231] [102150] train/loss = 1.4967008531093597
+I1205 07:42:21.106834 137274321021824 utils.py:1231] [102150] l2_grads = 2.4992504119873047
+I1205 07:42:21.106919 137274321021824 utils.py:1231] [102150] lr = 2.5396471883366287e-05
+I1205 07:42:21.106988 137274321021824 utils.py:1231] [102150] uptime = 640330.469333755
+I1205 07:42:21.107047 137274321021824 utils.py:1231] [102150] examples_seen = 104601600.0
+I1205 07:42:21.107094 137274321021824 utils.py:1231] [102150] progress = 0.9071694359830555
+I1205 07:42:21.107141 137274321021824 utils.py:1231] [102150] epoch = 81.64556221007878
+I1205 07:42:21.107190 137274321021824 utils.py:1231] [102150] img/sec/core = 165.18847854931113
+I1205 07:42:21.107244 137274321021824 utils.py:1231] [102150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.83525721257527
+I1205 07:42:21.107306 137274321021824 utils.py:1231] [102150] core_hours = 177.83525721257527
+I1205 07:42:21.107365 137274321021824 train.py:125] NOTE: Steps:102150/112603 [90.7%]
+Walltime:7d9h52m (0s eval)
+ETA:18h11m
+Total train time:8d4h2m
+I1205 07:47:30.928835 137274321021824 utils.py:1231] [102200] l2_params = 239.10388744428573
+I1205 07:47:30.929070 137274321021824 utils.py:1231] [102200] train/loss = 1.4143006354570389
+I1205 07:47:30.929193 137274321021824 utils.py:1231] [102200] l2_grads = 2.7552430629730225
+I1205 07:47:30.929280 137274321021824 utils.py:1231] [102200] lr = 2.5156170482632834e-05
+I1205 07:47:30.929337 137274321021824 utils.py:1231] [102200] uptime = 640640.291699226
+I1205 07:47:30.929393 137274321021824 utils.py:1231] [102200] examples_seen = 104652800.0
+I1205 07:47:30.929447 137274321021824 utils.py:1231] [102200] progress = 0.9076134738861309
+I1205 07:47:30.929500 137274321021824 utils.py:1231] [102200] epoch = 81.68552577454774
+I1205 07:47:30.929550 137274321021824 utils.py:1231] [102200] img/sec/core = 165.25598441596335
+I1205 07:47:30.929607 137274321021824 utils.py:1231] [102200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 177.92131898076167
+I1205 07:47:30.929657 137274321021824 utils.py:1231] [102200] core_hours = 177.92131898076167
+I1205 07:47:30.929720 137274321021824 train.py:125] NOTE: Steps:102200/112603 [90.8%]
+Walltime:7d9h57m (0s eval)
+ETA:18h6m
+Total train time:8d4h2m
+I1205 07:52:38.466123 137274321021824 utils.py:1231] [102250] l2_params = 239.08862240555234
+I1205 07:52:38.466349 137274321021824 utils.py:1231] [102250] train/loss = 1.3915291726589203
+I1205 07:52:38.466446 137274321021824 utils.py:1231] [102250] l2_grads = 2.8631203174591064
+I1205 07:52:38.466528 137274321021824 utils.py:1231] [102250] lr = 2.491698201830637e-05
+I1205 07:52:38.466593 137274321021824 utils.py:1231] [102250] uptime = 640947.828954606
+I1205 07:52:38.466649 137274321021824 utils.py:1231] [102250] examples_seen = 104704000.0
+I1205 07:52:38.466705 137274321021824 utils.py:1231] [102250] progress = 0.9080575117892064
+I1205 07:52:38.466766 137274321021824 utils.py:1231] [102250] epoch = 81.72548933901669
+I1205 07:52:38.466821 137274321021824 utils.py:1231] [102250] img/sec/core = 166.4838945666725
+I1205 07:52:38.466885 137274321021824 utils.py:1231] [102250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.00674599614499
+I1205 07:52:38.466943 137274321021824 utils.py:1231] [102250] core_hours = 178.00674599614499
+I1205 07:52:38.467009 137274321021824 train.py:125] NOTE: Steps:102250/112603 [90.8%]
+Walltime:7d10h2m (0s eval)
+ETA:18h1m
+Total train time:8d4h2m
+I1205 07:57:50.256552 137274321021824 utils.py:1231] [102300] l2_params = 239.07288236375268
+I1205 07:57:50.256765 137274321021824 utils.py:1231] [102300] train/loss = 1.5320264250040054
+I1205 07:57:50.256893 137274321021824 utils.py:1231] [102300] l2_grads = 2.844419002532959
+I1205 07:57:50.256975 137274321021824 utils.py:1231] [102300] lr = 2.4678907050995664e-05
+I1205 07:57:50.257035 137274321021824 utils.py:1231] [102300] uptime = 641259.619396665
+I1205 07:57:50.257087 137274321021824 utils.py:1231] [102300] examples_seen = 104755200.0
+I1205 07:57:50.257133 137274321021824 utils.py:1231] [102300] progress = 0.9085015496922817
+I1205 07:57:50.257179 137274321021824 utils.py:1231] [102300] epoch = 81.76545290348565
+I1205 07:57:50.257226 137274321021824 utils.py:1231] [102300] img/sec/core = 164.21285932268358
+I1205 07:57:50.257280 137274321021824 utils.py:1231] [102300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.0933544522725
+I1205 07:57:50.257328 137274321021824 utils.py:1231] [102300] core_hours = 178.0933544522725
+I1205 07:57:50.257384 137274321021824 train.py:125] NOTE: Steps:102300/112603 [90.9%]
+Walltime:7d10h7m (0s eval)
+ETA:17h56m
+Total train time:8d4h2m
+I1205 08:02:57.993194 137274321021824 utils.py:1231] [102350] l2_params = 239.05769178130308
+I1205 08:02:57.993471 137274321021824 utils.py:1231] [102350] train/loss = 2.0894617587327957
+I1205 08:02:57.993576 137274321021824 utils.py:1231] [102350] l2_grads = 2.5331900119781494
+I1205 08:02:57.993651 137274321021824 utils.py:1231] [102350] lr = 2.4441946138699314e-05
+I1205 08:02:57.993711 137274321021824 utils.py:1231] [102350] uptime = 641567.356072935
+I1205 08:02:57.993771 137274321021824 utils.py:1231] [102350] examples_seen = 104806400.0
+I1205 08:02:57.993827 137274321021824 utils.py:1231] [102350] progress = 0.9089455875953572
+I1205 08:02:57.993893 137274321021824 utils.py:1231] [102350] epoch = 81.8054164679546
+I1205 08:02:57.993955 137274321021824 utils.py:1231] [102350] img/sec/core = 166.37600893263107
+I1205 08:02:57.994017 137274321021824 utils.py:1231] [102350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.1788368623475
+I1205 08:02:57.994076 137274321021824 utils.py:1231] [102350] core_hours = 178.1788368623475
+I1205 08:02:57.994150 137274321021824 train.py:125] NOTE: Steps:102350/112603 [90.9%]
+Walltime:7d10h12m (0s eval)
+ETA:17h50m
+Total train time:8d4h1m
+I1205 08:08:09.781270 137274321021824 utils.py:1231] [102400] l2_params = 239.04369523419325
+I1205 08:08:09.781482 137274321021824 utils.py:1231] [102400] train/loss = 1.3327296823263168
+I1205 08:08:09.781585 137274321021824 utils.py:1231] [102400] l2_grads = 2.622781991958618
+I1205 08:08:09.781663 137274321021824 utils.py:1231] [102400] lr = 2.4206099836805217e-05
+I1205 08:08:09.781724 137274321021824 utils.py:1231] [102400] uptime = 641879.144085122
+I1205 08:08:09.781785 137274321021824 utils.py:1231] [102400] examples_seen = 104857600.0
+I1205 08:08:09.781844 137274321021824 utils.py:1231] [102400] progress = 0.9093896254984325
+I1205 08:08:09.781913 137274321021824 utils.py:1231] [102400] epoch = 81.84538003242356
+I1205 08:08:09.781981 137274321021824 utils.py:1231] [102400] img/sec/core = 164.2141390904167
+I1205 08:08:09.782040 137274321021824 utils.py:1231] [102400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.26544464351056
+I1205 08:08:09.782092 137274321021824 utils.py:1231] [102400] core_hours = 178.26544464351056
+I1205 08:08:09.782158 137274321021824 train.py:125] NOTE: Steps:102400/112603 [90.9%]
+Walltime:7d10h17m (0s eval)
+ETA:17h45m
+Total train time:8d4h1m
+I1205 08:13:15.636071 137274321021824 utils.py:1231] [102450] l2_params = 239.02907843207436
+I1205 08:13:15.636296 137274321021824 utils.py:1231] [102450] train/loss = 3.1004608273506165
+I1205 08:13:15.636402 137274321021824 utils.py:1231] [102450] l2_grads = 2.8055849075317383
+I1205 08:13:15.636470 137274321021824 utils.py:1231] [102450] lr = 2.3971368698088642e-05
+I1205 08:13:15.636526 137274321021824 utils.py:1231] [102450] uptime = 642184.998888474
+I1205 08:13:15.636587 137274321021824 utils.py:1231] [102450] examples_seen = 104908800.0
+I1205 08:13:15.636641 137274321021824 utils.py:1231] [102450] progress = 0.909833663401508
+I1205 08:13:15.636712 137274321021824 utils.py:1231] [102450] epoch = 81.88534359689253
+I1205 08:13:15.636786 137274321021824 utils.py:1231] [102450] img/sec/core = 167.3996923994324
+I1205 08:13:15.636849 137274321021824 utils.py:1231] [102450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.35040431110832
+I1205 08:13:15.636912 137274321021824 utils.py:1231] [102450] core_hours = 178.35040431110832
+I1205 08:13:15.636973 137274321021824 train.py:125] NOTE: Steps:102450/112603 [91.0%]
+Walltime:7d10h23m (0s eval)
+ETA:17h40m
+Total train time:8d4h1m
+I1205 08:18:22.931596 137274321021824 utils.py:1231] [102500] l2_params = 239.01419602901044
+I1205 08:18:22.931814 137274321021824 utils.py:1231] [102500] train/loss = 1.5008531510829926
+I1205 08:18:22.931916 137274321021824 utils.py:1231] [102500] l2_grads = 2.703458070755005
+I1205 08:18:22.931994 137274321021824 utils.py:1231] [102500] lr = 2.3737753272711056e-05
+I1205 08:18:22.932045 137274321021824 utils.py:1231] [102500] uptime = 642492.294407537
+I1205 08:18:22.932099 137274321021824 utils.py:1231] [102500] examples_seen = 104960000.0
+I1205 08:18:22.932150 137274321021824 utils.py:1231] [102500] progress = 0.9102777013045834
+I1205 08:18:22.932199 137274321021824 utils.py:1231] [102500] epoch = 81.92530716136147
+I1205 08:18:22.932256 137274321021824 utils.py:1231] [102500] img/sec/core = 166.6148603667095
+I1205 08:18:22.932311 137274321021824 utils.py:1231] [102500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.4357641775147
+I1205 08:18:22.932361 137274321021824 utils.py:1231] [102500] core_hours = 178.4357641775147
+I1205 08:18:22.932423 137274321021824 train.py:125] NOTE: Steps:102500/112603 [91.0%]
+Walltime:7d10h28m (0s eval)
+ETA:17h35m
+Total train time:8d4h1m
+I1205 08:18:22.932516 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:102500/112603 [91.0%]
+Walltime:7d10h28m (0s eval)
+ETA:17h35m
+Total train time:8d4h1m
+I1205 08:19:56.156876 137274321021824 utils.py:1231] [102500] val/acc@1 = 0.7615194515306123
+I1205 08:19:56.157078 137274321021824 utils.py:1231] [102500] val/loss = 0.9365957215124247
+I1205 08:19:56.157230 137274321021824 utils.py:1231] [102500] z/secs/eval/val = 93.22464763210155
+I1205 08:19:56.157290 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 93.22464763210155
+I1205 08:25:05.521626 137274321021824 utils.py:1231] [102550] l2_params = 238.9992008392887
+I1205 08:25:05.521845 137274321021824 utils.py:1231] [102550] train/loss = 2.0586300641298294
+I1205 08:25:05.521957 137274321021824 utils.py:1231] [102550] l2_grads = 2.599717378616333
+I1205 08:25:05.522031 137274321021824 utils.py:1231] [102550] lr = 2.3505254108219143e-05
+I1205 08:25:05.522091 137274321021824 utils.py:1231] [102550] uptime = 642894.884452757
+I1205 08:25:05.522151 137274321021824 utils.py:1231] [102550] examples_seen = 105011200.0
+I1205 08:25:05.522208 137274321021824 utils.py:1231] [102550] progress = 0.9107217392076588
+I1205 08:25:05.522264 137274321021824 utils.py:1231] [102550] epoch = 81.96527072583044
+I1205 08:25:05.522320 137274321021824 utils.py:1231] [102550] img/sec/core = 127.1765176707809
+I1205 08:25:05.522381 137274321021824 utils.py:1231] [102550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.54759474563136
+I1205 08:25:05.522436 137274321021824 utils.py:1231] [102550] core_hours = 178.54759474563136
+I1205 08:25:05.522503 137274321021824 train.py:125] NOTE: Steps:102550/112603 [91.1%]
+Walltime:7d10h34m (0s eval)
+ETA:17h30m
+Total train time:8d4h3m
+I1205 08:30:10.287449 137274321021824 utils.py:1231] [102600] l2_params = 238.98473523010858
+I1205 08:30:10.287675 137274321021824 utils.py:1231] [102600] train/loss = 3.5646126866340637
+I1205 08:30:10.287771 137274321021824 utils.py:1231] [102600] l2_grads = 2.844120979309082
+I1205 08:30:10.287837 137274321021824 utils.py:1231] [102600] lr = 2.3273871749543106e-05
+I1205 08:30:10.287900 137274321021824 utils.py:1231] [102600] uptime = 643199.650261217
+I1205 08:30:10.287955 137274321021824 utils.py:1231] [102600] examples_seen = 105062400.0
+I1205 08:30:10.288022 137274321021824 utils.py:1231] [102600] progress = 0.9111657771107342
+I1205 08:30:10.288077 137274321021824 utils.py:1231] [102600] epoch = 82.00523429029938
+I1205 08:30:10.288130 137274321021824 utils.py:1231] [102600] img/sec/core = 167.99784811395003
+I1205 08:30:10.288189 137274321021824 utils.py:1231] [102600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.63225191464804
+I1205 08:30:10.288241 137274321021824 utils.py:1231] [102600] core_hours = 178.63225191464804
+I1205 08:30:10.288305 137274321021824 train.py:125] NOTE: Steps:102600/112603 [91.1%]
+Walltime:7d10h39m (0s eval)
+ETA:17h24m
+Total train time:8d4h3m
+I1205 08:35:21.365361 137274321021824 utils.py:1231] [102650] l2_params = 238.9711930461989
+I1205 08:35:21.365581 137274321021824 utils.py:1231] [102650] train/loss = 1.4771282225847244
+I1205 08:35:21.365714 137274321021824 utils.py:1231] [102650] l2_grads = 2.857799768447876
+I1205 08:35:21.365805 137274321021824 utils.py:1231] [102650] lr = 2.304360673899582e-05
+I1205 08:35:21.365890 137274321021824 utils.py:1231] [102650] uptime = 643510.7282431499
+I1205 08:35:21.365973 137274321021824 utils.py:1231] [102650] examples_seen = 105113600.0
+I1205 08:35:21.366045 137274321021824 utils.py:1231] [102650] progress = 0.9116098150138096
+I1205 08:35:21.366119 137274321021824 utils.py:1231] [102650] epoch = 82.04519785476835
+I1205 08:35:21.366191 137274321021824 utils.py:1231] [102650] img/sec/core = 164.58895509693554
+I1205 08:35:21.366263 137274321021824 utils.py:1231] [102650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.71866246518496
+I1205 08:35:21.366326 137274321021824 utils.py:1231] [102650] core_hours = 178.71866246518496
+I1205 08:35:21.366401 137274321021824 train.py:125] NOTE: Steps:102650/112603 [91.2%]
+Walltime:7d10h45m (0s eval)
+ETA:17h19m
+Total train time:8d4h3m
+I1205 08:40:25.749159 137274321021824 utils.py:1231] [102700] l2_params = 238.9571387120301
+I1205 08:40:25.749350 137274321021824 utils.py:1231] [102700] train/loss = 1.4508144855499268
+I1205 08:40:25.749449 137274321021824 utils.py:1231] [102700] l2_grads = 2.805385112762451
+I1205 08:40:25.749523 137274321021824 utils.py:1231] [102700] lr = 2.281445961627094e-05
+I1205 08:40:25.749581 137274321021824 utils.py:1231] [102700] uptime = 643815.111943487
+I1205 08:40:25.749638 137274321021824 utils.py:1231] [102700] examples_seen = 105164800.0
+I1205 08:40:25.749692 137274321021824 utils.py:1231] [102700] progress = 0.912053852916885
+I1205 08:40:25.749760 137274321021824 utils.py:1231] [102700] epoch = 82.08516141923731
+I1205 08:40:25.749814 137274321021824 utils.py:1231] [102700] img/sec/core = 168.20874423731928
+I1205 08:40:25.749875 137274321021824 utils.py:1231] [102700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.80321349305638
+I1205 08:40:25.749936 137274321021824 utils.py:1231] [102700] core_hours = 178.80321349305638
+I1205 08:40:25.750019 137274321021824 train.py:125] NOTE: Steps:102700/112603 [91.2%]
+Walltime:7d10h50m (0s eval)
+ETA:17h14m
+Total train time:8d4h2m
+I1205 08:45:33.409533 137274321021824 utils.py:1231] [102750] l2_params = 238.94314463394227
+I1205 08:45:33.409811 137274321021824 utils.py:1231] [102750] train/loss = 1.5720027536153793
+I1205 08:45:33.409939 137274321021824 utils.py:1231] [102750] l2_grads = 2.7596523761749268
+I1205 08:45:33.410015 137274321021824 utils.py:1231] [102750] lr = 2.2586430918442426e-05
+I1205 08:45:33.410079 137274321021824 utils.py:1231] [102750] uptime = 644122.772438454
+I1205 08:45:33.410139 137274321021824 utils.py:1231] [102750] examples_seen = 105216000.0
+I1205 08:45:33.410197 137274321021824 utils.py:1231] [102750] progress = 0.9124978908199604
+I1205 08:45:33.410252 137274321021824 utils.py:1231] [102750] epoch = 82.12512498370626
+I1205 08:45:33.410301 137274321021824 utils.py:1231] [102750] img/sec/core = 166.41720610080066
+I1205 08:45:33.410355 137274321021824 utils.py:1231] [102750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.88867474165832
+I1205 08:45:33.410405 137274321021824 utils.py:1231] [102750] core_hours = 178.88867474165832
+I1205 08:45:33.410466 137274321021824 train.py:125] NOTE: Steps:102750/112603 [91.2%]
+Walltime:7d10h55m (0s eval)
+ETA:17h9m
+Total train time:8d4h2m
+I1205 08:50:40.309159 137274321021824 utils.py:1231] [102800] l2_params = 238.92851568257186
+I1205 08:50:40.309392 137274321021824 utils.py:1231] [102800] train/loss = 3.227092534303665
+I1205 08:50:40.309501 137274321021824 utils.py:1231] [102800] l2_grads = 2.6108155250549316
+I1205 08:50:40.309577 137274321021824 utils.py:1231] [102800] lr = 2.2359521179962716e-05
+I1205 08:50:40.309624 137274321021824 utils.py:1231] [102800] uptime = 644429.671987206
+I1205 08:50:40.309674 137274321021824 utils.py:1231] [102800] examples_seen = 105267200.0
+I1205 08:50:40.309728 137274321021824 utils.py:1231] [102800] progress = 0.9129419287230358
+I1205 08:50:40.309779 137274321021824 utils.py:1231] [102800] epoch = 82.16508854817522
+I1205 08:50:40.309839 137274321021824 utils.py:1231] [102800] img/sec/core = 166.82983148135452
+I1205 08:50:40.309900 137274321021824 utils.py:1231] [102800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 178.97392461631165
+I1205 08:50:40.309948 137274321021824 utils.py:1231] [102800] core_hours = 178.97392461631165
+I1205 08:50:40.310004 137274321021824 train.py:125] NOTE: Steps:102800/112603 [91.3%]
+Walltime:7d11h0m (0s eval)
+ETA:17h4m
+Total train time:8d4h2m
+I1205 08:55:52.069270 137274321021824 utils.py:1231] [102850] l2_params = 238.9154682834701
+I1205 08:55:52.069464 137274321021824 utils.py:1231] [102850] train/loss = 2.7749032974243164
+I1205 08:55:52.069568 137274321021824 utils.py:1231] [102850] l2_grads = 2.4968678951263428
+I1205 08:55:52.069644 137274321021824 utils.py:1231] [102850] lr = 2.213373093266157e-05
+I1205 08:55:52.069698 137274321021824 utils.py:1231] [102850] uptime = 644741.432059534
+I1205 08:55:52.069751 137274321021824 utils.py:1231] [102850] examples_seen = 105318400.0
+I1205 08:55:52.069803 137274321021824 utils.py:1231] [102850] progress = 0.9133859666261112
+I1205 08:55:52.069851 137274321021824 utils.py:1231] [102850] epoch = 82.20505211264418
+I1205 08:55:52.069911 137274321021824 utils.py:1231] [102850] img/sec/core = 164.228855920098
+I1205 08:55:52.069971 137274321021824 utils.py:1231] [102850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.0605246364028
+I1205 08:55:52.070022 137274321021824 utils.py:1231] [102850] core_hours = 179.0605246364028
+I1205 08:55:52.070083 137274321021824 train.py:125] NOTE: Steps:102850/112603 [91.3%]
+Walltime:7d11h5m (0s eval)
+ETA:16h58m
+Total train time:8d4h2m
+I1205 09:00:48.817584 137274321021824 utils.py:1231] [102900] l2_params = 238.90161928662607
+I1205 09:00:48.817787 137274321021824 utils.py:1231] [102900] train/loss = 1.469970941543579
+I1205 09:00:48.817890 137274321021824 utils.py:1231] [102900] l2_grads = 2.780503034591675
+I1205 09:00:48.817972 137274321021824 utils.py:1231] [102900] lr = 2.1909060705744993e-05
+I1205 09:00:48.818050 137274321021824 utils.py:1231] [102900] uptime = 645038.180411496
+I1205 09:00:48.818112 137274321021824 utils.py:1231] [102900] examples_seen = 105369600.0
+I1205 09:00:48.818166 137274321021824 utils.py:1231] [102900] progress = 0.9138300045291866
+I1205 09:00:48.818221 137274321021824 utils.py:1231] [102900] epoch = 82.24501567711313
+I1205 09:00:48.818277 137274321021824 utils.py:1231] [102900] img/sec/core = 172.53676275366115
+I1205 09:00:48.818343 137274321021824 utils.py:1231] [102900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.14295473417
+I1205 09:00:48.818416 137274321021824 utils.py:1231] [102900] core_hours = 179.14295473417
+I1205 09:00:48.818483 137274321021824 train.py:125] NOTE: Steps:102900/112603 [91.4%]
+Walltime:7d11h10m (0s eval)
+ETA:16h53m
+Total train time:8d4h2m
+I1205 09:06:00.606431 137274321021824 utils.py:1231] [102950] l2_params = 238.88873576252365
+I1205 09:06:00.606676 137274321021824 utils.py:1231] [102950] train/loss = 2.2162828147411346
+I1205 09:06:00.606799 137274321021824 utils.py:1231] [102950] l2_grads = 2.6609599590301514
+I1205 09:06:00.606878 137274321021824 utils.py:1231] [102950] lr = 2.168551102579386e-05
+I1205 09:06:00.606935 137274321021824 utils.py:1231] [102950] uptime = 645349.969296951
+I1205 09:06:00.606988 137274321021824 utils.py:1231] [102950] examples_seen = 105420800.0
+I1205 09:06:00.607037 137274321021824 utils.py:1231] [102950] progress = 0.9142740424322621
+I1205 09:06:00.607085 137274321021824 utils.py:1231] [102950] epoch = 82.28497924158209
+I1205 09:06:00.607136 137274321021824 utils.py:1231] [102950] img/sec/core = 164.21367915435263
+I1205 09:06:00.607218 137274321021824 utils.py:1231] [102950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.22956275790747
+I1205 09:06:00.607270 137274321021824 utils.py:1231] [102950] core_hours = 179.22956275790747
+I1205 09:06:00.607336 137274321021824 train.py:125] NOTE: Steps:102950/112603 [91.4%]
+Walltime:7d11h15m (0s eval)
+ETA:16h48m
+Total train time:8d4h2m
+I1205 09:11:08.034017 137274321021824 utils.py:1231] [103000] l2_params = 238.87583715715337
+I1205 09:11:08.034279 137274321021824 utils.py:1231] [103000] train/loss = 2.960147649049759
+I1205 09:11:08.034399 137274321021824 utils.py:1231] [103000] l2_grads = 2.5889015197753906
+I1205 09:11:08.034507 137274321021824 utils.py:1231] [103000] lr = 2.146308241676249e-05
+I1205 09:11:08.034572 137274321021824 utils.py:1231] [103000] uptime = 645657.396933186
+I1205 09:11:08.034646 137274321021824 utils.py:1231] [103000] examples_seen = 105472000.0
+I1205 09:11:08.034705 137274321021824 utils.py:1231] [103000] progress = 0.9147180803353374
+I1205 09:11:08.034780 137274321021824 utils.py:1231] [103000] epoch = 82.32494280605104
+I1205 09:11:08.034839 137274321021824 utils.py:1231] [103000] img/sec/core = 166.5432575517018
+I1205 09:11:08.034902 137274321021824 utils.py:1231] [103000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.31495932352834
+I1205 09:11:08.034957 137274321021824 utils.py:1231] [103000] core_hours = 179.31495932352834
+I1205 09:11:08.035018 137274321021824 train.py:125] NOTE: Steps:103000/112603 [91.5%]
+Walltime:7d11h20m (0s eval)
+ETA:16h43m
+Total train time:8d4h2m
+I1205 09:16:19.343648 137274321021824 utils.py:1231] [103050] l2_params = 238.86251758006134
+I1205 09:16:19.343887 137274321021824 utils.py:1231] [103050] train/loss = 1.4597970694303513
+I1205 09:16:19.344001 137274321021824 utils.py:1231] [103050] l2_grads = 2.7172558307647705
+I1205 09:16:19.344078 137274321021824 utils.py:1231] [103050] lr = 2.124177539997818e-05
+I1205 09:16:19.344144 137274321021824 utils.py:1231] [103050] uptime = 645968.706501549
+I1205 09:16:19.344222 137274321021824 utils.py:1231] [103050] examples_seen = 105523200.0
+I1205 09:16:19.344300 137274321021824 utils.py:1231] [103050] progress = 0.9151621182384129
+I1205 09:16:19.344370 137274321021824 utils.py:1231] [103050] epoch = 82.36490637052
+I1205 09:16:19.344439 137274321021824 utils.py:1231] [103050] img/sec/core = 164.466515659092
+I1205 09:16:19.344505 137274321021824 utils.py:1231] [103050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.40143420362918
+I1205 09:16:19.344564 137274321021824 utils.py:1231] [103050] core_hours = 179.40143420362918
+I1205 09:16:19.344632 137274321021824 train.py:125] NOTE: Steps:103050/112603 [91.5%]
+Walltime:7d11h26m (0s eval)
+ETA:16h37m
+Total train time:8d4h2m
+I1205 09:21:30.746156 137274321021824 utils.py:1231] [103100] l2_params = 238.8493069883596
+I1205 09:21:30.746397 137274321021824 utils.py:1231] [103100] train/loss = 2.6322492361068726
+I1205 09:21:30.746513 137274321021824 utils.py:1231] [103100] l2_grads = 2.7776975631713867
+I1205 09:21:30.746591 137274321021824 utils.py:1231] [103100] lr = 2.102159049413899e-05
+I1205 09:21:30.746649 137274321021824 utils.py:1231] [103100] uptime = 646280.1090059549
+I1205 09:21:30.746717 137274321021824 utils.py:1231] [103100] examples_seen = 105574400.0
+I1205 09:21:30.746772 137274321021824 utils.py:1231] [103100] progress = 0.9156061561414882
+I1205 09:21:30.746825 137274321021824 utils.py:1231] [103100] epoch = 82.40486993498897
+I1205 09:21:30.746889 137274321021824 utils.py:1231] [103100] img/sec/core = 164.4174317020158
+I1205 09:21:30.746949 137274321021824 utils.py:1231] [103100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.4879348992975
+I1205 09:21:30.747000 137274321021824 utils.py:1231] [103100] core_hours = 179.4879348992975
+I1205 09:21:30.747061 137274321021824 train.py:125] NOTE: Steps:103100/112603 [91.6%]
+Walltime:7d11h31m (0s eval)
+ETA:16h32m
+Total train time:8d4h2m
+I1205 09:26:42.186052 137274321021824 utils.py:1231] [103150] l2_params = 238.8353968518068
+I1205 09:26:42.186271 137274321021824 utils.py:1231] [103150] train/loss = 1.4626312851905823
+I1205 09:26:42.186425 137274321021824 utils.py:1231] [103150] l2_grads = 2.7092814445495605
+I1205 09:26:42.186529 137274321021824 utils.py:1231] [103150] lr = 2.080252821531308e-05
+I1205 09:26:42.186618 137274321021824 utils.py:1231] [103150] uptime = 646591.5489737419
+I1205 09:26:42.186704 137274321021824 utils.py:1231] [103150] examples_seen = 105625600.0
+I1205 09:26:42.186785 137274321021824 utils.py:1231] [103150] progress = 0.9160501940445637
+I1205 09:26:42.186866 137274321021824 utils.py:1231] [103150] epoch = 82.44483349945791
+I1205 09:26:42.186969 137274321021824 utils.py:1231] [103150] img/sec/core = 164.39765378801707
+I1205 09:26:42.187042 137274321021824 utils.py:1231] [103150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.57444600146056
+I1205 09:26:42.187117 137274321021824 utils.py:1231] [103150] core_hours = 179.57444600146056
+I1205 09:26:42.187203 137274321021824 train.py:125] NOTE: Steps:103150/112603 [91.6%]
+Walltime:7d11h36m (0s eval)
+ETA:16h27m
+Total train time:8d4h2m
+I1205 09:31:53.536288 137274321021824 utils.py:1231] [103200] l2_params = 238.82231940256364
+I1205 09:31:53.536500 137274321021824 utils.py:1231] [103200] train/loss = 1.5402800887823105
+I1205 09:31:53.536625 137274321021824 utils.py:1231] [103200] l2_grads = 2.7076618671417236
+I1205 09:31:53.536694 137274321021824 utils.py:1231] [103200] lr = 2.0584589076937544e-05
+I1205 09:31:53.536746 137274321021824 utils.py:1231] [103200] uptime = 646902.899108666
+I1205 09:31:53.536800 137274321021824 utils.py:1231] [103200] examples_seen = 105676800.0
+I1205 09:31:53.536849 137274321021824 utils.py:1231] [103200] progress = 0.916494231947639
+I1205 09:31:53.536904 137274321021824 utils.py:1231] [103200] epoch = 82.48479706392688
+I1205 09:31:53.536957 137274321021824 utils.py:1231] [103200] img/sec/core = 164.44508691955096
+I1205 09:31:53.537013 137274321021824 utils.py:1231] [103200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.66093215005054
+I1205 09:31:53.537065 137274321021824 utils.py:1231] [103200] core_hours = 179.66093215005054
+I1205 09:31:53.537127 137274321021824 train.py:125] NOTE: Steps:103200/112603 [91.6%]
+Walltime:7d11h41m (0s eval)
+ETA:16h22m
+Total train time:8d4h2m
+I1205 09:37:04.867017 137274321021824 utils.py:1231] [103250] l2_params = 238.80996513718586
+I1205 09:37:04.867310 137274321021824 utils.py:1231] [103250] train/loss = 2.4207723736763
+I1205 09:37:04.867469 137274321021824 utils.py:1231] [103250] l2_grads = 2.549515724182129
+I1205 09:37:04.867571 137274321021824 utils.py:1231] [103250] lr = 2.036777358981672e-05
+I1205 09:37:04.867626 137274321021824 utils.py:1231] [103250] uptime = 647214.2299881739
+I1205 09:37:04.867681 137274321021824 utils.py:1231] [103250] examples_seen = 105728000.0
+I1205 09:37:04.867730 137274321021824 utils.py:1231] [103250] progress = 0.9169382698507145
+I1205 09:37:04.867779 137274321021824 utils.py:1231] [103250] epoch = 82.52476062839582
+I1205 09:37:04.867856 137274321021824 utils.py:1231] [103250] img/sec/core = 164.4552576375417
+I1205 09:37:04.867923 137274321021824 utils.py:1231] [103250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.74741294991387
+I1205 09:37:04.867978 137274321021824 utils.py:1231] [103250] core_hours = 179.74741294991387
+I1205 09:37:04.868039 137274321021824 train.py:125] NOTE: Steps:103250/112603 [91.7%]
+Walltime:7d11h46m (0s eval)
+ETA:16h16m
+Total train time:8d4h2m
+I1205 09:42:16.328312 137274321021824 utils.py:1231] [103300] l2_params = 238.79657763825986
+I1205 09:42:16.328536 137274321021824 utils.py:1231] [103300] train/loss = 1.7242542803287506
+I1205 09:42:16.328651 137274321021824 utils.py:1231] [103300] l2_grads = 2.678133964538574
+I1205 09:42:16.328730 137274321021824 utils.py:1231] [103300] lr = 2.0152082262121817e-05
+I1205 09:42:16.328796 137274321021824 utils.py:1231] [103300] uptime = 647525.6911574979
+I1205 09:42:16.328859 137274321021824 utils.py:1231] [103300] examples_seen = 105779200.0
+I1205 09:42:16.328923 137274321021824 utils.py:1231] [103300] progress = 0.9173823077537898
+I1205 09:42:16.328977 137274321021824 utils.py:1231] [103300] epoch = 82.56472419286479
+I1205 09:42:16.329033 137274321021824 utils.py:1231] [103300] img/sec/core = 164.38646304168572
+I1205 09:42:16.329092 137274321021824 utils.py:1231] [103300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.83392994139277
+I1205 09:42:16.329145 137274321021824 utils.py:1231] [103300] core_hours = 179.83392994139277
+I1205 09:42:16.329210 137274321021824 train.py:125] NOTE: Steps:103300/112603 [91.7%]
+Walltime:7d11h52m (0s eval)
+ETA:16h11m
+Total train time:8d4h1m
+I1205 09:47:27.808260 137274321021824 utils.py:1231] [103350] l2_params = 238.78365856366761
+I1205 09:47:27.808456 137274321021824 utils.py:1231] [103350] train/loss = 2.232558697462082
+I1205 09:47:27.808560 137274321021824 utils.py:1231] [103350] l2_grads = 2.6421420574188232
+I1205 09:47:27.808632 137274321021824 utils.py:1231] [103350] lr = 1.9937515599388914e-05
+I1205 09:47:27.808692 137274321021824 utils.py:1231] [103350] uptime = 647837.171053357
+I1205 09:47:27.808765 137274321021824 utils.py:1231] [103350] examples_seen = 105830400.0
+I1205 09:47:27.808838 137274321021824 utils.py:1231] [103350] progress = 0.9178263456568653
+I1205 09:47:27.808923 137274321021824 utils.py:1231] [103350] epoch = 82.60468775733375
+I1205 09:47:27.808991 137274321021824 utils.py:1231] [103350] img/sec/core = 164.3765799355889
+I1205 09:47:27.809064 137274321021824 utils.py:1231] [103350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 179.92045213468694
+I1205 09:47:27.809128 137274321021824 utils.py:1231] [103350] core_hours = 179.92045213468694
+I1205 09:47:27.809209 137274321021824 train.py:125] NOTE: Steps:103350/112603 [91.8%]
+Walltime:7d11h57m (0s eval)
+ETA:16h6m
+Total train time:8d4h1m
+I1205 09:52:39.603596 137274321021824 utils.py:1231] [103400] l2_params = 238.77187861986596
+I1205 09:52:39.603858 137274321021824 utils.py:1231] [103400] train/loss = 1.5329100489616394
+I1205 09:52:39.603984 137274321021824 utils.py:1231] [103400] l2_grads = 2.7104597091674805
+I1205 09:52:39.604068 137274321021824 utils.py:1231] [103400] lr = 1.9724074104518298e-05
+I1205 09:52:39.604141 137274321021824 utils.py:1231] [103400] uptime = 648148.966502839
+I1205 09:52:39.604218 137274321021824 utils.py:1231] [103400] examples_seen = 105881600.0
+I1205 09:52:39.604281 137274321021824 utils.py:1231] [103400] progress = 0.9182703835599407
+I1205 09:52:39.604344 137274321021824 utils.py:1231] [103400] epoch = 82.6446513218027
+I1205 09:52:39.604421 137274321021824 utils.py:1231] [103400] img/sec/core = 164.21022207044277
+I1205 09:52:39.604483 137274321021824 utils.py:1231] [103400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.00706198176528
+I1205 09:52:39.604548 137274321021824 utils.py:1231] [103400] core_hours = 180.00706198176528
+I1205 09:52:39.604623 137274321021824 train.py:125] NOTE: Steps:103400/112603 [91.8%]
+Walltime:7d12h2m (0s eval)
+ETA:16h1m
+Total train time:8d4h1m
+I1205 09:57:51.261585 137274321021824 utils.py:1231] [103450] l2_params = 238.76127255176524
+I1205 09:57:51.261852 137274321021824 utils.py:1231] [103450] train/loss = 1.4466798305511475
+I1205 09:57:51.261989 137274321021824 utils.py:1231] [103450] l2_grads = 2.6909003257751465
+I1205 09:57:51.262063 137274321021824 utils.py:1231] [103450] lr = 1.9511758277772885e-05
+I1205 09:57:51.262116 137274321021824 utils.py:1231] [103450] uptime = 648460.6244777429
+I1205 09:57:51.262183 137274321021824 utils.py:1231] [103450] examples_seen = 105932800.0
+I1205 09:57:51.262235 137274321021824 utils.py:1231] [103450] progress = 0.9187144214630161
+I1205 09:57:51.262293 137274321021824 utils.py:1231] [103450] epoch = 82.68461488627166
+I1205 09:57:51.262346 137274321021824 utils.py:1231] [103450] img/sec/core = 164.28265638246995
+I1205 09:57:51.262403 137274321021824 utils.py:1231] [103450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.0936336414608
+I1205 09:57:51.262455 137274321021824 utils.py:1231] [103450] core_hours = 180.0936336414608
+I1205 09:57:51.262517 137274321021824 train.py:125] NOTE: Steps:103450/112603 [91.9%]
+Walltime:7d12h7m (0s eval)
+ETA:15h56m
+Total train time:8d4h1m
+I1205 10:03:03.053158 137274321021824 utils.py:1231] [103500] l2_params = 238.7481252315821
+I1205 10:03:03.053421 137274321021824 utils.py:1231] [103500] train/loss = 1.529567614197731
+I1205 10:03:03.053542 137274321021824 utils.py:1231] [103500] l2_grads = 2.8917346000671387
+I1205 10:03:03.053630 137274321021824 utils.py:1231] [103500] lr = 1.930056861677724e-05
+I1205 10:03:03.053703 137274321021824 utils.py:1231] [103500] uptime = 648772.416063879
+I1205 10:03:03.053763 137274321021824 utils.py:1231] [103500] examples_seen = 105984000.0
+I1205 10:03:03.053816 137274321021824 utils.py:1231] [103500] progress = 0.9191584593660915
+I1205 10:03:03.053865 137274321021824 utils.py:1231] [103500] epoch = 82.72457845074061
+I1205 10:03:03.053930 137274321021824 utils.py:1231] [103500] img/sec/core = 164.21225676581068
+I1205 10:03:03.053987 137274321021824 utils.py:1231] [103500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.1802424153875
+I1205 10:03:03.054038 137274321021824 utils.py:1231] [103500] core_hours = 180.1802424153875
+I1205 10:03:03.054097 137274321021824 train.py:125] NOTE: Steps:103500/112603 [91.9%]
+Walltime:7d12h12m (0s eval)
+ETA:15h50m
+Total train time:8d4h1m
+I1205 10:08:14.844167 137274321021824 utils.py:1231] [103550] l2_params = 238.73623284109425
+I1205 10:08:14.844463 137274321021824 utils.py:1231] [103550] train/loss = 1.474880337715149
+I1205 10:08:14.844645 137274321021824 utils.py:1231] [103550] l2_grads = 2.794769287109375
+I1205 10:08:14.844741 137274321021824 utils.py:1231] [103550] lr = 1.9090505616516854e-05
+I1205 10:08:14.844803 137274321021824 utils.py:1231] [103550] uptime = 649084.207164468
+I1205 10:08:14.844862 137274321021824 utils.py:1231] [103550] examples_seen = 106035200.0
+I1205 10:08:14.844948 137274321021824 utils.py:1231] [103550] progress = 0.9196024972691669
+I1205 10:08:14.845004 137274321021824 utils.py:1231] [103550] epoch = 82.76454201520957
+I1205 10:08:14.845060 137274321021824 utils.py:1231] [103550] img/sec/core = 164.21251249083988
+I1205 10:08:14.845122 137274321021824 utils.py:1231] [103550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.26685105444
+I1205 10:08:14.845175 137274321021824 utils.py:1231] [103550] core_hours = 180.26685105444
+I1205 10:08:14.845237 137274321021824 train.py:125] NOTE: Steps:103550/112603 [92.0%]
+Walltime:7d12h18m (0s eval)
+ETA:15h45m
+Total train time:8d4h1m
+I1205 10:13:26.635634 137274321021824 utils.py:1231] [103600] l2_params = 238.72391833113636
+I1205 10:13:26.635864 137274321021824 utils.py:1231] [103600] train/loss = 2.225042313337326
+I1205 10:13:26.636012 137274321021824 utils.py:1231] [103600] l2_grads = 2.61191463470459
+I1205 10:13:26.636114 137274321021824 utils.py:1231] [103600] lr = 1.8881569769336015e-05
+I1205 10:13:26.636192 137274321021824 utils.py:1231] [103600] uptime = 649395.998553286
+I1205 10:13:26.636269 137274321021824 utils.py:1231] [103600] examples_seen = 106086400.0
+I1205 10:13:26.636342 137274321021824 utils.py:1231] [103600] progress = 0.9200465351722423
+I1205 10:13:26.636408 137274321021824 utils.py:1231] [103600] epoch = 82.80450557967853
+I1205 10:13:26.636476 137274321021824 utils.py:1231] [103600] img/sec/core = 164.21236068804788
+I1205 10:13:26.636537 137274321021824 utils.py:1231] [103600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.3534597735561
+I1205 10:13:26.636600 137274321021824 utils.py:1231] [103600] core_hours = 180.3534597735561
+I1205 10:13:26.636687 137274321021824 train.py:125] NOTE: Steps:103600/112603 [92.0%]
+Walltime:7d12h23m (0s eval)
+ETA:15h40m
+Total train time:8d4h1m
+I1205 10:18:38.401195 137274321021824 utils.py:1231] [103650] l2_params = 238.7120008075087
+I1205 10:18:38.401403 137274321021824 utils.py:1231] [103650] train/loss = 1.8204628974199295
+I1205 10:18:38.401502 137274321021824 utils.py:1231] [103650] l2_grads = 2.574800729751587
+I1205 10:18:38.401583 137274321021824 utils.py:1231] [103650] lr = 1.867376156493759e-05
+I1205 10:18:38.401644 137274321021824 utils.py:1231] [103650] uptime = 649707.764005007
+I1205 10:18:38.401701 137274321021824 utils.py:1231] [103650] examples_seen = 106137600.0
+I1205 10:18:38.401757 137274321021824 utils.py:1231] [103650] progress = 0.9204905730753177
+I1205 10:18:38.401816 137274321021824 utils.py:1231] [103650] epoch = 82.84446914414748
+I1205 10:18:38.401889 137274321021824 utils.py:1231] [103650] img/sec/core = 164.22602221433368
+I1205 10:18:38.401960 137274321021824 utils.py:1231] [103650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.44006128792304
+I1205 10:18:38.402012 137274321021824 utils.py:1231] [103650] core_hours = 180.44006128792304
+I1205 10:18:38.402070 137274321021824 train.py:125] NOTE: Steps:103650/112603 [92.0%]
+Walltime:7d12h28m (0s eval)
+ETA:15h35m
+Total train time:8d4h1m
+I1205 10:23:50.191999 137274321021824 utils.py:1231] [103700] l2_params = 238.70071833801194
+I1205 10:23:50.192220 137274321021824 utils.py:1231] [103700] train/loss = 2.393092006444931
+I1205 10:23:50.192334 137274321021824 utils.py:1231] [103700] l2_grads = 2.5734152793884277
+I1205 10:23:50.192410 137274321021824 utils.py:1231] [103700] lr = 1.8467081490381257e-05
+I1205 10:23:50.192473 137274321021824 utils.py:1231] [103700] uptime = 650019.554834218
+I1205 10:23:50.192547 137274321021824 utils.py:1231] [103700] examples_seen = 106188800.0
+I1205 10:23:50.192626 137274321021824 utils.py:1231] [103700] progress = 0.9209346109783931
+I1205 10:23:50.192702 137274321021824 utils.py:1231] [103700] epoch = 82.88443270861644
+I1205 10:23:50.192772 137274321021824 utils.py:1231] [103700] img/sec/core = 164.21265541889895
+I1205 10:23:50.192836 137274321021824 utils.py:1231] [103700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.52666985159277
+I1205 10:23:50.192910 137274321021824 utils.py:1231] [103700] core_hours = 180.52666985159277
+I1205 10:23:50.192978 137274321021824 train.py:125] NOTE: Steps:103700/112603 [92.1%]
+Walltime:7d12h33m (0s eval)
+ETA:15h29m
+Total train time:8d4h1m
+I1205 10:29:01.979883 137274321021824 utils.py:1231] [103750] l2_params = 238.6883114544754
+I1205 10:29:01.980116 137274321021824 utils.py:1231] [103750] train/loss = 2.4725842475891113
+I1205 10:29:01.980228 137274321021824 utils.py:1231] [103750] l2_grads = 2.6509718894958496
+I1205 10:29:01.980307 137274321021824 utils.py:1231] [103750] lr = 1.8261530030082602e-05
+I1205 10:29:01.980365 137274321021824 utils.py:1231] [103750] uptime = 650331.3427278909
+I1205 10:29:01.980423 137274321021824 utils.py:1231] [103750] examples_seen = 106240000.0
+I1205 10:29:01.980471 137274321021824 utils.py:1231] [103750] progress = 0.9213786488814685
+I1205 10:29:01.980518 137274321021824 utils.py:1231] [103750] epoch = 82.92439627308539
+I1205 10:29:01.980568 137274321021824 utils.py:1231] [103750] img/sec/core = 164.21420151004324
+I1205 10:29:01.980630 137274321021824 utils.py:1231] [103750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.61327759983527
+I1205 10:29:01.980689 137274321021824 utils.py:1231] [103750] core_hours = 180.61327759983527
+I1205 10:29:01.980749 137274321021824 train.py:125] NOTE: Steps:103750/112603 [92.1%]
+Walltime:7d12h38m (0s eval)
+ETA:15h24m
+Total train time:8d4h1m
+I1205 10:34:13.780062 137274321021824 utils.py:1231] [103800] l2_params = 238.67762936249795
+I1205 10:34:13.780268 137274321021824 utils.py:1231] [103800] train/loss = 2.4387297928333282
+I1205 10:34:13.780370 137274321021824 utils.py:1231] [103800] l2_grads = 2.564509868621826
+I1205 10:34:13.780446 137274321021824 utils.py:1231] [103800] lr = 1.8057107665812057e-05
+I1205 10:34:13.780504 137274321021824 utils.py:1231] [103800] uptime = 650643.142866123
+I1205 10:34:13.780562 137274321021824 utils.py:1231] [103800] examples_seen = 106291200.0
+I1205 10:34:13.780617 137274321021824 utils.py:1231] [103800] progress = 0.9218226867845439
+I1205 10:34:13.780676 137274321021824 utils.py:1231] [103800] epoch = 82.96435983755435
+I1205 10:34:13.780732 137274321021824 utils.py:1231] [103800] img/sec/core = 164.20775273002948
+I1205 10:34:13.780793 137274321021824 utils.py:1231] [103800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.69988874934415
+I1205 10:34:13.780852 137274321021824 utils.py:1231] [103800] core_hours = 180.69988874934415
+I1205 10:34:13.780937 137274321021824 train.py:125] NOTE: Steps:103800/112603 [92.2%]
+Walltime:7d12h44m (0s eval)
+ETA:15h19m
+Total train time:8d4h1m
+I1205 10:39:25.564395 137274321021824 utils.py:1231] [103850] l2_params = 238.6663579829469
+I1205 10:39:25.564619 137274321021824 utils.py:1231] [103850] train/loss = 3.670030176639557
+I1205 10:39:25.564712 137274321021824 utils.py:1231] [103850] l2_grads = 2.922922372817993
+I1205 10:39:25.564776 137274321021824 utils.py:1231] [103850] lr = 1.7853814876693707e-05
+I1205 10:39:25.564829 137274321021824 utils.py:1231] [103850] uptime = 650954.9271909309
+I1205 10:39:25.564885 137274321021824 utils.py:1231] [103850] examples_seen = 106342400.0
+I1205 10:39:25.564936 137274321021824 utils.py:1231] [103850] progress = 0.9222667246876194
+I1205 10:39:25.564983 137274321021824 utils.py:1231] [103850] epoch = 83.00432340202332
+I1205 10:39:25.565033 137274321021824 utils.py:1231] [103850] img/sec/core = 164.21608120145558
+I1205 10:39:25.565088 137274321021824 utils.py:1231] [103850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.78649550623524
+I1205 10:39:25.565138 137274321021824 utils.py:1231] [103850] core_hours = 180.78649550623524
+I1205 10:39:25.565196 137274321021824 train.py:125] NOTE: Steps:103850/112603 [92.2%]
+Walltime:7d12h49m (0s eval)
+ETA:15h14m
+Total train time:8d4h1m
+I1205 10:44:37.350787 137274321021824 utils.py:1231] [103900] l2_params = 238.65553306735643
+I1205 10:44:37.351072 137274321021824 utils.py:1231] [103900] train/loss = 3.780633717775345
+I1205 10:44:37.351204 137274321021824 utils.py:1231] [103900] l2_grads = 3.005486011505127
+I1205 10:44:37.351298 137274321021824 utils.py:1231] [103900] lr = 1.765165213920418e-05
+I1205 10:44:37.351370 137274321021824 utils.py:1231] [103900] uptime = 651266.713728361
+I1205 10:44:37.351450 137274321021824 utils.py:1231] [103900] examples_seen = 106393600.0
+I1205 10:44:37.351525 137274321021824 utils.py:1231] [103900] progress = 0.9227107625906947
+I1205 10:44:37.351613 137274321021824 utils.py:1231] [103900] epoch = 83.04428696649227
+I1205 10:44:37.351739 137274321021824 utils.py:1231] [103900] img/sec/core = 164.21491582680142
+I1205 10:44:37.351814 137274321021824 utils.py:1231] [103900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.8731028777436
+I1205 10:44:37.351896 137274321021824 utils.py:1231] [103900] core_hours = 180.8731028777436
+I1205 10:44:37.352008 137274321021824 train.py:125] NOTE: Steps:103900/112603 [92.3%]
+Walltime:7d12h54m (0s eval)
+ETA:15h9m
+Total train time:8d4h1m
+I1205 10:49:49.136237 137274321021824 utils.py:1231] [103950] l2_params = 238.6445084783299
+I1205 10:49:49.136469 137274321021824 utils.py:1231] [103950] train/loss = 1.4802922755479813
+I1205 10:49:49.136604 137274321021824 utils.py:1231] [103950] l2_grads = 2.8503026962280273
+I1205 10:49:49.136710 137274321021824 utils.py:1231] [103950] lr = 1.7450619927171243e-05
+I1205 10:49:49.136795 137274321021824 utils.py:1231] [103950] uptime = 651578.499151755
+I1205 10:49:49.136875 137274321021824 utils.py:1231] [103950] examples_seen = 106444800.0
+I1205 10:49:49.136960 137274321021824 utils.py:1231] [103950] progress = 0.9231548004937702
+I1205 10:49:49.137036 137274321021824 utils.py:1231] [103950] epoch = 83.08425053096123
+I1205 10:49:49.137112 137274321021824 utils.py:1231] [103950] img/sec/core = 164.2155025807261
+I1205 10:49:49.137182 137274321021824 utils.py:1231] [103950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 180.9597099397975
+I1205 10:49:49.137255 137274321021824 utils.py:1231] [103950] core_hours = 180.9597099397975
+I1205 10:49:49.137333 137274321021824 train.py:125] NOTE: Steps:103950/112603 [92.3%]
+Walltime:7d12h59m (0s eval)
+ETA:15h3m
+Total train time:8d4h1m
+I1205 10:55:00.965121 137274321021824 utils.py:1231] [104000] l2_params = 238.63356527240046
+I1205 10:55:00.965316 137274321021824 utils.py:1231] [104000] train/loss = 3.3555429875850677
+I1205 10:55:00.965410 137274321021824 utils.py:1231] [104000] l2_grads = 2.7629523277282715
+I1205 10:55:00.965471 137274321021824 utils.py:1231] [104000] lr = 1.725071871177315e-05
+I1205 10:55:00.965521 137274321021824 utils.py:1231] [104000] uptime = 651890.3278837809
+I1205 10:55:00.965571 137274321021824 utils.py:1231] [104000] examples_seen = 106496000.0
+I1205 10:55:00.965617 137274321021824 utils.py:1231] [104000] progress = 0.9235988383968455
+I1205 10:55:00.965663 137274321021824 utils.py:1231] [104000] epoch = 83.12421409543018
+I1205 10:55:00.965712 137274321021824 utils.py:1231] [104000] img/sec/core = 164.19269535349662
+I1205 10:55:00.965766 137274321021824 utils.py:1231] [104000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.0463290320269
+I1205 10:55:00.965816 137274321021824 utils.py:1231] [104000] core_hours = 181.0463290320269
+I1205 10:55:00.965873 137274321021824 train.py:125] NOTE: Steps:104000/112603 [92.4%]
+Walltime:7d13h4m (0s eval)
+ETA:14h58m
+Total train time:8d4h1m
+I1205 11:00:13.079392 137274321021824 utils.py:1231] [104050] l2_params = 238.62215277310528
+I1205 11:00:13.079597 137274321021824 utils.py:1231] [104050] train/loss = 1.7171233147382736
+I1205 11:00:13.079697 137274321021824 utils.py:1231] [104050] l2_grads = 2.7809367179870605
+I1205 11:00:13.079769 137274321021824 utils.py:1231] [104050] lr = 1.705194896153734e-05
+I1205 11:00:13.079834 137274321021824 utils.py:1231] [104050] uptime = 652202.442195521
+I1205 11:00:13.079895 137274321021824 utils.py:1231] [104050] examples_seen = 106547200.0
+I1205 11:00:13.079953 137274321021824 utils.py:1231] [104050] progress = 0.924042876299921
+I1205 11:00:13.080020 137274321021824 utils.py:1231] [104050] epoch = 83.16417765989914
+I1205 11:00:13.080077 137274321021824 utils.py:1231] [104050] img/sec/core = 164.04246160504573
+I1205 11:00:13.080139 137274321021824 utils.py:1231] [104050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.13302745195472
+I1205 11:00:13.080207 137274321021824 utils.py:1231] [104050] core_hours = 181.13302745195472
+I1205 11:00:13.080271 137274321021824 train.py:125] NOTE: Steps:104050/112603 [92.4%]
+Walltime:7d13h10m (0s eval)
+ETA:14h53m
+Total train time:8d4h1m
+I1205 11:05:24.849386 137274321021824 utils.py:1231] [104100] l2_params = 238.61110783399207
+I1205 11:05:24.849685 137274321021824 utils.py:1231] [104100] train/loss = 1.375460535287857
+I1205 11:05:24.849838 137274321021824 utils.py:1231] [104100] l2_grads = 2.7916059494018555
+I1205 11:05:24.849961 137274321021824 utils.py:1231] [104100] lr = 1.6854311142339163e-05
+I1205 11:05:24.850031 137274321021824 utils.py:1231] [104100] uptime = 652514.2123908229
+I1205 11:05:24.850104 137274321021824 utils.py:1231] [104100] examples_seen = 106598400.0
+I1205 11:05:24.850171 137274321021824 utils.py:1231] [104100] progress = 0.9244869142029963
+I1205 11:05:24.850247 137274321021824 utils.py:1231] [104100] epoch = 83.2041412243681
+I1205 11:05:24.850315 137274321021824 utils.py:1231] [104100] img/sec/core = 164.2235235167659
+I1205 11:05:24.850395 137274321021824 utils.py:1231] [104100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.21963028398304
+I1205 11:05:24.850468 137274321021824 utils.py:1231] [104100] core_hours = 181.21963028398304
+I1205 11:05:24.850535 137274321021824 train.py:125] NOTE: Steps:104100/112603 [92.4%]
+Walltime:7d13h15m (0s eval)
+ETA:14h48m
+Total train time:8d4h1m
+I1205 11:10:36.634887 137274321021824 utils.py:1231] [104150] l2_params = 238.6015155731362
+I1205 11:10:36.635100 137274321021824 utils.py:1231] [104150] train/loss = 2.822864681482315
+I1205 11:10:36.635232 137274321021824 utils.py:1231] [104150] l2_grads = 2.572439670562744
+I1205 11:10:36.635328 137274321021824 utils.py:1231] [104150] lr = 1.6657805717401272e-05
+I1205 11:10:36.635416 137274321021824 utils.py:1231] [104150] uptime = 652825.997777051
+I1205 11:10:36.635473 137274321021824 utils.py:1231] [104150] examples_seen = 106649600.0
+I1205 11:10:36.635529 137274321021824 utils.py:1231] [104150] progress = 0.9249309521060718
+I1205 11:10:36.635584 137274321021824 utils.py:1231] [104150] epoch = 83.24410478883705
+I1205 11:10:36.635641 137274321021824 utils.py:1231] [104150] img/sec/core = 164.21552215586664
+I1205 11:10:36.635713 137274321021824 utils.py:1231] [104150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.30623733571304
+I1205 11:10:36.635769 137274321021824 utils.py:1231] [104150] core_hours = 181.30623733571304
+I1205 11:10:36.635835 137274321021824 train.py:125] NOTE: Steps:104150/112603 [92.5%]
+Walltime:7d13h20m (0s eval)
+ETA:14h42m
+Total train time:8d4h1m
+I1205 11:15:48.415227 137274321021824 utils.py:1231] [104200] l2_params = 238.5914124116384
+I1205 11:15:48.415482 137274321021824 utils.py:1231] [104200] train/loss = 2.0610059201717377
+I1205 11:15:48.415672 137274321021824 utils.py:1231] [104200] l2_grads = 2.8070499897003174
+I1205 11:15:48.415758 137274321021824 utils.py:1231] [104200] lr = 1.646243314729195e-05
+I1205 11:15:48.415822 137274321021824 utils.py:1231] [104200] uptime = 653137.778182424
+I1205 11:15:48.415892 137274321021824 utils.py:1231] [104200] examples_seen = 106700800.0
+I1205 11:15:48.415953 137274321021824 utils.py:1231] [104200] progress = 0.9253749900091471
+I1205 11:15:48.416010 137274321021824 utils.py:1231] [104200] epoch = 83.28406835330601
+I1205 11:15:48.416072 137274321021824 utils.py:1231] [104200] img/sec/core = 164.21814558465772
+I1205 11:15:48.416134 137274321021824 utils.py:1231] [104200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.39284300387223
+I1205 11:15:48.416191 137274321021824 utils.py:1231] [104200] core_hours = 181.39284300387223
+I1205 11:15:48.416258 137274321021824 train.py:125] NOTE: Steps:104200/112603 [92.5%]
+Walltime:7d13h25m (0s eval)
+ETA:14h37m
+Total train time:8d4h1m
+I1205 11:21:00.198668 137274321021824 utils.py:1231] [104250] l2_params = 238.5806965992582
+I1205 11:21:00.198956 137274321021824 utils.py:1231] [104250] train/loss = 3.7752205431461334
+I1205 11:21:00.199145 137274321021824 utils.py:1231] [104250] l2_grads = 3.059600591659546
+I1205 11:21:00.199221 137274321021824 utils.py:1231] [104250] lr = 1.6268193889924224e-05
+I1205 11:21:00.199283 137274321021824 utils.py:1231] [104250] uptime = 653449.561643855
+I1205 11:21:00.199345 137274321021824 utils.py:1231] [104250] examples_seen = 106752000.0
+I1205 11:21:00.199404 137274321021824 utils.py:1231] [104250] progress = 0.9258190279122226
+I1205 11:21:00.199463 137274321021824 utils.py:1231] [104250] epoch = 83.32403191777496
+I1205 11:21:00.199522 137274321021824 utils.py:1231] [104250] img/sec/core = 164.2165359413821
+I1205 11:21:00.199588 137274321021824 utils.py:1231] [104250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.47944952093636
+I1205 11:21:00.199648 137274321021824 utils.py:1231] [104250] core_hours = 181.47944952093636
+I1205 11:21:00.199715 137274321021824 train.py:125] NOTE: Steps:104250/112603 [92.6%]
+Walltime:7d13h30m (0s eval)
+ETA:14h32m
+Total train time:8d4h1m
+I1205 11:26:11.999893 137274321021824 utils.py:1231] [104300] l2_params = 238.57014465850793
+I1205 11:26:12.000094 137274321021824 utils.py:1231] [104300] train/loss = 2.029543772339821
+I1205 11:26:12.000191 137274321021824 utils.py:1231] [104300] l2_grads = 2.6836395263671875
+I1205 11:26:12.000266 137274321021824 utils.py:1231] [104300] lr = 1.6075088400555205e-05
+I1205 11:26:12.000325 137274321021824 utils.py:1231] [104300] uptime = 653761.362687152
+I1205 11:26:12.000384 137274321021824 utils.py:1231] [104300] examples_seen = 106803200.0
+I1205 11:26:12.000438 137274321021824 utils.py:1231] [104300] progress = 0.926263065815298
+I1205 11:26:12.000497 137274321021824 utils.py:1231] [104300] epoch = 83.36399548224392
+I1205 11:26:12.000553 137274321021824 utils.py:1231] [104300] img/sec/core = 164.20727608413614
+I1205 11:26:12.000614 137274321021824 utils.py:1231] [104300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.5660609218522
+I1205 11:26:12.000671 137274321021824 utils.py:1231] [104300] core_hours = 181.5660609218522
+I1205 11:26:12.000736 137274321021824 train.py:125] NOTE: Steps:104300/112603 [92.6%]
+Walltime:7d13h36m (0s eval)
+ETA:14h27m
+Total train time:8d4h1m
+I1205 11:31:23.793507 137274321021824 utils.py:1231] [104350] l2_params = 238.5604241707834
+I1205 11:31:23.793768 137274321021824 utils.py:1231] [104350] train/loss = 3.7174205780029297
+I1205 11:31:23.793877 137274321021824 utils.py:1231] [104350] l2_grads = 2.941331148147583
+I1205 11:31:23.793966 137274321021824 utils.py:1231] [104350] lr = 1.5883117131784358e-05
+I1205 11:31:23.794027 137274321021824 utils.py:1231] [104350] uptime = 654073.156389521
+I1205 11:31:23.794086 137274321021824 utils.py:1231] [104350] examples_seen = 106854400.0
+I1205 11:31:23.794136 137274321021824 utils.py:1231] [104350] progress = 0.9267071037183734
+I1205 11:31:23.794193 137274321021824 utils.py:1231] [104350] epoch = 83.40395904671288
+I1205 11:31:23.794246 137274321021824 utils.py:1231] [104350] img/sec/core = 164.21114221034009
+I1205 11:31:23.794310 137274321021824 utils.py:1231] [104350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.65267028362138
+I1205 11:31:23.794366 137274321021824 utils.py:1231] [104350] core_hours = 181.65267028362138
+I1205 11:31:23.794431 137274321021824 train.py:125] NOTE: Steps:104350/112603 [92.7%]
+Walltime:7d13h41m (0s eval)
+ETA:14h22m
+Total train time:8d4h1m
+I1205 11:36:35.597469 137274321021824 utils.py:1231] [104400] l2_params = 238.55094218120996
+I1205 11:36:35.597687 137274321021824 utils.py:1231] [104400] train/loss = 1.8593178242444992
+I1205 11:36:35.597781 137274321021824 utils.py:1231] [104400] l2_grads = 2.756946325302124
+I1205 11:36:35.597845 137274321021824 utils.py:1231] [104400] lr = 1.5692280533553088e-05
+I1205 11:36:35.597910 137274321021824 utils.py:1231] [104400] uptime = 654384.960271323
+I1205 11:36:35.597965 137274321021824 utils.py:1231] [104400] examples_seen = 106905600.0
+I1205 11:36:35.598016 137274321021824 utils.py:1231] [104400] progress = 0.9271511416214488
+I1205 11:36:35.598067 137274321021824 utils.py:1231] [104400] epoch = 83.44392261118183
+I1205 11:36:35.598119 137274321021824 utils.py:1231] [104400] img/sec/core = 164.20578122407863
+I1205 11:36:35.598178 137274321021824 utils.py:1231] [104400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.73928247301083
+I1205 11:36:35.598230 137274321021824 utils.py:1231] [104400] core_hours = 181.73928247301083
+I1205 11:36:35.598293 137274321021824 train.py:125] NOTE: Steps:104400/112603 [92.7%]
+Walltime:7d13h46m (0s eval)
+ETA:14h16m
+Total train time:8d4h1m
+I1205 11:41:47.278051 137274321021824 utils.py:1231] [104450] l2_params = 238.5412168229677
+I1205 11:41:47.278331 137274321021824 utils.py:1231] [104450] train/loss = 3.684897691011429
+I1205 11:41:47.278504 137274321021824 utils.py:1231] [104450] l2_grads = 2.9137022495269775
+I1205 11:41:47.278577 137274321021824 utils.py:1231] [104450] lr = 1.550257905314306e-05
+I1205 11:41:47.278640 137274321021824 utils.py:1231] [104450] uptime = 654696.641002072
+I1205 11:41:47.278710 137274321021824 utils.py:1231] [104450] examples_seen = 106956800.0
+I1205 11:41:47.278764 137274321021824 utils.py:1231] [104450] progress = 0.9275951795245242
+I1205 11:41:47.278822 137274321021824 utils.py:1231] [104450] epoch = 83.4838861756508
+I1205 11:41:47.278890 137274321021824 utils.py:1231] [104450] img/sec/core = 164.27066208730653
+I1205 11:41:47.278957 137274321021824 utils.py:1231] [104450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.82586045377445
+I1205 11:41:47.279013 137274321021824 utils.py:1231] [104450] core_hours = 181.82586045377445
+I1205 11:41:47.279077 137274321021824 train.py:125] NOTE: Steps:104450/112603 [92.8%]
+Walltime:7d13h51m (0s eval)
+ETA:14h11m
+Total train time:8d4h1m
+I1205 11:46:59.067798 137274321021824 utils.py:1231] [104500] l2_params = 238.5317500166798
+I1205 11:46:59.068011 137274321021824 utils.py:1231] [104500] train/loss = 3.7317376732826233
+I1205 11:46:59.068124 137274321021824 utils.py:1231] [104500] l2_grads = 2.935452461242676
+I1205 11:46:59.068200 137274321021824 utils.py:1231] [104500] lr = 1.5314013135175506e-05
+I1205 11:46:59.068263 137274321021824 utils.py:1231] [104500] uptime = 655008.430624088
+I1205 11:46:59.068325 137274321021824 utils.py:1231] [104500] examples_seen = 107008000.0
+I1205 11:46:59.068391 137274321021824 utils.py:1231] [104500] progress = 0.9280392174275996
+I1205 11:46:59.068449 137274321021824 utils.py:1231] [104500] epoch = 83.52384974011974
+I1205 11:46:59.068507 137274321021824 utils.py:1231] [104500] img/sec/core = 164.2132912216353
+I1205 11:46:59.068570 137274321021824 utils.py:1231] [104500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.91246868211223
+I1205 11:46:59.068662 137274321021824 utils.py:1231] [104500] core_hours = 181.91246868211223
+I1205 11:46:59.068736 137274321021824 train.py:125] NOTE: Steps:104500/112603 [92.8%]
+Walltime:7d13h56m (0s eval)
+ETA:14h6m
+Total train time:8d4h1m
+I1205 11:52:10.857507 137274321021824 utils.py:1231] [104550] l2_params = 238.52179849246494
+I1205 11:52:10.857721 137274321021824 utils.py:1231] [104550] train/loss = 3.5460202395915985
+I1205 11:52:10.857819 137274321021824 utils.py:1231] [104550] l2_grads = 2.799109697341919
+I1205 11:52:10.857902 137274321021824 utils.py:1231] [104550] lr = 1.5126583221610374e-05
+I1205 11:52:10.857958 137274321021824 utils.py:1231] [104550] uptime = 655320.220319654
+I1205 11:52:10.858013 137274321021824 utils.py:1231] [104550] examples_seen = 107059200.0
+I1205 11:52:10.858063 137274321021824 utils.py:1231] [104550] progress = 0.928483255330675
+I1205 11:52:10.858114 137274321021824 utils.py:1231] [104550] epoch = 83.5638133045887
+I1205 11:52:10.858165 137274321021824 utils.py:1231] [104550] img/sec/core = 164.21325248438848
+I1205 11:52:10.858223 137274321021824 utils.py:1231] [104550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 181.99907693088056
+I1205 11:52:10.858275 137274321021824 utils.py:1231] [104550] core_hours = 181.99907693088056
+I1205 11:52:10.858337 137274321021824 train.py:125] NOTE: Steps:104550/112603 [92.8%]
+Walltime:7d14h2m (0s eval)
+ETA:14h1m
+Total train time:8d4h1m
+I1205 11:57:22.627159 137274321021824 utils.py:1231] [104600] l2_params = 238.51141127079532
+I1205 11:57:22.627372 137274321021824 utils.py:1231] [104600] train/loss = 1.5627349317073822
+I1205 11:57:22.627472 137274321021824 utils.py:1231] [104600] l2_grads = 2.854505777359009
+I1205 11:57:22.627551 137274321021824 utils.py:1231] [104600] lr = 1.4940289751744715e-05
+I1205 11:57:22.627616 137274321021824 utils.py:1231] [104600] uptime = 655631.989975466
+I1205 11:57:22.627671 137274321021824 utils.py:1231] [104600] examples_seen = 107110400.0
+I1205 11:57:22.627718 137274321021824 utils.py:1231] [104600] progress = 0.9289272932337505
+I1205 11:57:22.627767 137274321021824 utils.py:1231] [104600] epoch = 83.60377686905767
+I1205 11:57:22.627817 137274321021824 utils.py:1231] [104600] img/sec/core = 164.22380769109623
+I1205 11:57:22.627874 137274321021824 utils.py:1231] [104600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.08567961305056
+I1205 11:57:22.627929 137274321021824 utils.py:1231] [104600] core_hours = 182.08567961305056
+I1205 11:57:22.627989 137274321021824 train.py:125] NOTE: Steps:104600/112603 [92.9%]
+Walltime:7d14h7m (0s eval)
+ETA:13h55m
+Total train time:8d4h1m
+I1205 12:02:34.413223 137274321021824 utils.py:1231] [104650] l2_params = 238.50241246392852
+I1205 12:02:34.413426 137274321021824 utils.py:1231] [104650] train/loss = 3.400609701871872
+I1205 12:02:34.413513 137274321021824 utils.py:1231] [104650] l2_grads = 2.7178890705108643
+I1205 12:02:34.413575 137274321021824 utils.py:1231] [104650] lr = 1.4755133162212328e-05
+I1205 12:02:34.413628 137274321021824 utils.py:1231] [104650] uptime = 655943.775990652
+I1205 12:02:34.413678 137274321021824 utils.py:1231] [104650] examples_seen = 107161600.0
+I1205 12:02:34.413725 137274321021824 utils.py:1231] [104650] progress = 0.9293713311368258
+I1205 12:02:34.413773 137274321021824 utils.py:1231] [104650] epoch = 83.64374043352662
+I1205 12:02:34.413821 137274321021824 utils.py:1231] [104650] img/sec/core = 164.21519088807463
+I1205 12:02:34.413880 137274321021824 utils.py:1231] [104650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.17228683949114
+I1205 12:02:34.413936 137274321021824 utils.py:1231] [104650] core_hours = 182.17228683949114
+I1205 12:02:34.413995 137274321021824 train.py:125] NOTE: Steps:104650/112603 [92.9%]
+Walltime:7d14h12m (0s eval)
+ETA:13h50m
+Total train time:8d4h1m
+I1205 12:07:46.199467 137274321021824 utils.py:1231] [104700] l2_params = 238.49280814107794
+I1205 12:07:46.199687 137274321021824 utils.py:1231] [104700] train/loss = 2.7855373322963715
+I1205 12:07:46.199795 137274321021824 utils.py:1231] [104700] l2_grads = 2.7615175247192383
+I1205 12:07:46.199887 137274321021824 utils.py:1231] [104700] lr = 1.4571113886982173e-05
+I1205 12:07:46.199948 137274321021824 utils.py:1231] [104700] uptime = 656255.5623095069
+I1205 12:07:46.200011 137274321021824 utils.py:1231] [104700] examples_seen = 107212800.0
+I1205 12:07:46.200066 137274321021824 utils.py:1231] [104700] progress = 0.9298153690399013
+I1205 12:07:46.200134 137274321021824 utils.py:1231] [104700] epoch = 83.68370399799558
+I1205 12:07:46.200191 137274321021824 utils.py:1231] [104700] img/sec/core = 164.2150309482582
+I1205 12:07:46.200250 137274321021824 utils.py:1231] [104700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.25889415028414
+I1205 12:07:46.200317 137274321021824 utils.py:1231] [104700] core_hours = 182.25889415028414
+I1205 12:07:46.200421 137274321021824 train.py:125] NOTE: Steps:104700/112603 [93.0%]
+Walltime:7d14h17m (0s eval)
+ETA:13h45m
+Total train time:8d4h1m
+I1205 12:12:57.987764 137274321021824 utils.py:1231] [104750] l2_params = 238.48357204098886
+I1205 12:12:57.988020 137274321021824 utils.py:1231] [104750] train/loss = 1.9985259026288986
+I1205 12:12:57.988155 137274321021824 utils.py:1231] [104750] l2_grads = 2.601783037185669
+I1205 12:12:57.988239 137274321021824 utils.py:1231] [104750] lr = 1.438823235735753e-05
+I1205 12:12:57.988307 137274321021824 utils.py:1231] [104750] uptime = 656567.350668483
+I1205 12:12:57.988365 137274321021824 utils.py:1231] [104750] examples_seen = 107264000.0
+I1205 12:12:57.988429 137274321021824 utils.py:1231] [104750] progress = 0.9302594069429767
+I1205 12:12:57.988484 137274321021824 utils.py:1231] [104750] epoch = 83.72366756246453
+I1205 12:12:57.988540 137274321021824 utils.py:1231] [104750] img/sec/core = 164.21395644196704
+I1205 12:12:57.988601 137274321021824 utils.py:1231] [104750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.34550202777748
+I1205 12:12:57.988656 137274321021824 utils.py:1231] [104750] core_hours = 182.34550202777748
+I1205 12:12:58.216718 137274321021824 train.py:125] NOTE: Steps:104750/112603 [93.0%]
+Walltime:7d14h22m (0s eval)
+ETA:13h40m
+Total train time:8d4h1m
+I1205 12:18:10.011135 137274321021824 utils.py:1231] [104800] l2_params = 238.4747088187342
+I1205 12:18:10.011434 137274321021824 utils.py:1231] [104800] train/loss = 1.4877128452062607
+I1205 12:18:10.011597 137274321021824 utils.py:1231] [104800] l2_grads = 2.5871317386627197
+I1205 12:18:10.011675 137274321021824 utils.py:1231] [104800] lr = 1.4206489001975344e-05
+I1205 12:18:10.011745 137274321021824 utils.py:1231] [104800] uptime = 656879.374106499
+I1205 12:18:10.011809 137274321021824 utils.py:1231] [104800] examples_seen = 107315200.0
+I1205 12:18:10.011868 137274321021824 utils.py:1231] [104800] progress = 0.930703444846052
+I1205 12:18:10.011935 137274321021824 utils.py:1231] [104800] epoch = 83.76363112693349
+I1205 12:18:10.011995 137274321021824 utils.py:1231] [104800] img/sec/core = 164.0902373409629
+I1205 12:18:10.012061 137274321021824 utils.py:1231] [104800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.43217520500417
+I1205 12:18:10.012116 137274321021824 utils.py:1231] [104800] core_hours = 182.43217520500417
+I1205 12:18:10.012195 137274321021824 train.py:125] NOTE: Steps:104800/112603 [93.1%]
+Walltime:7d14h27m (0s eval)
+ETA:13h35m
+Total train time:8d4h1m
+I1205 12:23:21.795251 137274321021824 utils.py:1231] [104850] l2_params = 238.4652047094333
+I1205 12:23:21.795549 137274321021824 utils.py:1231] [104850] train/loss = 1.7643668204545975
+I1205 12:23:21.795722 137274321021824 utils.py:1231] [104850] l2_grads = 2.510406494140625
+I1205 12:23:21.795798 137274321021824 utils.py:1231] [104850] lr = 1.4025884246804445e-05
+I1205 12:23:21.795888 137274321021824 utils.py:1231] [104850] uptime = 657191.158243877
+I1205 12:23:21.795945 137274321021824 utils.py:1231] [104850] examples_seen = 107366400.0
+I1205 12:23:21.795995 137274321021824 utils.py:1231] [104850] progress = 0.9311474827491275
+I1205 12:23:21.796044 137274321021824 utils.py:1231] [104850] epoch = 83.80359469140245
+I1205 12:23:21.796097 137274321021824 utils.py:1231] [104850] img/sec/core = 164.21617992043446
+I1205 12:23:21.796156 137274321021824 utils.py:1231] [104850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.51878190983138
+I1205 12:23:21.796209 137274321021824 utils.py:1231] [104850] core_hours = 182.51878190983138
+I1205 12:23:21.796274 137274321021824 train.py:125] NOTE: Steps:104850/112603 [93.1%]
+Walltime:7d14h33m (0s eval)
+ETA:13h29m
+Total train time:8d4h1m
+I1205 12:28:33.585438 137274321021824 utils.py:1231] [104900] l2_params = 238.45666408943293
+I1205 12:28:33.585640 137274321021824 utils.py:1231] [104900] train/loss = 2.9949841797351837
+I1205 12:28:33.585729 137274321021824 utils.py:1231] [104900] l2_grads = 2.6732800006866455
+I1205 12:28:33.585792 137274321021824 utils.py:1231] [104900] lr = 1.3846418515145505e-05
+I1205 12:28:33.585852 137274321021824 utils.py:1231] [104900] uptime = 657502.948213571
+I1205 12:28:33.585911 137274321021824 utils.py:1231] [104900] examples_seen = 107417600.0
+I1205 12:28:33.585963 137274321021824 utils.py:1231] [104900] progress = 0.9315915206522029
+I1205 12:28:33.586012 137274321021824 utils.py:1231] [104900] epoch = 83.8435582558714
+I1205 12:28:33.586065 137274321021824 utils.py:1231] [104900] img/sec/core = 164.21310810688217
+I1205 12:28:33.586128 137274321021824 utils.py:1231] [104900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.60539023474638
+I1205 12:28:33.586180 137274321021824 utils.py:1231] [104900] core_hours = 182.60539023474638
+I1205 12:28:33.586241 137274321021824 train.py:125] NOTE: Steps:104900/112603 [93.2%]
+Walltime:7d14h38m (0s eval)
+ETA:13h24m
+Total train time:8d4h1m
+I1205 12:33:45.379460 137274321021824 utils.py:1231] [104950] l2_params = 238.44878669975716
+I1205 12:33:45.379691 137274321021824 utils.py:1231] [104950] train/loss = 1.4959626346826553
+I1205 12:33:45.379801 137274321021824 utils.py:1231] [104950] l2_grads = 2.8426854610443115
+I1205 12:33:45.379890 137274321021824 utils.py:1231] [104950] lr = 1.3668092227629273e-05
+I1205 12:33:45.379954 137274321021824 utils.py:1231] [104950] uptime = 657814.74231495
+I1205 12:33:45.380018 137274321021824 utils.py:1231] [104950] examples_seen = 107468800.0
+I1205 12:33:45.380076 137274321021824 utils.py:1231] [104950] progress = 0.9320355585552783
+I1205 12:33:45.380133 137274321021824 utils.py:1231] [104950] epoch = 83.88352182034036
+I1205 12:33:45.380192 137274321021824 utils.py:1231] [104950] img/sec/core = 164.21093206559598
+I1205 12:33:45.380265 137274321021824 utils.py:1231] [104950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.69199970735164
+I1205 12:33:45.380324 137274321021824 utils.py:1231] [104950] core_hours = 182.69199970735164
+I1205 12:33:45.380388 137274321021824 train.py:125] NOTE: Steps:104950/112603 [93.2%]
+Walltime:7d14h43m (0s eval)
+ETA:13h19m
+Total train time:8d4h1m
+I1205 12:38:57.177954 137274321021824 utils.py:1231] [105000] l2_params = 238.43974486677894
+I1205 12:38:57.178178 137274321021824 utils.py:1231] [105000] train/loss = 2.576964169740677
+I1205 12:38:57.178344 137274321021824 utils.py:1231] [105000] l2_grads = 2.6277027130126953
+I1205 12:38:57.178432 137274321021824 utils.py:1231] [105000] lr = 1.3490905802215968e-05
+I1205 12:38:57.178502 137274321021824 utils.py:1231] [105000] uptime = 658126.540863429
+I1205 12:38:57.178563 137274321021824 utils.py:1231] [105000] examples_seen = 107520000.0
+I1205 12:38:57.178637 137274321021824 utils.py:1231] [105000] progress = 0.9324795964583537
+I1205 12:38:57.178709 137274321021824 utils.py:1231] [105000] epoch = 83.92348538480933
+I1205 12:38:57.178778 137274321021824 utils.py:1231] [105000] img/sec/core = 164.20858996863248
+I1205 12:38:57.178856 137274321021824 utils.py:1231] [105000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.77861041526248
+I1205 12:38:57.178950 137274321021824 utils.py:1231] [105000] core_hours = 182.77861041526248
+I1205 12:38:57.179027 137274321021824 train.py:125] NOTE: Steps:105000/112603 [93.2%]
+Walltime:7d14h48m (0s eval)
+ETA:13h14m
+Total train time:8d4h1m
+I1205 12:38:57.525446 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:105000/112603 [93.2%]
+Walltime:7d14h48m (0s eval)
+ETA:13h14m
+Total train time:8d4h1m
+I1205 12:40:35.637283 137274321021824 utils.py:1231] [105000] val/acc@1 = 0.7628148915816326
+I1205 12:40:35.637555 137274321021824 utils.py:1231] [105000] val/loss = 0.9271257472585659
+I1205 12:40:35.637739 137274321021824 utils.py:1231] [105000] z/secs/eval/val = 98.11204287398141
+I1205 12:40:35.637810 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.11204287398141
+I1205 12:45:46.608158 137274321021824 utils.py:1231] [105050] l2_params = 238.43090409258753
+I1205 12:45:46.608377 137274321021824 utils.py:1231] [105050] train/loss = 3.769624173641205
+I1205 12:45:46.608489 137274321021824 utils.py:1231] [105050] l2_grads = 2.9198873043060303
+I1205 12:45:46.608562 137274321021824 utils.py:1231] [105050] lr = 1.3314859654194123e-05
+I1205 12:45:46.608621 137274321021824 utils.py:1231] [105050] uptime = 658535.970982615
+I1205 12:45:46.608689 137274321021824 utils.py:1231] [105050] examples_seen = 107571200.0
+I1205 12:45:46.608755 137274321021824 utils.py:1231] [105050] progress = 0.9329236343614291
+I1205 12:45:46.608822 137274321021824 utils.py:1231] [105050] epoch = 83.96344894927827
+I1205 12:45:46.608901 137274321021824 utils.py:1231] [105050] img/sec/core = 125.05186502103217
+I1205 12:45:46.608982 137274321021824 utils.py:1231] [105050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.89234100392528
+I1205 12:45:46.609034 137274321021824 utils.py:1231] [105050] core_hours = 182.89234100392528
+I1205 12:45:46.609096 137274321021824 train.py:125] NOTE: Steps:105050/112603 [93.3%]
+Walltime:7d14h55m (0s eval)
+ETA:13h9m
+Total train time:8d4h2m
+I1205 12:50:58.398632 137274321021824 utils.py:1231] [105100] l2_params = 238.42277859817136
+I1205 12:50:58.398856 137274321021824 utils.py:1231] [105100] train/loss = 3.0150102972984314
+I1205 12:50:58.398973 137274321021824 utils.py:1231] [105100] l2_grads = 2.5252692699432373
+I1205 12:50:58.399095 137274321021824 utils.py:1231] [105100] lr = 1.3139954196179763e-05
+I1205 12:50:58.399196 137274321021824 utils.py:1231] [105100] uptime = 658847.761554042
+I1205 12:50:58.399271 137274321021824 utils.py:1231] [105100] examples_seen = 107622400.0
+I1205 12:50:58.399341 137274321021824 utils.py:1231] [105100] progress = 0.9333676722645045
+I1205 12:50:58.399439 137274321021824 utils.py:1231] [105100] epoch = 84.00341251374724
+I1205 12:50:58.399540 137274321021824 utils.py:1231] [105100] img/sec/core = 164.21279118760015
+I1205 12:50:58.399626 137274321021824 utils.py:1231] [105100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 182.97894949598833
+I1205 12:50:58.399705 137274321021824 utils.py:1231] [105100] core_hours = 182.97894949598833
+I1205 12:50:58.399811 137274321021824 train.py:125] NOTE: Steps:105100/112603 [93.3%]
+Walltime:7d15h0m (0s eval)
+ETA:13h3m
+Total train time:8d4h2m
+I1205 12:56:10.173393 137274321021824 utils.py:1231] [105150] l2_params = 238.41519622965683
+I1205 12:56:10.173673 137274321021824 utils.py:1231] [105150] train/loss = 2.2989587485790253
+I1205 12:56:10.173899 137274321021824 utils.py:1231] [105150] l2_grads = 2.6094017028808594
+I1205 12:56:10.173999 137274321021824 utils.py:1231] [105150] lr = 1.2966189838115386e-05
+I1205 12:56:10.174098 137274321021824 utils.py:1231] [105150] uptime = 659159.536450917
+I1205 12:56:10.174196 137274321021824 utils.py:1231] [105150] examples_seen = 107673600.0
+I1205 12:56:10.174283 137274321021824 utils.py:1231] [105150] progress = 0.9338117101675799
+I1205 12:56:10.174365 137274321021824 utils.py:1231] [105150] epoch = 84.04337607821618
+I1205 12:56:10.174458 137274321021824 utils.py:1231] [105150] img/sec/core = 164.2210470220241
+I1205 12:56:10.174566 137274321021824 utils.py:1231] [105150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.06555363400918
+I1205 12:56:10.174632 137274321021824 utils.py:1231] [105150] core_hours = 183.06555363400918
+I1205 12:56:10.174724 137274321021824 train.py:125] NOTE: Steps:105150/112603 [93.4%]
+Walltime:7d15h5m (0s eval)
+ETA:12h58m
+Total train time:8d4h2m
+I1205 13:01:21.955383 137274321021824 utils.py:1231] [105200] l2_params = 238.4068299964123
+I1205 13:01:21.955639 137274321021824 utils.py:1231] [105200] train/loss = 1.420710727572441
+I1205 13:01:21.955848 137274321021824 utils.py:1231] [105200] l2_grads = 2.891014337539673
+I1205 13:01:21.955950 137274321021824 utils.py:1231] [105200] lr = 1.2793566987268807e-05
+I1205 13:01:21.956013 137274321021824 utils.py:1231] [105200] uptime = 659471.318374495
+I1205 13:01:21.956067 137274321021824 utils.py:1231] [105200] examples_seen = 107724800.0
+I1205 13:01:21.956120 137274321021824 utils.py:1231] [105200] progress = 0.9342557480706553
+I1205 13:01:21.956168 137274321021824 utils.py:1231] [105200] epoch = 84.08333964268515
+I1205 13:01:21.956229 137274321021824 utils.py:1231] [105200] img/sec/core = 164.217345933432
+I1205 13:01:21.956285 137274321021824 utils.py:1231] [105200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.15215972389194
+I1205 13:01:21.956334 137274321021824 utils.py:1231] [105200] core_hours = 183.15215972389194
+I1205 13:01:21.956394 137274321021824 train.py:125] NOTE: Steps:105200/112603 [93.4%]
+Walltime:7d15h11m (0s eval)
+ETA:12h53m
+Total train time:8d4h2m
+I1205 13:06:33.754191 137274321021824 utils.py:1231] [105250] l2_params = 238.39914138880047
+I1205 13:06:33.754406 137274321021824 utils.py:1231] [105250] train/loss = 2.4006117582321167
+I1205 13:06:33.754503 137274321021824 utils.py:1231] [105250] l2_grads = 2.579993963241577
+I1205 13:06:33.754568 137274321021824 utils.py:1231] [105250] lr = 1.2622086048232793e-05
+I1205 13:06:33.754621 137274321021824 utils.py:1231] [105250] uptime = 659783.11698263
+I1205 13:06:33.754678 137274321021824 utils.py:1231] [105250] examples_seen = 107776000.0
+I1205 13:06:33.754728 137274321021824 utils.py:1231] [105250] progress = 0.9346997859737307
+I1205 13:06:33.754775 137274321021824 utils.py:1231] [105250] epoch = 84.12330320715411
+I1205 13:06:33.754826 137274321021824 utils.py:1231] [105250] img/sec/core = 164.20855855083667
+I1205 13:06:33.754885 137274321021824 utils.py:1231] [105250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.23877044837388
+I1205 13:06:33.754936 137274321021824 utils.py:1231] [105250] core_hours = 183.23877044837388
+I1205 13:06:33.754995 137274321021824 train.py:125] NOTE: Steps:105250/112603 [93.5%]
+Walltime:7d15h16m (0s eval)
+ETA:12h48m
+Total train time:8d4h2m
+I1205 13:11:45.471178 137274321021824 utils.py:1231] [105300] l2_params = 238.39082875492778
+I1205 13:11:45.471403 137274321021824 utils.py:1231] [105300] train/loss = 1.4506221860647202
+I1205 13:11:45.471516 137274321021824 utils.py:1231] [105300] l2_grads = 2.843141794204712
+I1205 13:11:45.471590 137274321021824 utils.py:1231] [105300] lr = 1.2451747422923274e-05
+I1205 13:11:45.471652 137274321021824 utils.py:1231] [105300] uptime = 660094.834013146
+I1205 13:11:45.471717 137274321021824 utils.py:1231] [105300] examples_seen = 107827200.0
+I1205 13:11:45.471776 137274321021824 utils.py:1231] [105300] progress = 0.9351438238768062
+I1205 13:11:45.471833 137274321021824 utils.py:1231] [105300] epoch = 84.16326677162306
+I1205 13:11:45.471903 137274321021824 utils.py:1231] [105300] img/sec/core = 164.2515326007195
+I1205 13:11:45.471977 137274321021824 utils.py:1231] [105300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.32535851240613
+I1205 13:11:45.472037 137274321021824 utils.py:1231] [105300] core_hours = 183.32535851240613
+I1205 13:11:45.472104 137274321021824 train.py:125] NOTE: Steps:105300/112603 [93.5%]
+Walltime:7d15h21m (0s eval)
+ETA:12h42m
+Total train time:8d4h2m
+I1205 13:16:57.218395 137274321021824 utils.py:1231] [105350] l2_params = 238.38213457421148
+I1205 13:16:57.218647 137274321021824 utils.py:1231] [105350] train/loss = 3.801688939332962
+I1205 13:16:57.218783 137274321021824 utils.py:1231] [105350] l2_grads = 3.0055580139160156
+I1205 13:16:57.218885 137274321021824 utils.py:1231] [105350] lr = 1.2282551510579047e-05
+I1205 13:16:57.218949 137274321021824 utils.py:1231] [105350] uptime = 660406.581309934
+I1205 13:16:57.219009 137274321021824 utils.py:1231] [105350] examples_seen = 107878400.0
+I1205 13:16:57.219066 137274321021824 utils.py:1231] [105350] progress = 0.9355878617798815
+I1205 13:16:57.219122 137274321021824 utils.py:1231] [105350] epoch = 84.20323033609202
+I1205 13:16:57.219185 137274321021824 utils.py:1231] [105350] img/sec/core = 164.2355860901428
+I1205 13:16:57.219254 137274321021824 utils.py:1231] [105350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.4119549837361
+I1205 13:16:57.219316 137274321021824 utils.py:1231] [105350] core_hours = 183.4119549837361
+I1205 13:16:57.219397 137274321021824 train.py:125] NOTE: Steps:105350/112603 [93.6%]
+Walltime:7d15h26m (0s eval)
+ETA:12h37m
+Total train time:8d4h2m
+I1205 13:22:09.005568 137274321021824 utils.py:1231] [105400] l2_params = 238.37483911628874
+I1205 13:22:09.005786 137274321021824 utils.py:1231] [105400] train/loss = 1.221925437450409
+I1205 13:22:09.005923 137274321021824 utils.py:1231] [105400] l2_grads = 2.764611005783081
+I1205 13:22:09.006023 137274321021824 utils.py:1231] [105400] lr = 1.21144987077606e-05
+I1205 13:22:09.006095 137274321021824 utils.py:1231] [105400] uptime = 660718.368457307
+I1205 13:22:09.006155 137274321021824 utils.py:1231] [105400] examples_seen = 107929600.0
+I1205 13:22:09.006210 137274321021824 utils.py:1231] [105400] progress = 0.936031899682957
+I1205 13:22:09.006264 137274321021824 utils.py:1231] [105400] epoch = 84.24319390056097
+I1205 13:22:09.006321 137274321021824 utils.py:1231] [105400] img/sec/core = 164.2145945764543
+I1205 13:22:09.006383 137274321021824 utils.py:1231] [105400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.49856252467305
+I1205 13:22:09.006442 137274321021824 utils.py:1231] [105400] core_hours = 183.49856252467305
+I1205 13:22:09.006506 137274321021824 train.py:125] NOTE: Steps:105400/112603 [93.6%]
+Walltime:7d15h31m (0s eval)
+ETA:12h32m
+Total train time:8d4h2m
+I1205 13:27:20.788432 137274321021824 utils.py:1231] [105450] l2_params = 238.36696954191908
+I1205 13:27:20.788673 137274321021824 utils.py:1231] [105450] train/loss = 3.8198936581611633
+I1205 13:27:20.788795 137274321021824 utils.py:1231] [105450] l2_grads = 3.048922300338745
+I1205 13:27:20.788890 137274321021824 utils.py:1231] [105450] lr = 1.1947589408349055e-05
+I1205 13:27:20.788945 137274321021824 utils.py:1231] [105450] uptime = 661030.15130702
+I1205 13:27:20.789000 137274321021824 utils.py:1231] [105450] examples_seen = 107980800.0
+I1205 13:27:20.789050 137274321021824 utils.py:1231] [105450] progress = 0.9364759375860323
+I1205 13:27:20.789099 137274321021824 utils.py:1231] [105450] epoch = 84.28315746502993
+I1205 13:27:20.789149 137274321021824 utils.py:1231] [105450] img/sec/core = 164.21685813421402
+I1205 13:27:20.789206 137274321021824 utils.py:1231] [105450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.58516887181557
+I1205 13:27:20.789254 137274321021824 utils.py:1231] [105450] core_hours = 183.58516887181557
+I1205 13:27:20.789313 137274321021824 train.py:125] NOTE: Steps:105450/112603 [93.6%]
+Walltime:7d15h37m (0s eval)
+ETA:12h27m
+Total train time:8d4h2m
+I1205 13:32:32.593059 137274321021824 utils.py:1231] [105500] l2_params = 238.35844470917522
+I1205 13:32:32.593262 137274321021824 utils.py:1231] [105500] train/loss = 2.724759519100189
+I1205 13:32:32.593352 137274321021824 utils.py:1231] [105500] l2_grads = 2.5746357440948486
+I1205 13:32:32.593417 137274321021824 utils.py:1231] [105500] lr = 1.1781824003545706e-05
+I1205 13:32:32.593473 137274321021824 utils.py:1231] [105500] uptime = 661341.955835378
+I1205 13:32:32.593529 137274321021824 utils.py:1231] [105500] examples_seen = 108032000.0
+I1205 13:32:32.593579 137274321021824 utils.py:1231] [105500] progress = 0.9369199754891078
+I1205 13:32:32.593629 137274321021824 utils.py:1231] [105500] epoch = 84.32312102949889
+I1205 13:32:32.593683 137274321021824 utils.py:1231] [105500] img/sec/core = 164.20544072794019
+I1205 13:32:32.593745 137274321021824 utils.py:1231] [105500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.67178124080388
+I1205 13:32:32.593797 137274321021824 utils.py:1231] [105500] core_hours = 183.67178124080388
+I1205 13:32:32.593858 137274321021824 train.py:125] NOTE: Steps:105500/112603 [93.7%]
+Walltime:7d15h42m (0s eval)
+ETA:12h21m
+Total train time:8d4h2m
+I1205 13:37:44.381559 137274321021824 utils.py:1231] [105550] l2_params = 238.35078885678297
+I1205 13:37:44.381766 137274321021824 utils.py:1231] [105550] train/loss = 1.3172641694545746
+I1205 13:37:44.381865 137274321021824 utils.py:1231] [105550] l2_grads = 2.80665922164917
+I1205 13:37:44.381954 137274321021824 utils.py:1231] [105550] lr = 1.1617202881870553e-05
+I1205 13:37:44.382054 137274321021824 utils.py:1231] [105550] uptime = 661653.744412964
+I1205 13:37:44.382118 137274321021824 utils.py:1231] [105550] examples_seen = 108083200.0
+I1205 13:37:44.382173 137274321021824 utils.py:1231] [105550] progress = 0.9373640133921831
+I1205 13:37:44.382235 137274321021824 utils.py:1231] [105550] epoch = 84.36308459396784
+I1205 13:37:44.382306 137274321021824 utils.py:1231] [105550] img/sec/core = 164.21384130368392
+I1205 13:37:44.382373 137274321021824 utils.py:1231] [105550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.75838917902223
+I1205 13:37:44.382438 137274321021824 utils.py:1231] [105550] core_hours = 183.75838917902223
+I1205 13:37:44.382511 137274321021824 train.py:125] NOTE: Steps:105550/112603 [93.7%]
+Walltime:7d15h47m (0s eval)
+ETA:12h16m
+Total train time:8d4h2m
+I1205 13:42:56.143070 137274321021824 utils.py:1231] [105600] l2_params = 238.34338380103503
+I1205 13:42:56.143320 137274321021824 utils.py:1231] [105600] train/loss = 1.6161503046751022
+I1205 13:42:56.143461 137274321021824 utils.py:1231] [105600] l2_grads = 2.814472198486328
+I1205 13:42:56.143556 137274321021824 utils.py:1231] [105600] lr = 1.1453726429161687e-05
+I1205 13:42:56.143645 137274321021824 utils.py:1231] [105600] uptime = 661965.5060007389
+I1205 13:42:56.143731 137274321021824 utils.py:1231] [105600] examples_seen = 108134400.0
+I1205 13:42:56.143803 137274321021824 utils.py:1231] [105600] progress = 0.9378080512952586
+I1205 13:42:56.143877 137274321021824 utils.py:1231] [105600] epoch = 84.4030481584368
+I1205 13:42:56.143947 137274321021824 utils.py:1231] [105600] img/sec/core = 164.228057617404
+I1205 13:42:56.144014 137274321021824 utils.py:1231] [105600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.84498962007083
+I1205 13:42:56.144076 137274321021824 utils.py:1231] [105600] core_hours = 183.84498962007083
+I1205 13:42:56.144144 137274321021824 train.py:125] NOTE: Steps:105600/112603 [93.8%]
+Walltime:7d15h52m (0s eval)
+ETA:12h11m
+Total train time:8d4h2m
+I1205 13:48:07.936909 137274321021824 utils.py:1231] [105650] l2_params = 238.33623832695906
+I1205 13:48:07.937133 137274321021824 utils.py:1231] [105650] train/loss = 1.559128686785698
+I1205 13:48:07.937230 137274321021824 utils.py:1231] [105650] l2_grads = 2.9498934745788574
+I1205 13:48:07.937304 137274321021824 utils.py:1231] [105650] lr = 1.129139502857432e-05
+I1205 13:48:07.937371 137274321021824 utils.py:1231] [105650] uptime = 662277.299732332
+I1205 13:48:07.937451 137274321021824 utils.py:1231] [105650] examples_seen = 108185600.0
+I1205 13:48:07.937512 137274321021824 utils.py:1231] [105650] progress = 0.9382520891983339
+I1205 13:48:07.937580 137274321021824 utils.py:1231] [105650] epoch = 84.44301172290575
+I1205 13:48:07.937678 137274321021824 utils.py:1231] [105650] img/sec/core = 164.21112681900826
+I1205 13:48:07.937748 137274321021824 utils.py:1231] [105650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 183.9315989899578
+I1205 13:48:07.937809 137274321021824 utils.py:1231] [105650] core_hours = 183.9315989899578
+I1205 13:48:07.937898 137274321021824 train.py:125] NOTE: Steps:105650/112603 [93.8%]
+Walltime:7d15h57m (0s eval)
+ETA:12h6m
+Total train time:8d4h2m
+I1205 13:53:19.739055 137274321021824 utils.py:1231] [105700] l2_params = 238.32841632022158
+I1205 13:53:19.739260 137274321021824 utils.py:1231] [105700] train/loss = 3.6422497034072876
+I1205 13:53:19.739360 137274321021824 utils.py:1231] [105700] l2_grads = 3.035428285598755
+I1205 13:53:19.739440 137274321021824 utils.py:1231] [105700] lr = 1.1130209060579847e-05
+I1205 13:53:19.739500 137274321021824 utils.py:1231] [105700] uptime = 662589.101861788
+I1205 13:53:19.739562 137274321021824 utils.py:1231] [105700] examples_seen = 108236800.0
+I1205 13:53:19.739619 137274321021824 utils.py:1231] [105700] progress = 0.9386961271014094
+I1205 13:53:19.739688 137274321021824 utils.py:1231] [105700] epoch = 84.48297528737471
+I1205 13:53:19.739757 137274321021824 utils.py:1231] [105700] img/sec/core = 164.2067040700949
+I1205 13:53:19.739836 137274321021824 utils.py:1231] [105700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.01821069258443
+I1205 13:53:19.739910 137274321021824 utils.py:1231] [105700] core_hours = 184.01821069258443
+I1205 13:53:19.740012 137274321021824 train.py:125] NOTE: Steps:105700/112603 [93.9%]
+Walltime:7d16h3m (0s eval)
+ETA:12h1m
+Total train time:8d4h2m
+I1205 13:58:31.532189 137274321021824 utils.py:1231] [105750] l2_params = 238.3218514778381
+I1205 13:58:31.532390 137274321021824 utils.py:1231] [105750] train/loss = 2.4212532937526703
+I1205 13:58:31.532490 137274321021824 utils.py:1231] [105750] l2_grads = 2.533691883087158
+I1205 13:58:31.532575 137274321021824 utils.py:1231] [105750] lr = 1.0970168902965231e-05
+I1205 13:58:31.532632 137274321021824 utils.py:1231] [105750] uptime = 662900.894993661
+I1205 13:58:31.532688 137274321021824 utils.py:1231] [105750] examples_seen = 108288000.0
+I1205 13:58:31.532738 137274321021824 utils.py:1231] [105750] progress = 0.9391401650044848
+I1205 13:58:31.532785 137274321021824 utils.py:1231] [105750] epoch = 84.52293885184368
+I1205 13:58:31.532834 137274321021824 utils.py:1231] [105750] img/sec/core = 164.21144267171655
+I1205 13:58:31.532901 137274321021824 utils.py:1231] [105750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.10481989588249
+I1205 13:58:31.532951 137274321021824 utils.py:1231] [105750] core_hours = 184.10481989588249
+I1205 13:58:31.533010 137274321021824 train.py:125] NOTE: Steps:105750/112603 [93.9%]
+Walltime:7d16h8m (0s eval)
+ETA:11h55m
+Total train time:8d4h2m
+I1205 14:03:43.318854 137274321021824 utils.py:1231] [105800] l2_params = 238.31530951585364
+I1205 14:03:43.319112 137274321021824 utils.py:1231] [105800] train/loss = 1.7509811371564865
+I1205 14:03:43.319251 137274321021824 utils.py:1231] [105800] l2_grads = 2.8776397705078125
+I1205 14:03:43.319345 137274321021824 utils.py:1231] [105800] lr = 1.0811274930831645e-05
+I1205 14:03:43.319414 137274321021824 utils.py:1231] [105800] uptime = 663212.681774988
+I1205 14:03:43.319489 137274321021824 utils.py:1231] [105800] examples_seen = 108339200.0
+I1205 14:03:43.319556 137274321021824 utils.py:1231] [105800] progress = 0.9395842029075602
+I1205 14:03:43.319615 137274321021824 utils.py:1231] [105800] epoch = 84.56290241631262
+I1205 14:03:43.319684 137274321021824 utils.py:1231] [105800] img/sec/core = 164.21478736875758
+I1205 14:03:43.319745 137274321021824 utils.py:1231] [105800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.19142733514
+I1205 14:03:43.319807 137274321021824 utils.py:1231] [105800] core_hours = 184.19142733514
+I1205 14:03:43.319879 137274321021824 train.py:125] NOTE: Steps:105800/112603 [94.0%]
+Walltime:7d16h13m (0s eval)
+ETA:11h50m
+Total train time:8d4h2m
+I1205 14:08:55.104654 137274321021824 utils.py:1231] [105850] l2_params = 238.30809984401867
+I1205 14:08:55.104859 137274321021824 utils.py:1231] [105850] train/loss = 1.5980489999055862
+I1205 14:08:55.104969 137274321021824 utils.py:1231] [105850] l2_grads = 2.8413586616516113
+I1205 14:08:55.105039 137274321021824 utils.py:1231] [105850] lr = 1.0653527516593905e-05
+I1205 14:08:55.105094 137274321021824 utils.py:1231] [105850] uptime = 663524.467455765
+I1205 14:08:55.105148 137274321021824 utils.py:1231] [105850] examples_seen = 108390400.0
+I1205 14:08:55.105198 137274321021824 utils.py:1231] [105850] progress = 0.9400282408106356
+I1205 14:08:55.105247 137274321021824 utils.py:1231] [105850] epoch = 84.60286598078159
+I1205 14:08:55.105299 137274321021824 utils.py:1231] [105850] img/sec/core = 164.2153670187721
+I1205 14:08:55.105357 137274321021824 utils.py:1231] [105850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.27803446868919
+I1205 14:08:55.105409 137274321021824 utils.py:1231] [105850] core_hours = 184.27803446868919
+I1205 14:08:55.105470 137274321021824 train.py:125] NOTE: Steps:105850/112603 [94.0%]
+Walltime:7d16h18m (0s eval)
+ETA:11h45m
+Total train time:8d4h2m
+I1205 14:14:06.888802 137274321021824 utils.py:1231] [105900] l2_params = 238.3011384684769
+I1205 14:14:06.889072 137274321021824 utils.py:1231] [105900] train/loss = 1.8418495804071426
+I1205 14:14:06.889208 137274321021824 utils.py:1231] [105900] l2_grads = 2.8935484886169434
+I1205 14:14:06.889327 137274321021824 utils.py:1231] [105900] lr = 1.049692702997959e-05
+I1205 14:14:06.889407 137274321021824 utils.py:1231] [105900] uptime = 663836.251767208
+I1205 14:14:06.889487 137274321021824 utils.py:1231] [105900] examples_seen = 108441600.0
+I1205 14:14:06.889569 137274321021824 utils.py:1231] [105900] progress = 0.940472278713711
+I1205 14:14:06.889632 137274321021824 utils.py:1231] [105900] epoch = 84.64282954525054
+I1205 14:14:06.889715 137274321021824 utils.py:1231] [105900] img/sec/core = 164.21608824074121
+I1205 14:14:06.889844 137274321021824 utils.py:1231] [105900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.36464122186777
+I1205 14:14:06.889945 137274321021824 utils.py:1231] [105900] core_hours = 184.36464122186777
+I1205 14:14:06.890016 137274321021824 train.py:125] NOTE: Steps:105900/112603 [94.0%]
+Walltime:7d16h23m (0s eval)
+ETA:11h40m
+Total train time:8d4h2m
+I1205 14:19:18.820603 137274321021824 utils.py:1231] [105950] l2_params = 238.29402254636165
+I1205 14:19:18.820806 137274321021824 utils.py:1231] [105950] train/loss = 3.283042073249817
+I1205 14:19:18.820909 137274321021824 utils.py:1231] [105950] l2_grads = 2.8700833320617676
+I1205 14:19:18.820981 137274321021824 utils.py:1231] [105950] lr = 1.0341473838027993e-05
+I1205 14:19:18.821041 137274321021824 utils.py:1231] [105950] uptime = 664148.183402602
+I1205 14:19:18.821099 137274321021824 utils.py:1231] [105950] examples_seen = 108492800.0
+I1205 14:19:18.821155 137274321021824 utils.py:1231] [105950] progress = 0.9409163166167864
+I1205 14:19:18.821211 137274321021824 utils.py:1231] [105950] epoch = 84.6827931097195
+I1205 14:19:18.821267 137274321021824 utils.py:1231] [105950] img/sec/core = 164.1385296984399
+I1205 14:19:18.821328 137274321021824 utils.py:1231] [105950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.45128889836613
+I1205 14:19:18.821384 137274321021824 utils.py:1231] [105950] core_hours = 184.45128889836613
+I1205 14:19:18.821450 137274321021824 train.py:125] NOTE: Steps:105950/112603 [94.1%]
+Walltime:7d16h29m (0s eval)
+ETA:11h34m
+Total train time:8d4h2m
+I1205 14:24:30.620060 137274321021824 utils.py:1231] [106000] l2_params = 238.28721235088614
+I1205 14:24:30.620276 137274321021824 utils.py:1231] [106000] train/loss = 1.9093187898397446
+I1205 14:24:30.620390 137274321021824 utils.py:1231] [106000] l2_grads = 2.7527475357055664
+I1205 14:24:30.620470 137274321021824 utils.py:1231] [106000] lr = 1.0187168305089621e-05
+I1205 14:24:30.620532 137274321021824 utils.py:1231] [106000] uptime = 664459.982892883
+I1205 14:24:30.620598 137274321021824 utils.py:1231] [106000] examples_seen = 108544000.0
+I1205 14:24:30.620658 137274321021824 utils.py:1231] [106000] progress = 0.9413603545198618
+I1205 14:24:30.620715 137274321021824 utils.py:1231] [106000] epoch = 84.72275667418846
+I1205 14:24:30.620775 137274321021824 utils.py:1231] [106000] img/sec/core = 164.20809397043067
+I1205 14:24:30.620842 137274321021824 utils.py:1231] [106000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.5378998678886
+I1205 14:24:30.620906 137274321021824 utils.py:1231] [106000] core_hours = 184.5378998678886
+I1205 14:24:30.620975 137274321021824 train.py:125] NOTE: Steps:106000/112603 [94.1%]
+Walltime:7d16h34m (0s eval)
+ETA:11h29m
+Total train time:8d4h2m
+I1205 14:29:42.780884 137274321021824 utils.py:1231] [106050] l2_params = 238.28090125086544
+I1205 14:29:42.781144 137274321021824 utils.py:1231] [106050] train/loss = 1.4792408049106598
+I1205 14:29:42.781283 137274321021824 utils.py:1231] [106050] l2_grads = 2.8543741703033447
+I1205 14:29:42.781382 137274321021824 utils.py:1231] [106050] lr = 1.0034010792824815e-05
+I1205 14:29:42.781456 137274321021824 utils.py:1231] [106050] uptime = 664772.143817456
+I1205 14:29:42.781529 137274321021824 utils.py:1231] [106050] examples_seen = 108595200.0
+I1205 14:29:42.781596 137274321021824 utils.py:1231] [106050] progress = 0.9418043924229372
+I1205 14:29:42.781663 137274321021824 utils.py:1231] [106050] epoch = 84.76272023865741
+I1205 14:29:42.781734 137274321021824 utils.py:1231] [106050] img/sec/core = 164.017966278264
+I1205 14:29:42.781807 137274321021824 utils.py:1231] [106050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.62461123582554
+I1205 14:29:42.781869 137274321021824 utils.py:1231] [106050] core_hours = 184.62461123582554
+I1205 14:29:42.781954 137274321021824 train.py:125] NOTE: Steps:106050/112603 [94.2%]
+Walltime:7d16h39m (0s eval)
+ETA:11h24m
+Total train time:8d4h2m
+I1205 14:34:54.578300 137274321021824 utils.py:1231] [106100] l2_params = 238.27400269290143
+I1205 14:34:54.578541 137274321021824 utils.py:1231] [106100] train/loss = 1.554186299443245
+I1205 14:34:54.578636 137274321021824 utils.py:1231] [106100] l2_grads = 2.884991407394409
+I1205 14:34:54.578700 137274321021824 utils.py:1231] [106100] lr = 9.882001660203335e-06
+I1205 14:34:54.578753 137274321021824 utils.py:1231] [106100] uptime = 665083.941115087
+I1205 14:34:54.578807 137274321021824 utils.py:1231] [106100] examples_seen = 108646400.0
+I1205 14:34:54.578856 137274321021824 utils.py:1231] [106100] progress = 0.9422484303260126
+I1205 14:34:54.578914 137274321021824 utils.py:1231] [106100] epoch = 84.80268380312637
+I1205 14:34:54.578968 137274321021824 utils.py:1231] [106100] img/sec/core = 164.20924872987982
+I1205 14:34:54.579024 137274321021824 utils.py:1231] [106100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.7112215962786
+I1205 14:34:54.579073 137274321021824 utils.py:1231] [106100] core_hours = 184.7112215962786
+I1205 14:34:54.579134 137274321021824 train.py:125] NOTE: Steps:106100/112603 [94.2%]
+Walltime:7d16h44m (0s eval)
+ETA:11h19m
+Total train time:8d4h2m
+I1205 14:40:06.377574 137274321021824 utils.py:1231] [106150] l2_params = 238.26786102513026
+I1205 14:40:06.377824 137274321021824 utils.py:1231] [106150] train/loss = 1.387568712234497
+I1205 14:40:06.377972 137274321021824 utils.py:1231] [106150] l2_grads = 2.773035764694214
+I1205 14:40:06.378065 137274321021824 utils.py:1231] [106150] lr = 9.731141263503459e-06
+I1205 14:40:06.378145 137274321021824 utils.py:1231] [106150] uptime = 665395.740502672
+I1205 14:40:06.378221 137274321021824 utils.py:1231] [106150] examples_seen = 108697600.0
+I1205 14:40:06.378287 137274321021824 utils.py:1231] [106150] progress = 0.942692468229088
+I1205 14:40:06.378355 137274321021824 utils.py:1231] [106150] epoch = 84.84264736759532
+I1205 14:40:06.378429 137274321021824 utils.py:1231] [106150] img/sec/core = 164.20814805493683
+I1205 14:40:06.378505 137274321021824 utils.py:1231] [106150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.79783253727447
+I1205 14:40:06.378563 137274321021824 utils.py:1231] [106150] core_hours = 184.79783253727447
+I1205 14:40:06.378630 137274321021824 train.py:125] NOTE: Steps:106150/112603 [94.3%]
+Walltime:7d16h49m (0s eval)
+ETA:11h14m
+Total train time:8d4h2m
+I1205 14:45:18.175248 137274321021824 utils.py:1231] [106200] l2_params = 238.26208080017517
+I1205 14:45:18.175454 137274321021824 utils.py:1231] [106200] train/loss = 2.632441520690918
+I1205 14:45:18.175556 137274321021824 utils.py:1231] [106200] l2_grads = 2.568948745727539
+I1205 14:45:18.175637 137274321021824 utils.py:1231] [106200] lr = 9.581429956310915e-06
+I1205 14:45:18.175720 137274321021824 utils.py:1231] [106200] uptime = 665707.538073351
+I1205 14:45:18.175808 137274321021824 utils.py:1231] [106200] examples_seen = 108748800.0
+I1205 14:45:18.175898 137274321021824 utils.py:1231] [106200] progress = 0.9431365061321635
+I1205 14:45:18.175961 137274321021824 utils.py:1231] [106200] epoch = 84.88261093206428
+I1205 14:45:18.176036 137274321021824 utils.py:1231] [106200] img/sec/core = 164.2091049282522
+I1205 14:45:18.176142 137274321021824 utils.py:1231] [106200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.88444297357415
+I1205 14:45:18.176213 137274321021824 utils.py:1231] [106200] core_hours = 184.88444297357415
+I1205 14:45:18.176292 137274321021824 train.py:125] NOTE: Steps:106200/112603 [94.3%]
+Walltime:7d16h55m (0s eval)
+ETA:11h8m
+Total train time:8d4h2m
+I1205 14:50:29.969474 137274321021824 utils.py:1231] [106250] l2_params = 238.2559452060639
+I1205 14:50:29.969728 137274321021824 utils.py:1231] [106250] train/loss = 2.4196856915950775
+I1205 14:50:29.969856 137274321021824 utils.py:1231] [106250] l2_grads = 2.5079774856567383
+I1205 14:50:29.969958 137274321021824 utils.py:1231] [106250] lr = 9.432868089518233e-06
+I1205 14:50:29.970031 137274321021824 utils.py:1231] [106250] uptime = 666019.332392798
+I1205 14:50:29.970117 137274321021824 utils.py:1231] [106250] examples_seen = 108800000.0
+I1205 14:50:29.970178 137274321021824 utils.py:1231] [106250] progress = 0.9435805440352388
+I1205 14:50:29.970241 137274321021824 utils.py:1231] [106250] epoch = 84.92257449653324
+I1205 14:50:29.970304 137274321021824 utils.py:1231] [106250] img/sec/core = 164.21081721696007
+I1205 14:50:29.970362 137274321021824 utils.py:1231] [106250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 184.9710525067539
+I1205 14:50:29.970417 137274321021824 utils.py:1231] [106250] core_hours = 184.9710525067539
+I1205 14:50:29.970482 137274321021824 train.py:125] NOTE: Steps:106250/112603 [94.4%]
+Walltime:7d17h0m (0s eval)
+ETA:11h3m
+Total train time:8d4h2m
+I1205 14:55:41.757143 137274321021824 utils.py:1231] [106300] l2_params = 238.24959330584107
+I1205 14:55:41.757342 137274321021824 utils.py:1231] [106300] train/loss = 2.038103073835373
+I1205 14:55:41.757436 137274321021824 utils.py:1231] [106300] l2_grads = 2.6385624408721924
+I1205 14:55:41.757500 137274321021824 utils.py:1231] [106300] lr = 9.285456011323937e-06
+I1205 14:55:41.757561 137274321021824 utils.py:1231] [106300] uptime = 666331.119923007
+I1205 14:55:41.757614 137274321021824 utils.py:1231] [106300] examples_seen = 108851200.0
+I1205 14:55:41.757666 137274321021824 utils.py:1231] [106300] progress = 0.9440245819383143
+I1205 14:55:41.757716 137274321021824 utils.py:1231] [106300] epoch = 84.96253806100219
+I1205 14:55:41.757769 137274321021824 utils.py:1231] [106300] img/sec/core = 164.21439294153106
+I1205 14:55:41.757825 137274321021824 utils.py:1231] [106300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.05766015403415
+I1205 14:55:41.757875 137274321021824 utils.py:1231] [106300] core_hours = 185.05766015403415
+I1205 14:55:41.757943 137274321021824 train.py:125] NOTE: Steps:106300/112603 [94.4%]
+Walltime:7d17h5m (0s eval)
+ETA:10h58m
+Total train time:8d4h2m
+I1205 15:00:53.548568 137274321021824 utils.py:1231] [106350] l2_params = 238.24388687233233
+I1205 15:00:53.548825 137274321021824 utils.py:1231] [106350] train/loss = 1.7290980815887451
+I1205 15:00:53.548946 137274321021824 utils.py:1231] [106350] l2_grads = 2.752502202987671
+I1205 15:00:53.549015 137274321021824 utils.py:1231] [106350] lr = 9.139194067231649e-06
+I1205 15:00:53.549071 137274321021824 utils.py:1231] [106350] uptime = 666642.911432835
+I1205 15:00:53.549126 137274321021824 utils.py:1231] [106350] examples_seen = 108902400.0
+I1205 15:00:53.549179 137274321021824 utils.py:1231] [106350] progress = 0.9444686198413896
+I1205 15:00:53.549230 137274321021824 utils.py:1231] [106350] epoch = 85.00250162547115
+I1205 15:00:53.549296 137274321021824 utils.py:1231] [106350] img/sec/core = 164.21229695521185
+I1205 15:00:53.549365 137274321021824 utils.py:1231] [106350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.14426890676415
+I1205 15:00:53.549422 137274321021824 utils.py:1231] [106350] core_hours = 185.14426890676415
+I1205 15:00:53.549486 137274321021824 train.py:125] NOTE: Steps:106350/112603 [94.4%]
+Walltime:7d17h10m (0s eval)
+ETA:10h53m
+Total train time:8d4h2m
+I1205 15:06:05.319126 137274321021824 utils.py:1231] [106400] l2_params = 238.23825656595704
+I1205 15:06:05.319345 137274321021824 utils.py:1231] [106400] train/loss = 1.4064976423978806
+I1205 15:06:05.319441 137274321021824 utils.py:1231] [106400] l2_grads = 2.778355360031128
+I1205 15:06:05.319504 137274321021824 utils.py:1231] [106400] lr = 8.994082600049378e-06
+I1205 15:06:05.319560 137274321021824 utils.py:1231] [106400] uptime = 666954.681921795
+I1205 15:06:05.319613 137274321021824 utils.py:1231] [106400] examples_seen = 108953600.0
+I1205 15:06:05.319663 137274321021824 utils.py:1231] [106400] progress = 0.9449126577444651
+I1205 15:06:05.319720 137274321021824 utils.py:1231] [106400] epoch = 85.0424651899401
+I1205 15:06:05.319798 137274321021824 utils.py:1231] [106400] img/sec/core = 164.22336883390483
+I1205 15:06:05.319878 137274321021824 utils.py:1231] [106400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.23087182036414
+I1205 15:06:05.319957 137274321021824 utils.py:1231] [106400] core_hours = 185.23087182036414
+I1205 15:06:05.320043 137274321021824 train.py:125] NOTE: Steps:106400/112603 [94.5%]
+Walltime:7d17h15m (0s eval)
+ETA:10h47m
+Total train time:8d4h1m
+I1205 15:11:17.101828 137274321021824 utils.py:1231] [106450] l2_params = 238.2325135106565
+I1205 15:11:17.102078 137274321021824 utils.py:1231] [106450] train/loss = 3.25536772608757
+I1205 15:11:17.102231 137274321021824 utils.py:1231] [106450] l2_grads = 2.7592132091522217
+I1205 15:11:17.102347 137274321021824 utils.py:1231] [106450] lr = 8.850121949888508e-06
+I1205 15:11:17.102443 137274321021824 utils.py:1231] [106450] uptime = 667266.46479894
+I1205 15:11:17.102535 137274321021824 utils.py:1231] [106450] examples_seen = 109004800.0
+I1205 15:11:17.102621 137274321021824 utils.py:1231] [106450] progress = 0.9453566956475404
+I1205 15:11:17.102700 137274321021824 utils.py:1231] [106450] epoch = 85.08242875440907
+I1205 15:11:17.102787 137274321021824 utils.py:1231] [106450] img/sec/core = 164.216843685713
+I1205 15:11:17.102891 137274321021824 utils.py:1231] [106450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.31747817512667
+I1205 15:11:17.102962 137274321021824 utils.py:1231] [106450] core_hours = 185.31747817512667
+I1205 15:11:17.103044 137274321021824 train.py:125] NOTE: Steps:106450/112603 [94.5%]
+Walltime:7d17h21m (0s eval)
+ETA:10h42m
+Total train time:8d4h1m
+I1205 15:16:28.885109 137274321021824 utils.py:1231] [106500] l2_params = 238.2267869449069
+I1205 15:16:28.885318 137274321021824 utils.py:1231] [106500] train/loss = 3.605606347322464
+I1205 15:16:28.885423 137274321021824 utils.py:1231] [106500] l2_grads = 2.9120521545410156
+I1205 15:16:28.885500 137274321021824 utils.py:1231] [106500] lr = 8.707312454163345e-06
+I1205 15:16:28.885572 137274321021824 utils.py:1231] [106500] uptime = 667578.247932621
+I1205 15:16:28.885635 137274321021824 utils.py:1231] [106500] examples_seen = 109056000.0
+I1205 15:16:28.885694 137274321021824 utils.py:1231] [106500] progress = 0.9458007335506159
+I1205 15:16:28.885753 137274321021824 utils.py:1231] [106500] epoch = 85.12239231887803
+I1205 15:16:28.885820 137274321021824 utils.py:1231] [106500] img/sec/core = 164.21670856761978
+I1205 15:16:28.885890 137274321021824 utils.py:1231] [106500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.40408460114915
+I1205 15:16:28.885953 137274321021824 utils.py:1231] [106500] core_hours = 185.40408460114915
+I1205 15:16:28.886021 137274321021824 train.py:125] NOTE: Steps:106500/112603 [94.6%]
+Walltime:7d17h26m (0s eval)
+ETA:10h37m
+Total train time:8d4h1m
+I1205 15:21:40.650396 137274321021824 utils.py:1231] [106550] l2_params = 238.22157865103702
+I1205 15:21:40.650604 137274321021824 utils.py:1231] [106550] train/loss = 1.4421917349100113
+I1205 15:21:40.650700 137274321021824 utils.py:1231] [106550] l2_grads = 2.8770296573638916
+I1205 15:21:40.650772 137274321021824 utils.py:1231] [106550] lr = 8.565654447589945e-06
+I1205 15:21:40.650830 137274321021824 utils.py:1231] [106550] uptime = 667890.013192152
+I1205 15:21:40.650907 137274321021824 utils.py:1231] [106550] examples_seen = 109107200.0
+I1205 15:21:40.650976 137274321021824 utils.py:1231] [106550] progress = 0.9462447714536912
+I1205 15:21:40.651033 137274321021824 utils.py:1231] [106550] epoch = 85.16235588334698
+I1205 15:21:40.651085 137274321021824 utils.py:1231] [106550] img/sec/core = 164.22612345270852
+I1205 15:21:40.651144 137274321021824 utils.py:1231] [106550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.49068606213
+I1205 15:21:40.651198 137274321021824 utils.py:1231] [106550] core_hours = 185.49068606213
+I1205 15:21:40.651260 137274321021824 train.py:125] NOTE: Steps:106550/112603 [94.6%]
+Walltime:7d17h31m (0s eval)
+ETA:10h32m
+Total train time:8d4h1m
+I1205 15:26:52.438479 137274321021824 utils.py:1231] [106600] l2_params = 238.21601462116712
+I1205 15:26:52.438778 137274321021824 utils.py:1231] [106600] train/loss = 2.9114649295806885
+I1205 15:26:52.439007 137274321021824 utils.py:1231] [106600] l2_grads = 2.750624895095825
+I1205 15:26:52.439085 137274321021824 utils.py:1231] [106600] lr = 8.425148262185491e-06
+I1205 15:26:52.439140 137274321021824 utils.py:1231] [106600] uptime = 668201.801501297
+I1205 15:26:52.439192 137274321021824 utils.py:1231] [106600] examples_seen = 109158400.0
+I1205 15:26:52.439242 137274321021824 utils.py:1231] [106600] progress = 0.9466888093567667
+I1205 15:26:52.439291 137274321021824 utils.py:1231] [106600] epoch = 85.20231944781594
+I1205 15:26:52.439342 137274321021824 utils.py:1231] [106600] img/sec/core = 164.21398268716237
+I1205 15:26:52.439407 137274321021824 utils.py:1231] [106600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.5772939257814
+I1205 15:26:52.439457 137274321021824 utils.py:1231] [106600] core_hours = 185.5772939257814
+I1205 15:26:52.439516 137274321021824 train.py:125] NOTE: Steps:106600/112603 [94.7%]
+Walltime:7d17h36m (0s eval)
+ETA:10h27m
+Total train time:8d4h1m
+I1205 15:32:04.226023 137274321021824 utils.py:1231] [106650] l2_params = 238.2105106637561
+I1205 15:32:04.226266 137274321021824 utils.py:1231] [106650] train/loss = 3.3439285159111023
+I1205 15:32:04.226379 137274321021824 utils.py:1231] [106650] l2_grads = 2.7818243503570557
+I1205 15:32:04.226459 137274321021824 utils.py:1231] [106650] lr = 8.285794227267833e-06
+I1205 15:32:04.226516 137274321021824 utils.py:1231] [106650] uptime = 668513.588877953
+I1205 15:32:04.226569 137274321021824 utils.py:1231] [106650] examples_seen = 109209600.0
+I1205 15:32:04.226621 137274321021824 utils.py:1231] [106650] progress = 0.9471328472598421
+I1205 15:32:04.226668 137274321021824 utils.py:1231] [106650] epoch = 85.24228301228489
+I1205 15:32:04.226716 137274321021824 utils.py:1231] [106650] img/sec/core = 164.21447381590326
+I1205 15:32:04.226775 137274321021824 utils.py:1231] [106650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.66390153040805
+I1205 15:32:04.226822 137274321021824 utils.py:1231] [106650] core_hours = 185.66390153040805
+I1205 15:32:04.226898 137274321021824 train.py:125] NOTE: Steps:106650/112603 [94.7%]
+Walltime:7d17h41m (0s eval)
+ETA:10h21m
+Total train time:8d4h1m
+I1205 15:37:15.915989 137274321021824 utils.py:1231] [106700] l2_params = 238.20505011666722
+I1205 15:37:15.916197 137274321021824 utils.py:1231] [106700] train/loss = 1.5662232637405396
+I1205 15:37:15.916311 137274321021824 utils.py:1231] [106700] l2_grads = 2.9189867973327637
+I1205 15:37:15.916384 137274321021824 utils.py:1231] [106700] lr = 8.147592669454026e-06
+I1205 15:37:15.916444 137274321021824 utils.py:1231] [106700] uptime = 668825.278805083
+I1205 15:37:15.916507 137274321021824 utils.py:1231] [106700] examples_seen = 109260800.0
+I1205 15:37:15.916565 137274321021824 utils.py:1231] [106700] progress = 0.9475768851629175
+I1205 15:37:15.916621 137274321021824 utils.py:1231] [106700] epoch = 85.28224657675385
+I1205 15:37:15.916678 137274321021824 utils.py:1231] [106700] img/sec/core = 164.26581529740747
+I1205 15:37:15.916741 137274321021824 utils.py:1231] [106700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.75048206572194
+I1205 15:37:15.916796 137274321021824 utils.py:1231] [106700] core_hours = 185.75048206572194
+I1205 15:37:15.916862 137274321021824 train.py:125] NOTE: Steps:106700/112603 [94.8%]
+Walltime:7d17h47m (0s eval)
+ETA:10h16m
+Total train time:8d4h1m
+I1205 15:42:27.724255 137274321021824 utils.py:1231] [106750] l2_params = 238.19954133307886
+I1205 15:42:27.724461 137274321021824 utils.py:1231] [106750] train/loss = 1.5064576119184494
+I1205 15:42:27.724562 137274321021824 utils.py:1231] [106750] l2_grads = 2.7931487560272217
+I1205 15:42:27.724641 137274321021824 utils.py:1231] [106750] lr = 8.010543912660262e-06
+I1205 15:42:27.724701 137274321021824 utils.py:1231] [106750] uptime = 669137.087063208
+I1205 15:42:27.724761 137274321021824 utils.py:1231] [106750] examples_seen = 109312000.0
+I1205 15:42:27.724818 137274321021824 utils.py:1231] [106750] progress = 0.948020923065993
+I1205 15:42:27.724877 137274321021824 utils.py:1231] [106750] epoch = 85.32221014122281
+I1205 15:42:27.724941 137274321021824 utils.py:1231] [106750] img/sec/core = 164.2034765464051
+I1205 15:42:27.724999 137274321021824 utils.py:1231] [106750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.83709547075665
+I1205 15:42:27.725051 137274321021824 utils.py:1231] [106750] core_hours = 185.83709547075665
+I1205 15:42:27.725115 137274321021824 train.py:125] NOTE: Steps:106750/112603 [94.8%]
+Walltime:7d17h52m (0s eval)
+ETA:10h11m
+Total train time:8d4h1m
+I1205 15:47:39.510430 137274321021824 utils.py:1231] [106800] l2_params = 238.1944878047895
+I1205 15:47:39.510688 137274321021824 utils.py:1231] [106800] train/loss = 2.193016469478607
+I1205 15:47:39.510823 137274321021824 utils.py:1231] [106800] l2_grads = 2.6502065658569336
+I1205 15:47:39.510917 137274321021824 utils.py:1231] [106800] lr = 7.874648278100641e-06
+I1205 15:47:39.510984 137274321021824 utils.py:1231] [106800] uptime = 669448.8733446139
+I1205 15:47:39.511049 137274321021824 utils.py:1231] [106800] examples_seen = 109363200.0
+I1205 15:47:39.511109 137274321021824 utils.py:1231] [106800] progress = 0.9484649609690683
+I1205 15:47:39.511177 137274321021824 utils.py:1231] [106800] epoch = 85.36217370569176
+I1205 15:47:39.511246 137274321021824 utils.py:1231] [106800] img/sec/core = 164.21505067228154
+I1205 15:47:39.511312 137274321021824 utils.py:1231] [106800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 185.92370277114722
+I1205 15:47:39.511373 137274321021824 utils.py:1231] [106800] core_hours = 185.92370277114722
+I1205 15:47:39.511455 137274321021824 train.py:125] NOTE: Steps:106800/112603 [94.8%]
+Walltime:7d17h57m (0s eval)
+ETA:10h6m
+Total train time:8d4h1m
+I1205 15:52:51.308320 137274321021824 utils.py:1231] [106850] l2_params = 238.18949253549619
+I1205 15:52:51.308562 137274321021824 utils.py:1231] [106850] train/loss = 1.4624857753515244
+I1205 15:52:51.308681 137274321021824 utils.py:1231] [106850] l2_grads = 2.7233235836029053
+I1205 15:52:51.308743 137274321021824 utils.py:1231] [106850] lr = 7.739906084286541e-06
+I1205 15:52:51.308795 137274321021824 utils.py:1231] [106850] uptime = 669760.671157181
+I1205 15:52:51.308847 137274321021824 utils.py:1231] [106850] examples_seen = 109414400.0
+I1205 15:52:51.308900 137274321021824 utils.py:1231] [106850] progress = 0.9489089988721437
+I1205 15:52:51.308948 137274321021824 utils.py:1231] [106850] epoch = 85.40213727016072
+I1205 15:52:51.309000 137274321021824 utils.py:1231] [106850] img/sec/core = 164.20897753727124
+I1205 15:52:51.309056 137274321021824 utils.py:1231] [106850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.01031327463807
+I1205 15:52:51.309106 137274321021824 utils.py:1231] [106850] core_hours = 186.01031327463807
+I1205 15:52:51.309165 137274321021824 train.py:125] NOTE: Steps:106850/112603 [94.9%]
+Walltime:7d18h2m (0s eval)
+ETA:10h0m
+Total train time:8d4h1m
+I1205 15:58:03.102320 137274321021824 utils.py:1231] [106900] l2_params = 238.18458391359914
+I1205 15:58:03.102539 137274321021824 utils.py:1231] [106900] train/loss = 1.4994824826717377
+I1205 15:58:03.102658 137274321021824 utils.py:1231] [106900] l2_grads = 2.8939332962036133
+I1205 15:58:03.102746 137274321021824 utils.py:1231] [106900] lr = 7.606317647026073e-06
+I1205 15:58:03.102819 137274321021824 utils.py:1231] [106900] uptime = 670072.465181413
+I1205 15:58:03.102892 137274321021824 utils.py:1231] [106900] examples_seen = 109465600.0
+I1205 15:58:03.102943 137274321021824 utils.py:1231] [106900] progress = 0.9493530367752191
+I1205 15:58:03.102990 137274321021824 utils.py:1231] [106900] epoch = 85.44210083462968
+I1205 15:58:03.103041 137274321021824 utils.py:1231] [106900] img/sec/core = 164.21097269622337
+I1205 15:58:03.103096 137274321021824 utils.py:1231] [106900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.0969227258136
+I1205 15:58:03.103147 137274321021824 utils.py:1231] [106900] core_hours = 186.0969227258136
+I1205 15:58:03.103209 137274321021824 train.py:125] NOTE: Steps:106900/112603 [94.9%]
+Walltime:7d18h7m (0s eval)
+ETA:9h55m
+Total train time:8d4h1m
+I1205 16:03:14.911077 137274321021824 utils.py:1231] [106950] l2_params = 238.179859722502
+I1205 16:03:14.911344 137274321021824 utils.py:1231] [106950] train/loss = 1.413047343492508
+I1205 16:03:14.911487 137274321021824 utils.py:1231] [106950] l2_grads = 2.8482046127319336
+I1205 16:03:14.911572 137274321021824 utils.py:1231] [106950] lr = 7.473883279423053e-06
+I1205 16:03:14.911639 137274321021824 utils.py:1231] [106950] uptime = 670384.2740004869
+I1205 16:03:14.911711 137274321021824 utils.py:1231] [106950] examples_seen = 109516800.0
+I1205 16:03:14.911775 137274321021824 utils.py:1231] [106950] progress = 0.9497970746782946
+I1205 16:03:14.911840 137274321021824 utils.py:1231] [106950] epoch = 85.48206439909863
+I1205 16:03:14.911915 137274321021824 utils.py:1231] [106950] img/sec/core = 164.2031811417657
+I1205 16:03:14.911975 137274321021824 utils.py:1231] [106950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.18353628666748
+I1205 16:03:14.912029 137274321021824 utils.py:1231] [106950] core_hours = 186.18353628666748
+I1205 16:03:14.912114 137274321021824 train.py:125] NOTE: Steps:106950/112603 [95.0%]
+Walltime:7d18h13m (0s eval)
+ETA:9h50m
+Total train time:8d4h1m
+I1205 16:08:26.704467 137274321021824 utils.py:1231] [107000] l2_params = 238.17483640367655
+I1205 16:08:26.704687 137274321021824 utils.py:1231] [107000] train/loss = 1.5695439130067825
+I1205 16:08:26.704802 137274321021824 utils.py:1231] [107000] l2_grads = 2.84242844581604
+I1205 16:08:26.704876 137274321021824 utils.py:1231] [107000] lr = 7.342603291876466e-06
+I1205 16:08:26.704942 137274321021824 utils.py:1231] [107000] uptime = 670696.067303814
+I1205 16:08:26.705003 137274321021824 utils.py:1231] [107000] examples_seen = 109568000.0
+I1205 16:08:26.705061 137274321021824 utils.py:1231] [107000] progress = 0.9502411125813699
+I1205 16:08:26.705117 137274321021824 utils.py:1231] [107000] epoch = 85.5220279635676
+I1205 16:08:26.705177 137274321021824 utils.py:1231] [107000] img/sec/core = 164.21135237242157
+I1205 16:08:26.705239 137274321021824 utils.py:1231] [107000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.27014553759167
+I1205 16:08:26.705297 137274321021824 utils.py:1231] [107000] core_hours = 186.27014553759167
+I1205 16:08:26.705363 137274321021824 train.py:125] NOTE: Steps:107000/112603 [95.0%]
+Walltime:7d18h18m (0s eval)
+ETA:9h45m
+Total train time:8d4h1m
+I1205 16:13:38.846014 137274321021824 utils.py:1231] [107050] l2_params = 238.16997205320277
+I1205 16:13:38.846236 137274321021824 utils.py:1231] [107050] train/loss = 3.7427389919757843
+I1205 16:13:38.846343 137274321021824 utils.py:1231] [107050] l2_grads = 3.080749988555908
+I1205 16:13:38.846420 137274321021824 utils.py:1231] [107050] lr = 7.212477992079683e-06
+I1205 16:13:38.846482 137274321021824 utils.py:1231] [107050] uptime = 671008.208843308
+I1205 16:13:38.846543 137274321021824 utils.py:1231] [107050] examples_seen = 109619200.0
+I1205 16:13:38.846606 137274321021824 utils.py:1231] [107050] progress = 0.9506851504844454
+I1205 16:13:38.846665 137274321021824 utils.py:1231] [107050] epoch = 85.56199152803654
+I1205 16:13:38.846724 137274321021824 utils.py:1231] [107050] img/sec/core = 164.02815236636386
+I1205 16:13:38.846787 137274321021824 utils.py:1231] [107050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.35685152078446
+I1205 16:13:38.846844 137274321021824 utils.py:1231] [107050] core_hours = 186.35685152078446
+I1205 16:13:38.846922 137274321021824 train.py:125] NOTE: Steps:107050/112603 [95.1%]
+Walltime:7d18h23m (0s eval)
+ETA:9h40m
+Total train time:8d4h1m
+I1205 16:18:50.636272 137274321021824 utils.py:1231] [107100] l2_params = 238.16529218354063
+I1205 16:18:50.636490 137274321021824 utils.py:1231] [107100] train/loss = 2.309813439846039
+I1205 16:18:50.636600 137274321021824 utils.py:1231] [107100] l2_grads = 2.7008161544799805
+I1205 16:18:50.636682 137274321021824 utils.py:1231] [107100] lr = 7.083507685019521e-06
+I1205 16:18:50.636747 137274321021824 utils.py:1231] [107100] uptime = 671319.999107692
+I1205 16:18:50.636812 137274321021824 utils.py:1231] [107100] examples_seen = 109670400.0
+I1205 16:18:50.636875 137274321021824 utils.py:1231] [107100] progress = 0.9511291883875208
+I1205 16:18:50.636943 137274321021824 utils.py:1231] [107100] epoch = 85.6019550925055
+I1205 16:18:50.637004 137274321021824 utils.py:1231] [107100] img/sec/core = 164.21295290009832
+I1205 16:18:50.637070 137274321021824 utils.py:1231] [107100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.4434599275578
+I1205 16:18:50.637131 137274321021824 utils.py:1231] [107100] core_hours = 186.4434599275578
+I1205 16:18:50.637202 137274321021824 train.py:125] NOTE: Steps:107100/112603 [95.1%]
+Walltime:7d18h28m (0s eval)
+ETA:9h34m
+Total train time:8d4h1m
+I1205 16:24:02.425867 137274321021824 utils.py:1231] [107150] l2_params = 238.16067857357646
+I1205 16:24:02.426117 137274321021824 utils.py:1231] [107150] train/loss = 1.3901580274105072
+I1205 16:24:02.426225 137274321021824 utils.py:1231] [107150] l2_grads = 2.8493690490722656
+I1205 16:24:02.426297 137274321021824 utils.py:1231] [107150] lr = 6.9556926729760105e-06
+I1205 16:24:02.426358 137274321021824 utils.py:1231] [107150] uptime = 671631.788720059
+I1205 16:24:02.426410 137274321021824 utils.py:1231] [107150] examples_seen = 109721600.0
+I1205 16:24:02.426459 137274321021824 utils.py:1231] [107150] progress = 0.9515732262905962
+I1205 16:24:02.426506 137274321021824 utils.py:1231] [107150] epoch = 85.64191865697447
+I1205 16:24:02.426555 137274321021824 utils.py:1231] [107150] img/sec/core = 164.21329630361294
+I1205 16:24:02.426621 137274321021824 utils.py:1231] [107150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.53006815321527
+I1205 16:24:02.426681 137274321021824 utils.py:1231] [107150] core_hours = 186.53006815321527
+I1205 16:24:02.426743 137274321021824 train.py:125] NOTE: Steps:107150/112603 [95.2%]
+Walltime:7d18h33m (0s eval)
+ETA:9h29m
+Total train time:8d4h1m
+I1205 16:29:14.211315 137274321021824 utils.py:1231] [107200] l2_params = 238.15631863087884
+I1205 16:29:14.211518 137274321021824 utils.py:1231] [107200] train/loss = 1.5401839762926102
+I1205 16:29:14.211624 137274321021824 utils.py:1231] [107200] l2_grads = 3.0231263637542725
+I1205 16:29:14.211697 137274321021824 utils.py:1231] [107200] lr = 6.8290332555212965e-06
+I1205 16:29:14.211762 137274321021824 utils.py:1231] [107200] uptime = 671943.5741237929
+I1205 16:29:14.211819 137274321021824 utils.py:1231] [107200] examples_seen = 109772800.0
+I1205 16:29:14.211894 137274321021824 utils.py:1231] [107200] progress = 0.9520172641936716
+I1205 16:29:14.211950 137274321021824 utils.py:1231] [107200] epoch = 85.68188222144342
+I1205 16:29:14.212005 137274321021824 utils.py:1231] [107200] img/sec/core = 164.21551293558454
+I1205 16:29:14.212063 137274321021824 utils.py:1231] [107200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.61667520980802
+I1205 16:29:14.212119 137274321021824 utils.py:1231] [107200] core_hours = 186.61667520980802
+I1205 16:29:14.212179 137274321021824 train.py:125] NOTE: Steps:107200/112603 [95.2%]
+Walltime:7d18h39m (0s eval)
+ETA:9h24m
+Total train time:8d4h1m
+I1205 16:34:26.018638 137274321021824 utils.py:1231] [107250] l2_params = 238.15192252688485
+I1205 16:34:26.018860 137274321021824 utils.py:1231] [107250] train/loss = 1.8179631233215332
+I1205 16:34:26.018969 137274321021824 utils.py:1231] [107250] l2_grads = 2.848109722137451
+I1205 16:34:26.019035 137274321021824 utils.py:1231] [107250] lr = 6.703529729519084e-06
+I1205 16:34:26.019088 137274321021824 utils.py:1231] [107250] uptime = 672255.381449996
+I1205 16:34:26.019141 137274321021824 utils.py:1231] [107250] examples_seen = 109824000.0
+I1205 16:34:26.019193 137274321021824 utils.py:1231] [107250] progress = 0.952461302096747
+I1205 16:34:26.019249 137274321021824 utils.py:1231] [107250] epoch = 85.72184578591238
+I1205 16:34:26.019300 137274321021824 utils.py:1231] [107250] img/sec/core = 164.2039673136188
+I1205 16:34:26.019358 137274321021824 utils.py:1231] [107250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.70328835597556
+I1205 16:34:26.019410 137274321021824 utils.py:1231] [107250] core_hours = 186.70328835597556
+I1205 16:34:26.019476 137274321021824 train.py:125] NOTE: Steps:107250/112603 [95.2%]
+Walltime:7d18h44m (0s eval)
+ETA:9h19m
+Total train time:8d4h1m
+I1205 16:39:37.808701 137274321021824 utils.py:1231] [107300] l2_params = 238.14764091033854
+I1205 16:39:37.808912 137274321021824 utils.py:1231] [107300] train/loss = 2.3719807267189026
+I1205 16:39:37.809021 137274321021824 utils.py:1231] [107300] l2_grads = 2.693272352218628
+I1205 16:39:37.809100 137274321021824 utils.py:1231] [107300] lr = 6.5791823891237354e-06
+I1205 16:39:37.809161 137274321021824 utils.py:1231] [107300] uptime = 672567.1715220109
+I1205 16:39:37.809222 137274321021824 utils.py:1231] [107300] examples_seen = 109875200.0
+I1205 16:39:37.809279 137274321021824 utils.py:1231] [107300] progress = 0.9529053399998224
+I1205 16:39:37.809336 137274321021824 utils.py:1231] [107300] epoch = 85.76180935038133
+I1205 16:39:37.809400 137274321021824 utils.py:1231] [107300] img/sec/core = 164.21305421666048
+I1205 16:39:37.809463 137274321021824 utils.py:1231] [107300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.789896709313
+I1205 16:39:37.809519 137274321021824 utils.py:1231] [107300] core_hours = 186.789896709313
+I1205 16:39:37.809581 137274321021824 train.py:125] NOTE: Steps:107300/112603 [95.3%]
+Walltime:7d18h49m (0s eval)
+ETA:9h13m
+Total train time:8d4h1m
+I1205 16:44:49.609932 137274321021824 utils.py:1231] [107350] l2_params = 238.14368731720626
+I1205 16:44:49.610142 137274321021824 utils.py:1231] [107350] train/loss = 1.4228817224502563
+I1205 16:44:49.610239 137274321021824 utils.py:1231] [107350] l2_grads = 2.6943702697753906
+I1205 16:44:49.610319 137274321021824 utils.py:1231] [107350] lr = 6.455991525779844e-06
+I1205 16:44:49.610379 137274321021824 utils.py:1231] [107350] uptime = 672878.9727413
+I1205 16:44:49.610442 137274321021824 utils.py:1231] [107350] examples_seen = 109926400.0
+I1205 16:44:49.610500 137274321021824 utils.py:1231] [107350] progress = 0.9533493779028978
+I1205 16:44:49.610557 137274321021824 utils.py:1231] [107350] epoch = 85.80177291485029
+I1205 16:44:49.610615 137274321021824 utils.py:1231] [107350] img/sec/core = 164.20718339953592
+I1205 16:44:49.610677 137274321021824 utils.py:1231] [107350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.87650815911556
+I1205 16:44:49.610734 137274321021824 utils.py:1231] [107350] core_hours = 186.87650815911556
+I1205 16:44:49.610807 137274321021824 train.py:125] NOTE: Steps:107350/112603 [95.3%]
+Walltime:7d18h54m (0s eval)
+ETA:9h8m
+Total train time:8d4h1m
+I1205 16:50:01.400597 137274321021824 utils.py:1231] [107400] l2_params = 238.13897882103697
+I1205 16:50:01.400849 137274321021824 utils.py:1231] [107400] train/loss = 1.9579756259918213
+I1205 16:50:01.400995 137274321021824 utils.py:1231] [107400] l2_grads = 2.7055954933166504
+I1205 16:50:01.401067 137274321021824 utils.py:1231] [107400] lr = 6.33395742822172e-06
+I1205 16:50:01.401125 137274321021824 utils.py:1231] [107400] uptime = 673190.763487257
+I1205 16:50:01.401186 137274321021824 utils.py:1231] [107400] examples_seen = 109977600.0
+I1205 16:50:01.401242 137274321021824 utils.py:1231] [107400] progress = 0.9537934158059732
+I1205 16:50:01.401292 137274321021824 utils.py:1231] [107400] epoch = 85.84173647931925
+I1205 16:50:01.401349 137274321021824 utils.py:1231] [107400] img/sec/core = 164.21269926680588
+I1205 16:50:01.401419 137274321021824 utils.py:1231] [107400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 186.96311669965914
+I1205 16:50:01.401470 137274321021824 utils.py:1231] [107400] core_hours = 186.96311669965914
+I1205 16:50:01.401537 137274321021824 train.py:125] NOTE: Steps:107400/112603 [95.4%]
+Walltime:7d18h59m (0s eval)
+ETA:9h3m
+Total train time:8d4h1m
+I1205 16:55:13.207695 137274321021824 utils.py:1231] [107450] l2_params = 238.13477661470688
+I1205 16:55:13.207983 137274321021824 utils.py:1231] [107450] train/loss = 3.4853015542030334
+I1205 16:55:13.208187 137274321021824 utils.py:1231] [107450] l2_grads = 2.8955533504486084
+I1205 16:55:13.208332 137274321021824 utils.py:1231] [107450] lr = 6.213080382471957e-06
+I1205 16:55:13.208441 137274321021824 utils.py:1231] [107450] uptime = 673502.570790275
+I1205 16:55:13.208540 137274321021824 utils.py:1231] [107450] examples_seen = 110028800.0
+I1205 16:55:13.208627 137274321021824 utils.py:1231] [107450] progress = 0.9542374537090486
+I1205 16:55:13.208739 137274321021824 utils.py:1231] [107450] epoch = 85.8817000437882
+I1205 16:55:13.208831 137274321021824 utils.py:1231] [107450] img/sec/core = 164.20397952332635
+I1205 16:55:13.208919 137274321021824 utils.py:1231] [107450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.0497298393864
+I1205 16:55:13.208989 137274321021824 utils.py:1231] [107450] core_hours = 187.0497298393864
+I1205 16:55:13.209073 137274321021824 train.py:125] NOTE: Steps:107450/112603 [95.4%]
+Walltime:7d19h5m (0s eval)
+ETA:8h58m
+Total train time:8d4h1m
+I1205 17:00:25.012003 137274321021824 utils.py:1231] [107500] l2_params = 238.13073022385748
+I1205 17:00:25.012296 137274321021824 utils.py:1231] [107500] train/loss = 1.3712719827890396
+I1205 17:00:25.012476 137274321021824 utils.py:1231] [107500] l2_grads = 2.7314555644989014
+I1205 17:00:25.012553 137274321021824 utils.py:1231] [107500] lr = 6.093360671841717e-06
+I1205 17:00:25.012614 137274321021824 utils.py:1231] [107500] uptime = 673814.3749757219
+I1205 17:00:25.012672 137274321021824 utils.py:1231] [107500] examples_seen = 110080000.0
+I1205 17:00:25.012725 137274321021824 utils.py:1231] [107500] progress = 0.954681491612124
+I1205 17:00:25.012782 137274321021824 utils.py:1231] [107500] epoch = 85.92166360825716
+I1205 17:00:25.012836 137274321021824 utils.py:1231] [107500] img/sec/core = 164.20562131522047
+I1205 17:00:25.012902 137274321021824 utils.py:1231] [107500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.13634211312166
+I1205 17:00:25.012959 137274321021824 utils.py:1231] [107500] core_hours = 187.13634211312166
+I1205 17:00:25.013024 137274321021824 train.py:125] NOTE: Steps:107500/112603 [95.5%]
+Walltime:7d19h10m (0s eval)
+ETA:8h53m
+Total train time:8d4h1m
+I1205 17:00:25.013126 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:107500/112603 [95.5%]
+Walltime:7d19h10m (0s eval)
+ETA:8h53m
+Total train time:8d4h1m
+I1205 17:02:02.872891 137274321021824 utils.py:1231] [107500] val/acc@1 = 0.7644491390306123
+I1205 17:02:02.873178 137274321021824 utils.py:1231] [107500] val/loss = 0.9242822094535341
+I1205 17:02:02.873372 137274321021824 utils.py:1231] [107500] z/secs/eval/val = 97.8601558599621
+I1205 17:02:02.873456 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.8601558599621
+I1205 17:07:14.602486 137274321021824 utils.py:1231] [107550] l2_params = 238.12700390998816
+I1205 17:07:14.602705 137274321021824 utils.py:1231] [107550] train/loss = 1.4887017905712128
+I1205 17:07:14.602817 137274321021824 utils.py:1231] [107550] l2_grads = 2.778320074081421
+I1205 17:07:14.602898 137274321021824 utils.py:1231] [107550] lr = 5.974798576929316e-06
+I1205 17:07:14.602959 137274321021824 utils.py:1231] [107550] uptime = 674223.965320779
+I1205 17:07:14.603019 137274321021824 utils.py:1231] [107550] examples_seen = 110131200.0
+I1205 17:07:14.603076 137274321021824 utils.py:1231] [107550] progress = 0.9551255295151995
+I1205 17:07:14.603131 137274321021824 utils.py:1231] [107550] epoch = 85.96162717272611
+I1205 17:07:14.603193 137274321021824 utils.py:1231] [107550] img/sec/core = 125.00294652420392
+I1205 17:07:14.603251 137274321021824 utils.py:1231] [107550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.25011720897083
+I1205 17:07:14.603330 137274321021824 utils.py:1231] [107550] core_hours = 187.25011720897083
+I1205 17:07:14.603404 137274321021824 train.py:125] NOTE: Steps:107550/112603 [95.5%]
+Walltime:7d19h17m (0s eval)
+ETA:8h47m
+Total train time:8d4h3m
+I1205 17:12:26.396059 137274321021824 utils.py:1231] [107600] l2_params = 238.12300087860564
+I1205 17:12:26.396286 137274321021824 utils.py:1231] [107600] train/loss = 1.4116557091474533
+I1205 17:12:26.396383 137274321021824 utils.py:1231] [107600] l2_grads = 2.810039520263672
+I1205 17:12:26.396457 137274321021824 utils.py:1231] [107600] lr = 5.857394375620027e-06
+I1205 17:12:26.396511 137274321021824 utils.py:1231] [107600] uptime = 674535.758872859
+I1205 17:12:26.396565 137274321021824 utils.py:1231] [107600] examples_seen = 110182400.0
+I1205 17:12:26.396615 137274321021824 utils.py:1231] [107600] progress = 0.9555695674182748
+I1205 17:12:26.396666 137274321021824 utils.py:1231] [107600] epoch = 86.00159073719507
+I1205 17:12:26.396718 137274321021824 utils.py:1231] [107600] img/sec/core = 164.2112213625044
+I1205 17:12:26.396775 137274321021824 utils.py:1231] [107600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.33672652899304
+I1205 17:12:26.396828 137274321021824 utils.py:1231] [107600] core_hours = 187.33672652899304
+I1205 17:12:26.396896 137274321021824 train.py:125] NOTE: Steps:107600/112603 [95.6%]
+Walltime:7d19h22m (0s eval)
+ETA:8h42m
+Total train time:8d4h3m
+I1205 17:17:38.124112 137274321021824 utils.py:1231] [107650] l2_params = 238.11931963926324
+I1205 17:17:38.124374 137274321021824 utils.py:1231] [107650] train/loss = 2.8129963278770447
+I1205 17:17:38.124485 137274321021824 utils.py:1231] [107650] l2_grads = 2.7113518714904785
+I1205 17:17:38.124564 137274321021824 utils.py:1231] [107650] lr = 5.741148343085121e-06
+I1205 17:17:38.124637 137274321021824 utils.py:1231] [107650] uptime = 674847.486997603
+I1205 17:17:38.124699 137274321021824 utils.py:1231] [107650] examples_seen = 110233600.0
+I1205 17:17:38.124761 137274321021824 utils.py:1231] [107650] progress = 0.9560136053213503
+I1205 17:17:38.124827 137274321021824 utils.py:1231] [107650] epoch = 86.04155430166404
+I1205 17:17:38.124903 137274321021824 utils.py:1231] [107650] img/sec/core = 164.24568698138043
+I1205 17:17:38.124969 137274321021824 utils.py:1231] [107650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.42331767475525
+I1205 17:17:38.125052 137274321021824 utils.py:1231] [107650] core_hours = 187.42331767475525
+I1205 17:17:38.125160 137274321021824 train.py:125] NOTE: Steps:107650/112603 [95.6%]
+Walltime:7d19h27m (0s eval)
+ETA:8h37m
+Total train time:8d4h3m
+I1205 17:22:49.869972 137274321021824 utils.py:1231] [107700] l2_params = 238.11569299006402
+I1205 17:22:49.870184 137274321021824 utils.py:1231] [107700] train/loss = 2.1362767964601517
+I1205 17:22:49.870289 137274321021824 utils.py:1231] [107700] l2_grads = 2.6649296283721924
+I1205 17:22:49.870369 137274321021824 utils.py:1231] [107700] lr = 5.6260607517813776e-06
+I1205 17:22:49.870436 137274321021824 utils.py:1231] [107700] uptime = 675159.232797266
+I1205 17:22:49.870497 137274321021824 utils.py:1231] [107700] examples_seen = 110284800.0
+I1205 17:22:49.870556 137274321021824 utils.py:1231] [107700] progress = 0.9564576432244256
+I1205 17:22:49.870614 137274321021824 utils.py:1231] [107700] epoch = 86.08151786613298
+I1205 17:22:49.870678 137274321021824 utils.py:1231] [107700] img/sec/core = 164.2363748135367
+I1205 17:22:49.870741 137274321021824 utils.py:1231] [107700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.50991373021722
+I1205 17:22:49.870800 137274321021824 utils.py:1231] [107700] core_hours = 187.50991373021722
+I1205 17:22:49.870868 137274321021824 train.py:125] NOTE: Steps:107700/112603 [95.6%]
+Walltime:7d19h32m (0s eval)
+ETA:8h32m
+Total train time:8d4h2m
+I1205 17:28:01.668013 137274321021824 utils.py:1231] [107750] l2_params = 238.1121851674336
+I1205 17:28:01.668270 137274321021824 utils.py:1231] [107750] train/loss = 1.452351838350296
+I1205 17:28:01.668400 137274321021824 utils.py:1231] [107750] l2_grads = 2.9010701179504395
+I1205 17:28:01.668483 137274321021824 utils.py:1231] [107750] lr = 5.512131871450583e-06
+I1205 17:28:01.668540 137274321021824 utils.py:1231] [107750] uptime = 675471.030902052
+I1205 17:28:01.668598 137274321021824 utils.py:1231] [107750] examples_seen = 110336000.0
+I1205 17:28:01.668651 137274321021824 utils.py:1231] [107750] progress = 0.9569016811275011
+I1205 17:28:01.668710 137274321021824 utils.py:1231] [107750] epoch = 86.12148143060195
+I1205 17:28:01.668765 137274321021824 utils.py:1231] [107750] img/sec/core = 164.20882363972822
+I1205 17:28:01.668827 137274321021824 utils.py:1231] [107750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.59652431487999
+I1205 17:28:01.668890 137274321021824 utils.py:1231] [107750] core_hours = 187.59652431487999
+I1205 17:28:01.668969 137274321021824 train.py:125] NOTE: Steps:107750/112603 [95.7%]
+Walltime:7d19h37m (0s eval)
+ETA:8h26m
+Total train time:8d4h2m
+I1205 17:33:13.458824 137274321021824 utils.py:1231] [107800] l2_params = 238.10867492216903
+I1205 17:33:13.459040 137274321021824 utils.py:1231] [107800] train/loss = 2.3825110495090485
+I1205 17:33:13.459151 137274321021824 utils.py:1231] [107800] l2_grads = 2.805629014968872
+I1205 17:33:13.459226 137274321021824 utils.py:1231] [107800] lr = 5.399361969118581e-06
+I1205 17:33:13.459290 137274321021824 utils.py:1231] [107800] uptime = 675782.821651077
+I1205 17:33:13.459352 137274321021824 utils.py:1231] [107800] examples_seen = 110387200.0
+I1205 17:33:13.459424 137274321021824 utils.py:1231] [107800] progress = 0.9573457190305764
+I1205 17:33:13.459486 137274321021824 utils.py:1231] [107800] epoch = 86.1614449950709
+I1205 17:33:13.459545 137274321021824 utils.py:1231] [107800] img/sec/core = 164.2126976508977
+I1205 17:33:13.459607 137274321021824 utils.py:1231] [107800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.68313285627582
+I1205 17:33:13.459662 137274321021824 utils.py:1231] [107800] core_hours = 187.68313285627582
+I1205 17:33:13.459727 137274321021824 train.py:125] NOTE: Steps:107800/112603 [95.7%]
+Walltime:7d19h43m (0s eval)
+ETA:8h21m
+Total train time:8d4h2m
+I1205 17:38:25.253273 137274321021824 utils.py:1231] [107850] l2_params = 238.10502682209034
+I1205 17:38:25.253537 137274321021824 utils.py:1231] [107850] train/loss = 1.9331664144992828
+I1205 17:38:25.253678 137274321021824 utils.py:1231] [107850] l2_grads = 2.7586727142333984
+I1205 17:38:25.253773 137274321021824 utils.py:1231] [107850] lr = 5.287751309094891e-06
+I1205 17:38:25.253873 137274321021824 utils.py:1231] [107850] uptime = 676094.616230261
+I1205 17:38:25.253967 137274321021824 utils.py:1231] [107850] examples_seen = 110438400.0
+I1205 17:38:25.254056 137274321021824 utils.py:1231] [107850] progress = 0.9577897569336519
+I1205 17:38:25.254176 137274321021824 utils.py:1231] [107850] epoch = 86.20140855953986
+I1205 17:38:25.254256 137274321021824 utils.py:1231] [107850] img/sec/core = 164.21068042295877
+I1205 17:38:25.254340 137274321021824 utils.py:1231] [107850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.76974246160475
+I1205 17:38:25.254414 137274321021824 utils.py:1231] [107850] core_hours = 187.76974246160475
+I1205 17:38:25.254490 137274321021824 train.py:125] NOTE: Steps:107850/112603 [95.8%]
+Walltime:7d19h48m (0s eval)
+ETA:8h16m
+Total train time:8d4h2m
+I1205 17:43:37.038938 137274321021824 utils.py:1231] [107900] l2_params = 238.10124168445728
+I1205 17:43:37.039161 137274321021824 utils.py:1231] [107900] train/loss = 1.3901874870061874
+I1205 17:43:37.039264 137274321021824 utils.py:1231] [107900] l2_grads = 2.895263433456421
+I1205 17:43:37.039342 137274321021824 utils.py:1231] [107900] lr = 5.1773001529719314e-06
+I1205 17:43:37.039411 137274321021824 utils.py:1231] [107900] uptime = 676406.40177158
+I1205 17:43:37.039474 137274321021824 utils.py:1231] [107900] examples_seen = 110489600.0
+I1205 17:43:37.039533 137274321021824 utils.py:1231] [107900] progress = 0.9582337948367272
+I1205 17:43:37.039592 137274321021824 utils.py:1231] [107900] epoch = 86.24137212400882
+I1205 17:43:37.039651 137274321021824 utils.py:1231] [107900] img/sec/core = 164.2154404704267
+I1205 17:43:37.039715 137274321021824 utils.py:1231] [107900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.85634955641552
+I1205 17:43:37.039773 137274321021824 utils.py:1231] [107900] core_hours = 187.85634955641552
+I1205 17:43:37.039843 137274321021824 train.py:125] NOTE: Steps:107900/112603 [95.8%]
+Walltime:7d19h53m (0s eval)
+ETA:8h11m
+Total train time:8d4h2m
+I1205 17:48:48.976046 137274321021824 utils.py:1231] [107950] l2_params = 238.09805433112896
+I1205 17:48:48.976314 137274321021824 utils.py:1231] [107950] train/loss = 1.519781306385994
+I1205 17:48:48.976490 137274321021824 utils.py:1231] [107950] l2_grads = 2.9437410831451416
+I1205 17:48:48.976591 137274321021824 utils.py:1231] [107950] lr = 5.068008759624456e-06
+I1205 17:48:48.976660 137274321021824 utils.py:1231] [107950] uptime = 676718.339021467
+I1205 17:48:48.976739 137274321021824 utils.py:1231] [107950] examples_seen = 110540800.0
+I1205 17:48:48.976812 137274321021824 utils.py:1231] [107950] progress = 0.9586778327398027
+I1205 17:48:48.976877 137274321021824 utils.py:1231] [107950] epoch = 86.28133568847777
+I1205 17:48:48.976942 137274321021824 utils.py:1231] [107950] img/sec/core = 164.13557540351908
+I1205 17:48:48.977005 137274321021824 utils.py:1231] [107950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 187.9429987924953
+I1205 17:48:48.977063 137274321021824 utils.py:1231] [107950] core_hours = 187.9429987924953
+I1205 17:48:48.977130 137274321021824 train.py:125] NOTE: Steps:107950/112603 [95.9%]
+Walltime:7d19h58m (0s eval)
+ETA:8h6m
+Total train time:8d4h2m
+I1205 17:54:00.780168 137274321021824 utils.py:1231] [108000] l2_params = 238.0949205877225
+I1205 17:54:00.780429 137274321021824 utils.py:1231] [108000] train/loss = 1.5058180540800095
+I1205 17:54:00.780560 137274321021824 utils.py:1231] [108000] l2_grads = 2.8213253021240234
+I1205 17:54:00.780650 137274321021824 utils.py:1231] [108000] lr = 4.959877385209223e-06
+I1205 17:54:00.780728 137274321021824 utils.py:1231] [108000] uptime = 677030.14308591
+I1205 17:54:00.780799 137274321021824 utils.py:1231] [108000] examples_seen = 110592000.0
+I1205 17:54:00.780865 137274321021824 utils.py:1231] [108000] progress = 0.9591218706428781
+I1205 17:54:00.780933 137274321021824 utils.py:1231] [108000] epoch = 86.32129925294673
+I1205 17:54:00.780991 137274321021824 utils.py:1231] [108000] img/sec/core = 164.2056850396255
+I1205 17:54:00.781056 137274321021824 utils.py:1231] [108000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.02961103261833
+I1205 17:54:00.781111 137274321021824 utils.py:1231] [108000] core_hours = 188.02961103261833
+I1205 17:54:00.781179 137274321021824 train.py:125] NOTE: Steps:108000/112603 [95.9%]
+Walltime:7d20h3m (0s eval)
+ETA:8h0m
+Total train time:8d4h2m
+I1205 17:59:12.951320 137274321021824 utils.py:1231] [108050] l2_params = 238.09172413917042
+I1205 17:59:12.951532 137274321021824 utils.py:1231] [108050] train/loss = 1.4829940795898438
+I1205 17:59:12.951644 137274321021824 utils.py:1231] [108050] l2_grads = 2.7973859310150146
+I1205 17:59:12.951724 137274321021824 utils.py:1231] [108050] lr = 4.852906283163787e-06
+I1205 17:59:12.951806 137274321021824 utils.py:1231] [108050] uptime = 677342.314165156
+I1205 17:59:12.951879 137274321021824 utils.py:1231] [108050] examples_seen = 110643200.0
+I1205 17:59:12.951959 137274321021824 utils.py:1231] [108050] progress = 0.9595659085459535
+I1205 17:59:12.952015 137274321021824 utils.py:1231] [108050] epoch = 86.36126281741568
+I1205 17:59:12.952071 137274321021824 utils.py:1231] [108050] img/sec/core = 164.01263090632588
+I1205 17:59:12.952141 137274321021824 utils.py:1231] [108050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.11632522129776
+I1205 17:59:12.952207 137274321021824 utils.py:1231] [108050] core_hours = 188.11632522129776
+I1205 17:59:12.952278 137274321021824 train.py:125] NOTE: Steps:108050/112603 [96.0%]
+Walltime:7d20h9m (0s eval)
+ETA:7h55m
+Total train time:8d4h2m
+I1205 18:04:24.767462 137274321021824 utils.py:1231] [108100] l2_params = 238.08853863128428
+I1205 18:04:24.767731 137274321021824 utils.py:1231] [108100] train/loss = 1.7672018110752106
+I1205 18:04:24.767849 137274321021824 utils.py:1231] [108100] l2_grads = 2.680154323577881
+I1205 18:04:24.767941 137274321021824 utils.py:1231] [108100] lr = 4.747095704206642e-06
+I1205 18:04:24.767998 137274321021824 utils.py:1231] [108100] uptime = 677654.13035987
+I1205 18:04:24.768054 137274321021824 utils.py:1231] [108100] examples_seen = 110694400.0
+I1205 18:04:24.768106 137274321021824 utils.py:1231] [108100] progress = 0.9600099464490289
+I1205 18:04:24.768157 137274321021824 utils.py:1231] [108100] epoch = 86.40122638188464
+I1205 18:04:24.768210 137274321021824 utils.py:1231] [108100] img/sec/core = 164.19929711140654
+I1205 18:04:24.768269 137274321021824 utils.py:1231] [108100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.20294083094055
+I1205 18:04:24.768322 137274321021824 utils.py:1231] [108100] core_hours = 188.20294083094055
+I1205 18:04:24.768388 137274321021824 train.py:125] NOTE: Steps:108100/112603 [96.0%]
+Walltime:7d20h14m (0s eval)
+ETA:7h50m
+Total train time:8d4h2m
+I1205 18:09:36.569930 137274321021824 utils.py:1231] [108150] l2_params = 238.08544332238625
+I1205 18:09:36.570193 137274321021824 utils.py:1231] [108150] train/loss = 1.3475275188684464
+I1205 18:09:36.570300 137274321021824 utils.py:1231] [108150] l2_grads = 2.7731332778930664
+I1205 18:09:36.570378 137274321021824 utils.py:1231] [108150] lr = 4.642445896335857e-06
+I1205 18:09:36.570437 137274321021824 utils.py:1231] [108150] uptime = 677965.932799791
+I1205 18:09:36.570491 137274321021824 utils.py:1231] [108150] examples_seen = 110745600.0
+I1205 18:09:36.570538 137274321021824 utils.py:1231] [108150] progress = 0.9604539843521043
+I1205 18:09:36.570583 137274321021824 utils.py:1231] [108150] epoch = 86.4411899463536
+I1205 18:09:36.570633 137274321021824 utils.py:1231] [108150] img/sec/core = 164.20654056773884
+I1205 18:09:36.570689 137274321021824 utils.py:1231] [108150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.2895526198075
+I1205 18:09:36.570741 137274321021824 utils.py:1231] [108150] core_hours = 188.2895526198075
+I1205 18:09:36.570802 137274321021824 train.py:125] NOTE: Steps:108150/112603 [96.0%]
+Walltime:7d20h19m (0s eval)
+ETA:7h45m
+Total train time:8d4h2m
+I1205 18:14:48.378293 137274321021824 utils.py:1231] [108200] l2_params = 238.08195582589562
+I1205 18:14:48.378544 137274321021824 utils.py:1231] [108200] train/loss = 2.597558945417404
+I1205 18:14:48.378672 137274321021824 utils.py:1231] [108200] l2_grads = 2.6974287033081055
+I1205 18:14:48.378763 137274321021824 utils.py:1231] [108200] lr = 4.538957104829124e-06
+I1205 18:14:48.378838 137274321021824 utils.py:1231] [108200] uptime = 678277.741195254
+I1205 18:14:48.378919 137274321021824 utils.py:1231] [108200] examples_seen = 110796800.0
+I1205 18:14:48.378978 137274321021824 utils.py:1231] [108200] progress = 0.9608980222551797
+I1205 18:14:48.379037 137274321021824 utils.py:1231] [108200] epoch = 86.48115351082255
+I1205 18:14:48.379104 137274321021824 utils.py:1231] [108200] img/sec/core = 164.20340422190023
+I1205 18:14:48.379167 137274321021824 utils.py:1231] [108200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.37616606299167
+I1205 18:14:48.379226 137274321021824 utils.py:1231] [108200] core_hours = 188.37616606299167
+I1205 18:14:48.379294 137274321021824 train.py:125] NOTE: Steps:108200/112603 [96.1%]
+Walltime:7d20h24m (0s eval)
+ETA:7h39m
+Total train time:8d4h2m
+I1205 18:20:00.170715 137274321021824 utils.py:1231] [108250] l2_params = 238.07922171655372
+I1205 18:20:00.170945 137274321021824 utils.py:1231] [108250] train/loss = 2.440184235572815
+I1205 18:20:00.171067 137274321021824 utils.py:1231] [108250] l2_grads = 2.5330147743225098
+I1205 18:20:00.171147 137274321021824 utils.py:1231] [108250] lr = 4.436629572243023e-06
+I1205 18:20:00.171210 137274321021824 utils.py:1231] [108250] uptime = 678589.533570583
+I1205 18:20:00.171270 137274321021824 utils.py:1231] [108250] examples_seen = 110848000.0
+I1205 18:20:00.171328 137274321021824 utils.py:1231] [108250] progress = 0.9613420601582551
+I1205 18:20:00.171384 137274321021824 utils.py:1231] [108250] epoch = 86.52111707529151
+I1205 18:20:00.171442 137274321021824 utils.py:1231] [108250] img/sec/core = 164.21184112017627
+I1205 18:20:00.171504 137274321021824 utils.py:1231] [108250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.4627750561386
+I1205 18:20:00.171561 137274321021824 utils.py:1231] [108250] core_hours = 188.4627750561386
+I1205 18:20:00.171627 137274321021824 train.py:125] NOTE: Steps:108250/112603 [96.1%]
+Walltime:7d20h29m (0s eval)
+ETA:7h34m
+Total train time:8d4h2m
+I1205 18:25:11.970578 137274321021824 utils.py:1231] [108300] l2_params = 238.07654445917436
+I1205 18:25:11.970773 137274321021824 utils.py:1231] [108300] train/loss = 1.443441465497017
+I1205 18:25:11.970874 137274321021824 utils.py:1231] [108300] l2_grads = 2.946077823638916
+I1205 18:25:11.970947 137274321021824 utils.py:1231] [108300] lr = 4.335463538412158e-06
+I1205 18:25:11.971001 137274321021824 utils.py:1231] [108300] uptime = 678901.333362441
+I1205 18:25:11.971053 137274321021824 utils.py:1231] [108300] examples_seen = 110899200.0
+I1205 18:25:11.971109 137274321021824 utils.py:1231] [108300] progress = 0.9617860980613305
+I1205 18:25:11.971167 137274321021824 utils.py:1231] [108300] epoch = 86.56108063976046
+I1205 18:25:11.971224 137274321021824 utils.py:1231] [108300] img/sec/core = 164.2079351461104
+I1205 18:25:11.971281 137274321021824 utils.py:1231] [108300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.5493861094325
+I1205 18:25:11.971332 137274321021824 utils.py:1231] [108300] core_hours = 188.5493861094325
+I1205 18:25:12.198701 137274321021824 train.py:125] NOTE: Steps:108300/112603 [96.2%]
+Walltime:7d20h35m (0s eval)
+ETA:7h29m
+Total train time:8d4h2m
+I1205 18:30:24.006091 137274321021824 utils.py:1231] [108350] l2_params = 238.07366175785666
+I1205 18:30:24.006358 137274321021824 utils.py:1231] [108350] train/loss = 1.6802430003881454
+I1205 18:30:24.006458 137274321021824 utils.py:1231] [108350] l2_grads = 2.741751194000244
+I1205 18:30:24.006535 137274321021824 utils.py:1231] [108350] lr = 4.235459240449071e-06
+I1205 18:30:24.006587 137274321021824 utils.py:1231] [108350] uptime = 679213.368950155
+I1205 18:30:24.006638 137274321021824 utils.py:1231] [108350] examples_seen = 110950400.0
+I1205 18:30:24.006685 137274321021824 utils.py:1231] [108350] progress = 0.9622301359644059
+I1205 18:30:24.006732 137274321021824 utils.py:1231] [108350] epoch = 86.60104420422942
+I1205 18:30:24.006781 137274321021824 utils.py:1231] [108350] img/sec/core = 164.0838481761198
+I1205 18:30:24.006836 137274321021824 utils.py:1231] [108350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.63606266157527
+I1205 18:30:24.006888 137274321021824 utils.py:1231] [108350] core_hours = 188.63606266157527
+I1205 18:30:24.006948 137274321021824 train.py:125] NOTE: Steps:108350/112603 [96.2%]
+Walltime:7d20h40m (0s eval)
+ETA:7h24m
+Total train time:8d4h2m
+I1205 18:35:35.791546 137274321021824 utils.py:1231] [108400] l2_params = 238.0711220655621
+I1205 18:35:35.791806 137274321021824 utils.py:1231] [108400] train/loss = 2.956108510494232
+I1205 18:35:35.791918 137274321021824 utils.py:1231] [108400] l2_grads = 2.6721789836883545
+I1205 18:35:35.791990 137274321021824 utils.py:1231] [108400] lr = 4.1366169127431095e-06
+I1205 18:35:35.792039 137274321021824 utils.py:1231] [108400] uptime = 679525.154402079
+I1205 18:35:35.792088 137274321021824 utils.py:1231] [108400] examples_seen = 111001600.0
+I1205 18:35:35.792135 137274321021824 utils.py:1231] [108400] progress = 0.9626741738674813
+I1205 18:35:35.792182 137274321021824 utils.py:1231] [108400] epoch = 86.64100776869839
+I1205 18:35:35.792230 137274321021824 utils.py:1231] [108400] img/sec/core = 164.21548755419963
+I1205 18:35:35.792294 137274321021824 utils.py:1231] [108400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.72266973155413
+I1205 18:35:35.792364 137274321021824 utils.py:1231] [108400] core_hours = 188.72266973155413
+I1205 18:35:35.792430 137274321021824 train.py:125] NOTE: Steps:108400/112603 [96.3%]
+Walltime:7d20h45m (0s eval)
+ETA:7h19m
+Total train time:8d4h2m
+I1205 18:40:47.592309 137274321021824 utils.py:1231] [108450] l2_params = 238.06842506924107
+I1205 18:40:47.592600 137274321021824 utils.py:1231] [108450] train/loss = 1.3694669157266617
+I1205 18:40:47.592773 137274321021824 utils.py:1231] [108450] l2_grads = 2.77163028717041
+I1205 18:40:47.592871 137274321021824 utils.py:1231] [108450] lr = 4.0389367869605165e-06
+I1205 18:40:47.592954 137274321021824 utils.py:1231] [108450] uptime = 679836.9553156759
+I1205 18:40:47.593029 137274321021824 utils.py:1231] [108450] examples_seen = 111052800.0
+I1205 18:40:47.593097 137274321021824 utils.py:1231] [108450] progress = 0.9631182117705568
+I1205 18:40:47.593155 137274321021824 utils.py:1231] [108450] epoch = 86.68097133316734
+I1205 18:40:47.593212 137274321021824 utils.py:1231] [108450] img/sec/core = 164.20734438955802
+I1205 18:40:47.593270 137274321021824 utils.py:1231] [108450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.8092810964422
+I1205 18:40:47.593325 137274321021824 utils.py:1231] [108450] core_hours = 188.8092810964422
+I1205 18:40:47.593391 137274321021824 train.py:125] NOTE: Steps:108450/112603 [96.3%]
+Walltime:7d20h50m (0s eval)
+ETA:7h13m
+Total train time:8d4h2m
+I1205 18:45:59.379722 137274321021824 utils.py:1231] [108500] l2_params = 238.0656625215138
+I1205 18:45:59.379953 137274321021824 utils.py:1231] [108500] train/loss = 1.9048417657613754
+I1205 18:45:59.380099 137274321021824 utils.py:1231] [108500] l2_grads = 2.7359402179718018
+I1205 18:45:59.380199 137274321021824 utils.py:1231] [108500] lr = 3.942419092043431e-06
+I1205 18:45:59.380268 137274321021824 utils.py:1231] [108500] uptime = 680148.74262912
+I1205 18:45:59.380346 137274321021824 utils.py:1231] [108500] examples_seen = 111104000.0
+I1205 18:45:59.380411 137274321021824 utils.py:1231] [108500] progress = 0.9635622496736321
+I1205 18:45:59.380473 137274321021824 utils.py:1231] [108500] epoch = 86.7209348976363
+I1205 18:45:59.380537 137274321021824 utils.py:1231] [108500] img/sec/core = 164.21450710881416
+I1205 18:45:59.380614 137274321021824 utils.py:1231] [108500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.89588868350998
+I1205 18:45:59.380671 137274321021824 utils.py:1231] [108500] core_hours = 188.89588868350998
+I1205 18:45:59.380741 137274321021824 train.py:125] NOTE: Steps:108500/112603 [96.4%]
+Walltime:7d20h55m (0s eval)
+ETA:7h8m
+Total train time:8d4h2m
+I1205 18:51:11.158635 137274321021824 utils.py:1231] [108550] l2_params = 238.06282391367134
+I1205 18:51:11.158838 137274321021824 utils.py:1231] [108550] train/loss = 1.4385537207126617
+I1205 18:51:11.158940 137274321021824 utils.py:1231] [108550] l2_grads = 2.875760793685913
+I1205 18:51:11.159003 137274321021824 utils.py:1231] [108550] lr = 3.847064054209455e-06
+I1205 18:51:11.159055 137274321021824 utils.py:1231] [108550] uptime = 680460.521417376
+I1205 18:51:11.159108 137274321021824 utils.py:1231] [108550] examples_seen = 111155200.0
+I1205 18:51:11.159159 137274321021824 utils.py:1231] [108550] progress = 0.9640062875767076
+I1205 18:51:11.159213 137274321021824 utils.py:1231] [108550] epoch = 86.76089846210525
+I1205 18:51:11.159265 137274321021824 utils.py:1231] [108550] img/sec/core = 164.2189973423113
+I1205 18:51:11.159322 137274321021824 utils.py:1231] [108550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 188.98249390247
+I1205 18:51:11.159375 137274321021824 utils.py:1231] [108550] core_hours = 188.98249390247
+I1205 18:51:11.159436 137274321021824 train.py:125] NOTE: Steps:108550/112603 [96.4%]
+Walltime:7d21h1m (0s eval)
+ETA:7h3m
+Total train time:8d4h2m
+I1205 18:56:22.948595 137274321021824 utils.py:1231] [108600] l2_params = 238.06055972366946
+I1205 18:56:22.948909 137274321021824 utils.py:1231] [108600] train/loss = 2.639501839876175
+I1205 18:56:22.949068 137274321021824 utils.py:1231] [108600] l2_grads = 2.7173919677734375
+I1205 18:56:22.949156 137274321021824 utils.py:1231] [108600] lr = 3.752871896951259e-06
+I1205 18:56:22.949211 137274321021824 utils.py:1231] [108600] uptime = 680772.311572773
+I1205 18:56:22.949264 137274321021824 utils.py:1231] [108600] examples_seen = 111206400.0
+I1205 18:56:22.949316 137274321021824 utils.py:1231] [108600] progress = 0.9644503254797829
+I1205 18:56:22.949366 137274321021824 utils.py:1231] [108600] epoch = 86.80086202657421
+I1205 18:56:22.949419 137274321021824 utils.py:1231] [108600] img/sec/core = 164.21301030111928
+I1205 18:56:22.949477 137274321021824 utils.py:1231] [108600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.06910227896918
+I1205 18:56:22.949530 137274321021824 utils.py:1231] [108600] core_hours = 189.06910227896918
+I1205 18:56:22.949595 137274321021824 train.py:125] NOTE: Steps:108600/112603 [96.4%]
+Walltime:7d21h6m (0s eval)
+ETA:6h58m
+Total train time:8d4h2m
+I1205 19:01:34.737571 137274321021824 utils.py:1231] [108650] l2_params = 238.05825874569567
+I1205 19:01:34.737778 137274321021824 utils.py:1231] [108650] train/loss = 1.778140977025032
+I1205 19:01:34.737879 137274321021824 utils.py:1231] [108650] l2_grads = 2.602328062057495
+I1205 19:01:34.737949 137274321021824 utils.py:1231] [108650] lr = 3.6598428410359082e-06
+I1205 19:01:34.738002 137274321021824 utils.py:1231] [108650] uptime = 681084.100363664
+I1205 19:01:34.738056 137274321021824 utils.py:1231] [108650] examples_seen = 111257600.0
+I1205 19:01:34.738106 137274321021824 utils.py:1231] [108650] progress = 0.9648943633828584
+I1205 19:01:34.738157 137274321021824 utils.py:1231] [108650] epoch = 86.84082559104317
+I1205 19:01:34.738208 137274321021824 utils.py:1231] [108650] img/sec/core = 164.21372895957637
+I1205 19:01:34.738264 137274321021824 utils.py:1231] [108650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.1557102764389
+I1205 19:01:34.738317 137274321021824 utils.py:1231] [108650] core_hours = 189.1557102764389
+I1205 19:01:34.738380 137274321021824 train.py:125] NOTE: Steps:108650/112603 [96.5%]
+Walltime:7d21h11m (0s eval)
+ETA:6h52m
+Total train time:8d4h2m
+I1205 19:06:46.524790 137274321021824 utils.py:1231] [108700] l2_params = 238.05580806027146
+I1205 19:06:46.525063 137274321021824 utils.py:1231] [108700] train/loss = 3.624351292848587
+I1205 19:06:46.525244 137274321021824 utils.py:1231] [108700] l2_grads = 2.9472925662994385
+I1205 19:06:46.525337 137274321021824 utils.py:1231] [108700] lr = 3.567977104504382e-06
+I1205 19:06:46.525401 137274321021824 utils.py:1231] [108700] uptime = 681395.8877621259
+I1205 19:06:46.525471 137274321021824 utils.py:1231] [108700] examples_seen = 111308800.0
+I1205 19:06:46.754875 137274321021824 utils.py:1231] [108700] progress = 0.9653384012859337
+I1205 19:06:46.755163 137274321021824 utils.py:1231] [108700] epoch = 86.88078915551212
+I1205 19:06:46.755250 137274321021824 utils.py:1231] [108700] img/sec/core = 164.21446233096623
+I1205 19:06:46.755365 137274321021824 utils.py:1231] [108700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.24231788712274
+I1205 19:06:46.755443 137274321021824 utils.py:1231] [108700] core_hours = 189.24231788712274
+I1205 19:06:46.755527 137274321021824 train.py:125] NOTE: Steps:108700/112603 [96.5%]
+Walltime:7d21h16m (0s eval)
+ETA:6h47m
+Total train time:8d4h2m
+I1205 19:11:58.550757 137274321021824 utils.py:1231] [108750] l2_params = 238.0533370366259
+I1205 19:11:58.551084 137274321021824 utils.py:1231] [108750] train/loss = 3.1052693724632263
+I1205 19:11:58.551256 137274321021824 utils.py:1231] [108750] l2_grads = 2.7363038063049316
+I1205 19:11:58.551348 137274321021824 utils.py:1231] [108750] lr = 3.4772749026711092e-06
+I1205 19:11:58.551417 137274321021824 utils.py:1231] [108750] uptime = 681707.913778086
+I1205 19:11:58.551490 137274321021824 utils.py:1231] [108750] examples_seen = 111360000.0
+I1205 19:11:58.551553 137274321021824 utils.py:1231] [108750] progress = 0.9657824391890092
+I1205 19:11:58.551621 137274321021824 utils.py:1231] [108750] epoch = 86.92075271998108
+I1205 19:11:58.551694 137274321021824 utils.py:1231] [108750] img/sec/core = 164.08888163530168
+I1205 19:11:58.551771 137274321021824 utils.py:1231] [108750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.328991780445
+I1205 19:11:58.551837 137274321021824 utils.py:1231] [108750] core_hours = 189.328991780445
+I1205 19:11:58.551934 137274321021824 train.py:125] NOTE: Steps:108750/112603 [96.6%]
+Walltime:7d21h21m (0s eval)
+ETA:6h42m
+Total train time:8d4h2m
+I1205 19:17:10.349307 137274321021824 utils.py:1231] [108800] l2_params = 238.05102455946866
+I1205 19:17:10.349566 137274321021824 utils.py:1231] [108800] train/loss = 1.367810234427452
+I1205 19:17:10.349699 137274321021824 utils.py:1231] [108800] l2_grads = 2.866917848587036
+I1205 19:17:10.349825 137274321021824 utils.py:1231] [108800] lr = 3.387736448123376e-06
+I1205 19:17:10.349915 137274321021824 utils.py:1231] [108800] uptime = 682019.7122738659
+I1205 19:17:10.349986 137274321021824 utils.py:1231] [108800] examples_seen = 111411200.0
+I1205 19:17:10.350051 137274321021824 utils.py:1231] [108800] progress = 0.9662264770920845
+I1205 19:17:10.350116 137274321021824 utils.py:1231] [108800] epoch = 86.96071628445004
+I1205 19:17:10.350184 137274321021824 utils.py:1231] [108800] img/sec/core = 164.20861772254824
+I1205 19:17:10.350254 137274321021824 utils.py:1231] [108800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.4156024737172
+I1205 19:17:10.350320 137274321021824 utils.py:1231] [108800] core_hours = 189.4156024737172
+I1205 19:17:10.350392 137274321021824 train.py:125] NOTE: Steps:108800/112603 [96.6%]
+Walltime:7d21h26m (0s eval)
+ETA:6h37m
+Total train time:8d4h2m
+I1205 19:22:22.137997 137274321021824 utils.py:1231] [108850] l2_params = 238.04887170007433
+I1205 19:22:22.138193 137274321021824 utils.py:1231] [108850] train/loss = 1.4387988448143005
+I1205 19:22:22.138286 137274321021824 utils.py:1231] [108850] l2_grads = 2.7298452854156494
+I1205 19:22:22.138347 137274321021824 utils.py:1231] [108850] lr = 3.2993619507210312e-06
+I1205 19:22:22.138409 137274321021824 utils.py:1231] [108850] uptime = 682331.500771173
+I1205 19:22:22.138461 137274321021824 utils.py:1231] [108850] examples_seen = 111462400.0
+I1205 19:22:22.138509 137274321021824 utils.py:1231] [108850] progress = 0.96667051499516
+I1205 19:22:22.138556 137274321021824 utils.py:1231] [108850] epoch = 87.00067984891899
+I1205 19:22:22.138606 137274321021824 utils.py:1231] [108850] img/sec/core = 164.21388358522123
+I1205 19:22:22.138662 137274321021824 utils.py:1231] [108850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.50221038963585
+I1205 19:22:22.138713 137274321021824 utils.py:1231] [108850] core_hours = 189.50221038963585
+I1205 19:22:22.138771 137274321021824 train.py:125] NOTE: Steps:108850/112603 [96.7%]
+Walltime:7d21h32m (0s eval)
+ETA:6h32m
+Total train time:8d4h2m
+I1205 19:27:33.915101 137274321021824 utils.py:1231] [108900] l2_params = 238.04672219785138
+I1205 19:27:33.915352 137274321021824 utils.py:1231] [108900] train/loss = 2.4409776628017426
+I1205 19:27:33.915478 137274321021824 utils.py:1231] [108900] l2_grads = 2.7930803298950195
+I1205 19:27:33.915561 137274321021824 utils.py:1231] [108900] lr = 3.2121516175956662e-06
+I1205 19:27:33.915622 137274321021824 utils.py:1231] [108900] uptime = 682643.277984329
+I1205 19:27:33.915683 137274321021824 utils.py:1231] [108900] examples_seen = 111513600.0
+I1205 19:27:33.915733 137274321021824 utils.py:1231] [108900] progress = 0.9671145528982354
+I1205 19:27:33.915788 137274321021824 utils.py:1231] [108900] epoch = 87.04064341338795
+I1205 19:27:33.915852 137274321021824 utils.py:1231] [108900] img/sec/core = 164.21982697748751
+I1205 19:27:33.915930 137274321021824 utils.py:1231] [108900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.58881517106803
+I1205 19:27:33.915994 137274321021824 utils.py:1231] [108900] core_hours = 189.58881517106803
+I1205 19:27:33.916064 137274321021824 train.py:125] NOTE: Steps:108900/112603 [96.7%]
+Walltime:7d21h37m (0s eval)
+ETA:6h26m
+Total train time:8d4h2m
+I1205 19:32:45.702722 137274321021824 utils.py:1231] [108950] l2_params = 238.0445955900375
+I1205 19:32:45.702992 137274321021824 utils.py:1231] [108950] train/loss = 1.5977507084608078
+I1205 19:32:45.703124 137274321021824 utils.py:1231] [108950] l2_grads = 2.8634564876556396
+I1205 19:32:45.703229 137274321021824 utils.py:1231] [108950] lr = 3.12610565315044e-06
+I1205 19:32:45.703306 137274321021824 utils.py:1231] [108950] uptime = 682955.065667097
+I1205 19:32:45.703382 137274321021824 utils.py:1231] [108950] examples_seen = 111564800.0
+I1205 19:32:45.703443 137274321021824 utils.py:1231] [108950] progress = 0.9675585908013108
+I1205 19:32:45.703513 137274321021824 utils.py:1231] [108950] epoch = 87.0806069778569
+I1205 19:32:45.703582 137274321021824 utils.py:1231] [108950] img/sec/core = 164.2143125907405
+I1205 19:32:45.703645 137274321021824 utils.py:1231] [108950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.67542286072583
+I1205 19:32:45.703709 137274321021824 utils.py:1231] [108950] core_hours = 189.67542286072583
+I1205 19:32:45.703772 137274321021824 train.py:125] NOTE: Steps:108950/112603 [96.8%]
+Walltime:7d21h42m (0s eval)
+ETA:6h21m
+Total train time:8d4h2m
+I1205 19:37:57.495180 137274321021824 utils.py:1231] [109000] l2_params = 238.0424510494651
+I1205 19:37:57.495385 137274321021824 utils.py:1231] [109000] train/loss = 1.4260676205158234
+I1205 19:37:57.495485 137274321021824 utils.py:1231] [109000] l2_grads = 2.94299054145813
+I1205 19:37:57.495564 137274321021824 utils.py:1231] [109000] lr = 3.041224259059364e-06
+I1205 19:37:57.495626 137274321021824 utils.py:1231] [109000] uptime = 683266.857987372
+I1205 19:37:57.495687 137274321021824 utils.py:1231] [109000] examples_seen = 111616000.0
+I1205 19:37:57.495746 137274321021824 utils.py:1231] [109000] progress = 0.9680026287043862
+I1205 19:37:57.495806 137274321021824 utils.py:1231] [109000] epoch = 87.12057054232586
+I1205 19:37:57.495865 137274321021824 utils.py:1231] [109000] img/sec/core = 164.21187011548233
+I1205 19:37:57.495935 137274321021824 utils.py:1231] [109000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.76203183858
+I1205 19:37:57.495995 137274321021824 utils.py:1231] [109000] core_hours = 189.76203183858
+I1205 19:37:57.496063 137274321021824 train.py:125] NOTE: Steps:109000/112603 [96.8%]
+Walltime:7d21h47m (0s eval)
+ETA:6h16m
+Total train time:8d4h2m
+I1205 19:43:09.606321 137274321021824 utils.py:1231] [109050] l2_params = 238.04041259890062
+I1205 19:43:09.606585 137274321021824 utils.py:1231] [109050] train/loss = 3.271276205778122
+I1205 19:43:09.606724 137274321021824 utils.py:1231] [109050] l2_grads = 2.869370937347412
+I1205 19:43:09.606823 137274321021824 utils.py:1231] [109050] lr = 2.957507634266968e-06
+I1205 19:43:09.606904 137274321021824 utils.py:1231] [109050] uptime = 683578.969264783
+I1205 19:43:09.606968 137274321021824 utils.py:1231] [109050] examples_seen = 111667200.0
+I1205 19:43:09.607029 137274321021824 utils.py:1231] [109050] progress = 0.9684466666074616
+I1205 19:43:09.607087 137274321021824 utils.py:1231] [109050] epoch = 87.16053410679483
+I1205 19:43:09.607147 137274321021824 utils.py:1231] [109050] img/sec/core = 164.04405641702905
+I1205 19:43:09.607211 137274321021824 utils.py:1231] [109050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.8487294156386
+I1205 19:43:09.607270 137274321021824 utils.py:1231] [109050] core_hours = 189.8487294156386
+I1205 19:43:09.607340 137274321021824 train.py:125] NOTE: Steps:109050/112603 [96.8%]
+Walltime:7d21h52m (0s eval)
+ETA:6h11m
+Total train time:8d4h2m
+I1205 19:48:21.403103 137274321021824 utils.py:1231] [109100] l2_params = 238.03828647232498
+I1205 19:48:21.403361 137274321021824 utils.py:1231] [109100] train/loss = 3.7639551162719727
+I1205 19:48:21.403583 137274321021824 utils.py:1231] [109100] l2_grads = 2.9252753257751465
+I1205 19:48:21.403672 137274321021824 utils.py:1231] [109100] lr = 2.87495597498796e-06
+I1205 19:48:21.403735 137274321021824 utils.py:1231] [109100] uptime = 683890.766096119
+I1205 19:48:21.403805 137274321021824 utils.py:1231] [109100] examples_seen = 111718400.0
+I1205 19:48:21.403861 137274321021824 utils.py:1231] [109100] progress = 0.968890704510537
+I1205 19:48:21.403928 137274321021824 utils.py:1231] [109100] epoch = 87.20049767126378
+I1205 19:48:21.403989 137274321021824 utils.py:1231] [109100] img/sec/core = 164.20949430633388
+I1205 19:48:21.404051 137274321021824 utils.py:1231] [109100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 189.93533964656527
+I1205 19:48:21.404106 137274321021824 utils.py:1231] [109100] core_hours = 189.93533964656527
+I1205 19:48:21.404178 137274321021824 train.py:125] NOTE: Steps:109100/112603 [96.9%]
+Walltime:7d21h58m (0s eval)
+ETA:6h5m
+Total train time:8d4h2m
+I1205 19:53:33.258228 137274321021824 utils.py:1231] [109150] l2_params = 238.03628198611713
+I1205 19:53:33.258476 137274321021824 utils.py:1231] [109150] train/loss = 1.5500036627054214
+I1205 19:53:33.258602 137274321021824 utils.py:1231] [109150] l2_grads = 2.869983434677124
+I1205 19:53:33.258681 137274321021824 utils.py:1231] [109150] lr = 2.7935694747063474e-06
+I1205 19:53:33.258738 137274321021824 utils.py:1231] [109150] uptime = 684202.621100202
+I1205 19:53:33.258795 137274321021824 utils.py:1231] [109150] examples_seen = 111769600.0
+I1205 19:53:33.258847 137274321021824 utils.py:1231] [109150] progress = 0.9693347424136124
+I1205 19:53:33.258911 137274321021824 utils.py:1231] [109150] epoch = 87.24046123573274
+I1205 19:53:33.258975 137274321021824 utils.py:1231] [109150] img/sec/core = 164.17886302818644
+I1205 19:53:33.259035 137274321021824 utils.py:1231] [109150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.0219660365883
+I1205 19:53:33.259099 137274321021824 utils.py:1231] [109150] core_hours = 190.0219660365883
+I1205 19:53:33.259171 137274321021824 train.py:125] NOTE: Steps:109150/112603 [96.9%]
+Walltime:7d22h3m (0s eval)
+ETA:6h0m
+Total train time:8d4h2m
+I1205 19:58:45.068199 137274321021824 utils.py:1231] [109200] l2_params = 238.03407165755013
+I1205 19:58:45.068472 137274321021824 utils.py:1231] [109200] train/loss = 3.3858854174613953
+I1205 19:58:45.068577 137274321021824 utils.py:1231] [109200] l2_grads = 2.97988224029541
+I1205 19:58:45.068670 137274321021824 utils.py:1231] [109200] lr = 2.7133483241754344e-06
+I1205 19:58:45.068772 137274321021824 utils.py:1231] [109200] uptime = 684514.431117377
+I1205 19:58:45.068876 137274321021824 utils.py:1231] [109200] examples_seen = 111820800.0
+I1205 19:58:45.068953 137274321021824 utils.py:1231] [109200] progress = 0.9697787803166878
+I1205 19:58:45.069009 137274321021824 utils.py:1231] [109200] epoch = 87.28042480020169
+I1205 19:58:45.069067 137274321021824 utils.py:1231] [109200] img/sec/core = 164.20255020626166
+I1205 19:58:45.069130 137274321021824 utils.py:1231] [109200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.10857993024806
+I1205 19:58:45.069185 137274321021824 utils.py:1231] [109200] core_hours = 190.10857993024806
+I1205 19:58:45.069251 137274321021824 train.py:125] NOTE: Steps:109200/112603 [97.0%]
+Walltime:7d22h8m (0s eval)
+ETA:5h55m
+Total train time:8d4h2m
+I1205 20:03:56.881946 137274321021824 utils.py:1231] [109250] l2_params = 238.032361520784
+I1205 20:03:56.882193 137274321021824 utils.py:1231] [109250] train/loss = 2.942538857460022
+I1205 20:03:56.882326 137274321021824 utils.py:1231] [109250] l2_grads = 2.7158749103546143
+I1205 20:03:56.882411 137274321021824 utils.py:1231] [109250] lr = 2.6342927114170945e-06
+I1205 20:03:56.882484 137274321021824 utils.py:1231] [109250] uptime = 684826.2448459769
+I1205 20:03:56.882560 137274321021824 utils.py:1231] [109250] examples_seen = 111872000.0
+I1205 20:03:56.882618 137274321021824 utils.py:1231] [109250] progress = 0.9702228182197632
+I1205 20:03:56.882677 137274321021824 utils.py:1231] [109250] epoch = 87.32038836467065
+I1205 20:03:56.882744 137274321021824 utils.py:1231] [109250] img/sec/core = 164.200595752769
+I1205 20:03:56.882813 137274321021824 utils.py:1231] [109250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.19519485485912
+I1205 20:03:56.882888 137274321021824 utils.py:1231] [109250] core_hours = 190.19519485485912
+I1205 20:03:56.882956 137274321021824 train.py:125] NOTE: Steps:109250/112603 [97.0%]
+Walltime:7d22h13m (0s eval)
+ETA:5h50m
+Total train time:8d4h2m
+I1205 20:09:08.692190 137274321021824 utils.py:1231] [109300] l2_params = 238.03094566789514
+I1205 20:09:08.692480 137274321021824 utils.py:1231] [109300] train/loss = 2.914584517478943
+I1205 20:09:08.692656 137274321021824 utils.py:1231] [109300] l2_grads = 2.678227424621582
+I1205 20:09:08.692750 137274321021824 utils.py:1231] [109300] lr = 2.5564028217213927e-06
+I1205 20:09:08.692826 137274321021824 utils.py:1231] [109300] uptime = 685138.055183868
+I1205 20:09:08.692898 137274321021824 utils.py:1231] [109300] examples_seen = 111923200.0
+I1205 20:09:08.692962 137274321021824 utils.py:1231] [109300] progress = 0.9706668561228387
+I1205 20:09:08.693018 137274321021824 utils.py:1231] [109300] epoch = 87.36035192913961
+I1205 20:09:08.693074 137274321021824 utils.py:1231] [109300] img/sec/core = 164.2023813138647
+I1205 20:09:08.693136 137274321021824 utils.py:1231] [109300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.28180883760666
+I1205 20:09:08.693194 137274321021824 utils.py:1231] [109300] core_hours = 190.28180883760666
+I1205 20:09:08.693259 137274321021824 train.py:125] NOTE: Steps:109300/112603 [97.1%]
+Walltime:7d22h18m (0s eval)
+ETA:5h45m
+Total train time:8d4h2m
+I1205 20:14:20.495807 137274321021824 utils.py:1231] [109350] l2_params = 238.02946766104282
+I1205 20:14:20.496021 137274321021824 utils.py:1231] [109350] train/loss = 2.3322669863700867
+I1205 20:14:20.496123 137274321021824 utils.py:1231] [109350] l2_grads = 2.5377256870269775
+I1205 20:14:20.496195 137274321021824 utils.py:1231] [109350] lr = 2.4796788376463488e-06
+I1205 20:14:20.496254 137274321021824 utils.py:1231] [109350] uptime = 685449.858615183
+I1205 20:14:20.496315 137274321021824 utils.py:1231] [109350] examples_seen = 111974400.0
+I1205 20:14:20.496374 137274321021824 utils.py:1231] [109350] progress = 0.9711108940259141
+I1205 20:14:20.496430 137274321021824 utils.py:1231] [109350] epoch = 87.40031549360856
+I1205 20:14:20.496488 137274321021824 utils.py:1231] [109350] img/sec/core = 164.20601846512517
+I1205 20:14:20.496551 137274321021824 utils.py:1231] [109350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.36842090186082
+I1205 20:14:20.496607 137274321021824 utils.py:1231] [109350] core_hours = 190.36842090186082
+I1205 20:14:20.496671 137274321021824 train.py:125] NOTE: Steps:109350/112603 [97.1%]
+Walltime:7d22h24m (0s eval)
+ETA:5h39m
+Total train time:8d4h2m
+I1205 20:19:32.294524 137274321021824 utils.py:1231] [109400] l2_params = 238.02799157717834
+I1205 20:19:32.294770 137274321021824 utils.py:1231] [109400] train/loss = 1.3760341852903366
+I1205 20:19:32.294895 137274321021824 utils.py:1231] [109400] l2_grads = 2.795724391937256
+I1205 20:19:32.294970 137274321021824 utils.py:1231] [109400] lr = 2.4041209390172293e-06
+I1205 20:19:32.295025 137274321021824 utils.py:1231] [109400] uptime = 685761.657386986
+I1205 20:19:32.295080 137274321021824 utils.py:1231] [109400] examples_seen = 112025600.0
+I1205 20:19:32.295132 137274321021824 utils.py:1231] [109400] progress = 0.9715549319289895
+I1205 20:19:32.295186 137274321021824 utils.py:1231] [109400] epoch = 87.44027905807752
+I1205 20:19:32.295240 137274321021824 utils.py:1231] [109400] img/sec/core = 164.2084723552488
+I1205 20:19:32.295300 137274321021824 utils.py:1231] [109400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.4550316718061
+I1205 20:19:32.295352 137274321021824 utils.py:1231] [109400] core_hours = 190.4550316718061
+I1205 20:19:32.295417 137274321021824 train.py:125] NOTE: Steps:109400/112603 [97.2%]
+Walltime:7d22h29m (0s eval)
+ETA:5h34m
+Total train time:8d4h2m
+I1205 20:24:44.093239 137274321021824 utils.py:1231] [109450] l2_params = 238.02646386727403
+I1205 20:24:44.093445 137274321021824 utils.py:1231] [109450] train/loss = 1.50863479077816
+I1205 20:24:44.093545 137274321021824 utils.py:1231] [109450] l2_grads = 2.778456211090088
+I1205 20:24:44.093619 137274321021824 utils.py:1231] [109450] lr = 2.329729302926042e-06
+I1205 20:24:44.093677 137274321021824 utils.py:1231] [109450] uptime = 686073.456038433
+I1205 20:24:44.093783 137274321021824 utils.py:1231] [109450] examples_seen = 112076800.0
+I1205 20:24:44.093844 137274321021824 utils.py:1231] [109450] progress = 0.9719989698320649
+I1205 20:24:44.093908 137274321021824 utils.py:1231] [109450] epoch = 87.48024262254647
+I1205 20:24:44.093966 137274321021824 utils.py:1231] [109450] img/sec/core = 164.20853574055366
+I1205 20:24:44.094026 137274321021824 utils.py:1231] [109450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.54164240831918
+I1205 20:24:44.094078 137274321021824 utils.py:1231] [109450] core_hours = 190.54164240831918
+I1205 20:24:44.094141 137274321021824 train.py:125] NOTE: Steps:109450/112603 [97.2%]
+Walltime:7d22h34m (0s eval)
+ETA:5h29m
+Total train time:8d4h2m
+I1205 20:29:55.895176 137274321021824 utils.py:1231] [109500] l2_params = 238.02473620248466
+I1205 20:29:55.895386 137274321021824 utils.py:1231] [109500] train/loss = 1.4049250781536102
+I1205 20:29:55.895491 137274321021824 utils.py:1231] [109500] l2_grads = 2.7905337810516357
+I1205 20:29:55.895556 137274321021824 utils.py:1231] [109500] lr = 2.2565041037317023e-06
+I1205 20:29:55.895609 137274321021824 utils.py:1231] [109500] uptime = 686385.257970533
+I1205 20:29:55.895662 137274321021824 utils.py:1231] [109500] examples_seen = 112128000.0
+I1205 20:29:55.895717 137274321021824 utils.py:1231] [109500] progress = 0.9724430077351403
+I1205 20:29:55.895766 137274321021824 utils.py:1231] [109500] epoch = 87.52020618701543
+I1205 20:29:55.895817 137274321021824 utils.py:1231] [109500] img/sec/core = 164.2068080052295
+I1205 20:29:55.895873 137274321021824 utils.py:1231] [109500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.62825405612472
+I1205 20:29:55.895930 137274321021824 utils.py:1231] [109500] core_hours = 190.62825405612472
+I1205 20:29:55.895998 137274321021824 train.py:125] NOTE: Steps:109500/112603 [97.2%]
+Walltime:7d22h39m (0s eval)
+ETA:5h24m
+Total train time:8d4h2m
+I1205 20:35:07.699937 137274321021824 utils.py:1231] [109550] l2_params = 238.02294031422375
+I1205 20:35:07.700160 137274321021824 utils.py:1231] [109550] train/loss = 3.29221647977829
+I1205 20:35:07.700261 137274321021824 utils.py:1231] [109550] l2_grads = 2.8225314617156982
+I1205 20:35:07.700334 137274321021824 utils.py:1231] [109550] lr = 2.1844455130589277e-06
+I1205 20:35:07.700394 137274321021824 utils.py:1231] [109550] uptime = 686697.062755579
+I1205 20:35:07.700453 137274321021824 utils.py:1231] [109550] examples_seen = 112179200.0
+I1205 20:35:07.700510 137274321021824 utils.py:1231] [109550] progress = 0.9728870456382157
+I1205 20:35:07.700566 137274321021824 utils.py:1231] [109550] epoch = 87.5601697514844
+I1205 20:35:07.700628 137274321021824 utils.py:1231] [109550] img/sec/core = 164.2053055486046
+I1205 20:35:07.700690 137274321021824 utils.py:1231] [109550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.71486649641528
+I1205 20:35:07.700746 137274321021824 utils.py:1231] [109550] core_hours = 190.71486649641528
+I1205 20:35:07.700812 137274321021824 train.py:125] NOTE: Steps:109550/112603 [97.3%]
+Walltime:7d22h44m (0s eval)
+ETA:5h18m
+Total train time:8d4h2m
+I1205 20:40:19.501771 137274321021824 utils.py:1231] [109600] l2_params = 238.02136371738487
+I1205 20:40:19.501986 137274321021824 utils.py:1231] [109600] train/loss = 1.997459501028061
+I1205 20:40:19.502105 137274321021824 utils.py:1231] [109600] l2_grads = 2.703460931777954
+I1205 20:40:19.502192 137274321021824 utils.py:1231] [109600] lr = 2.1135536997983407e-06
+I1205 20:40:19.502275 137274321021824 utils.py:1231] [109600] uptime = 687008.864631579
+I1205 20:40:19.502333 137274321021824 utils.py:1231] [109600] examples_seen = 112230400.0
+I1205 20:40:19.502387 137274321021824 utils.py:1231] [109600] progress = 0.973331083541291
+I1205 20:40:19.502440 137274321021824 utils.py:1231] [109600] epoch = 87.60013331595334
+I1205 20:40:19.502496 137274321021824 utils.py:1231] [109600] img/sec/core = 164.20683754961752
+I1205 20:40:19.502556 137274321021824 utils.py:1231] [109600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.8014781286375
+I1205 20:40:19.502612 137274321021824 utils.py:1231] [109600] core_hours = 190.8014781286375
+I1205 20:40:19.502675 137274321021824 train.py:125] NOTE: Steps:109600/112603 [97.3%]
+Walltime:7d22h50m (0s eval)
+ETA:5h13m
+Total train time:8d4h1m
+I1205 20:45:31.311254 137274321021824 utils.py:1231] [109650] l2_params = 238.02001841162834
+I1205 20:45:31.311463 137274321021824 utils.py:1231] [109650] train/loss = 3.265958070755005
+I1205 20:45:31.311578 137274321021824 utils.py:1231] [109650] l2_grads = 2.7545254230499268
+I1205 20:45:31.311650 137274321021824 utils.py:1231] [109650] lr = 2.0438288301056953e-06
+I1205 20:45:31.311712 137274321021824 utils.py:1231] [109650] uptime = 687320.674073001
+I1205 20:45:31.311765 137274321021824 utils.py:1231] [109650] examples_seen = 112281600.0
+I1205 20:45:31.311819 137274321021824 utils.py:1231] [109650] progress = 0.9737751214443665
+I1205 20:45:31.311869 137274321021824 utils.py:1231] [109650] epoch = 87.6400968804223
+I1205 20:45:31.311935 137274321021824 utils.py:1231] [109650] img/sec/core = 164.20285340464426
+I1205 20:45:31.311994 137274321021824 utils.py:1231] [109650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.88809186236583
+I1205 20:45:31.312046 137274321021824 utils.py:1231] [109650] core_hours = 190.88809186236583
+I1205 20:45:31.312109 137274321021824 train.py:125] NOTE: Steps:109650/112603 [97.4%]
+Walltime:7d22h55m (0s eval)
+ETA:5h8m
+Total train time:8d4h1m
+I1205 20:50:43.125334 137274321021824 utils.py:1231] [109700] l2_params = 238.01887246195784
+I1205 20:50:43.125541 137274321021824 utils.py:1231] [109700] train/loss = 1.8518698066473007
+I1205 20:50:43.125653 137274321021824 utils.py:1231] [109700] l2_grads = 2.828583002090454
+I1205 20:50:43.125751 137274321021824 utils.py:1231] [109700] lr = 1.9752710674018254e-06
+I1205 20:50:43.125838 137274321021824 utils.py:1231] [109700] uptime = 687632.488197418
+I1205 20:50:43.125913 137274321021824 utils.py:1231] [109700] examples_seen = 112332800.0
+I1205 20:50:43.126003 137274321021824 utils.py:1231] [109700] progress = 0.9742191593474419
+I1205 20:50:43.126074 137274321021824 utils.py:1231] [109700] epoch = 87.68006044489125
+I1205 20:50:43.126147 137274321021824 utils.py:1231] [109700] img/sec/core = 164.20038731640082
+I1205 20:50:43.126255 137274321021824 utils.py:1231] [109700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 190.9747068969261
+I1205 20:50:43.126319 137274321021824 utils.py:1231] [109700] core_hours = 190.9747068969261
+I1205 20:50:43.126425 137274321021824 train.py:125] NOTE: Steps:109700/112603 [97.4%]
+Walltime:7d23h0m (0s eval)
+ETA:5h3m
+Total train time:8d4h1m
+I1205 20:55:54.930050 137274321021824 utils.py:1231] [109750] l2_params = 238.01761675822488
+I1205 20:55:54.930258 137274321021824 utils.py:1231] [109750] train/loss = 1.5174070298671722
+I1205 20:55:54.930359 137274321021824 utils.py:1231] [109750] l2_grads = 2.8144028186798096
+I1205 20:55:54.930424 137274321021824 utils.py:1231] [109750] lr = 1.907880572372029e-06
+I1205 20:55:54.930479 137274321021824 utils.py:1231] [109750] uptime = 687944.292840588
+I1205 20:55:54.930538 137274321021824 utils.py:1231] [109750] examples_seen = 112384000.0
+I1205 20:55:54.930588 137274321021824 utils.py:1231] [109750] progress = 0.9746631972505173
+I1205 20:55:54.930639 137274321021824 utils.py:1231] [109750] epoch = 87.72002400936022
+I1205 20:55:54.930693 137274321021824 utils.py:1231] [109750] img/sec/core = 164.2053802646194
+I1205 20:55:54.930752 137274321021824 utils.py:1231] [109750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.06131929780668
+I1205 20:55:54.930807 137274321021824 utils.py:1231] [109750] core_hours = 191.06131929780668
+I1205 20:55:54.930869 137274321021824 train.py:125] NOTE: Steps:109750/112603 [97.5%]
+Walltime:7d23h5m (0s eval)
+ETA:4h58m
+Total train time:8d4h1m
+I1205 21:01:06.726023 137274321021824 utils.py:1231] [109800] l2_params = 238.01629248569122
+I1205 21:01:06.726344 137274321021824 utils.py:1231] [109800] train/loss = 1.6866580992937088
+I1205 21:01:06.726576 137274321021824 utils.py:1231] [109800] l2_grads = 2.816906452178955
+I1205 21:01:06.726689 137274321021824 utils.py:1231] [109800] lr = 1.8416575029657945e-06
+I1205 21:01:06.726768 137274321021824 utils.py:1231] [109800] uptime = 688256.089125677
+I1205 21:01:06.726858 137274321021824 utils.py:1231] [109800] examples_seen = 112435200.0
+I1205 21:01:06.726954 137274321021824 utils.py:1231] [109800] progress = 0.9751072351535927
+I1205 21:01:06.727039 137274321021824 utils.py:1231] [109800] epoch = 87.75998757382918
+I1205 21:01:06.727103 137274321021824 utils.py:1231] [109800] img/sec/core = 164.20978199078652
+I1205 21:01:06.727181 137274321021824 utils.py:1231] [109800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.14792937699806
+I1205 21:01:06.727249 137274321021824 utils.py:1231] [109800] core_hours = 191.14792937699806
+I1205 21:01:06.727323 137274321021824 train.py:125] NOTE: Steps:109800/112603 [97.5%]
+Walltime:7d23h10m (0s eval)
+ETA:4h52m
+Total train time:8d4h1m
+I1205 21:06:18.540538 137274321021824 utils.py:1231] [109850] l2_params = 238.01505656097174
+I1205 21:06:18.540768 137274321021824 utils.py:1231] [109850] train/loss = 2.2863596379756927
+I1205 21:06:18.540932 137274321021824 utils.py:1231] [109850] l2_grads = 2.6391079425811768
+I1205 21:06:18.541025 137274321021824 utils.py:1231] [109850] lr = 1.776602014396351e-06
+I1205 21:06:18.541090 137274321021824 utils.py:1231] [109850] uptime = 688567.90345039
+I1205 21:06:18.541156 137274321021824 utils.py:1231] [109850] examples_seen = 112486400.0
+I1205 21:06:18.541231 137274321021824 utils.py:1231] [109850] progress = 0.9755512730566681
+I1205 21:06:18.541332 137274321021824 utils.py:1231] [109850] epoch = 87.79995113829813
+I1205 21:06:18.541424 137274321021824 utils.py:1231] [109850] img/sec/core = 164.2002818412258
+I1205 21:06:18.541522 137274321021824 utils.py:1231] [109850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.2345444671961
+I1205 21:06:18.541618 137274321021824 utils.py:1231] [109850] core_hours = 191.2345444671961
+I1205 21:06:18.541728 137274321021824 train.py:125] NOTE: Steps:109850/112603 [97.6%]
+Walltime:7d23h16m (0s eval)
+ETA:4h47m
+Total train time:8d4h1m
+I1205 21:11:30.340243 137274321021824 utils.py:1231] [109900] l2_params = 238.01368890980467
+I1205 21:11:30.340515 137274321021824 utils.py:1231] [109900] train/loss = 1.5092905759811401
+I1205 21:11:30.340650 137274321021824 utils.py:1231] [109900] l2_grads = 2.8894400596618652
+I1205 21:11:30.340744 137274321021824 utils.py:1231] [109900] lr = 1.7127142591404525e-06
+I1205 21:11:30.340816 137274321021824 utils.py:1231] [109900] uptime = 688879.7031741589
+I1205 21:11:30.340902 137274321021824 utils.py:1231] [109900] examples_seen = 112537600.0
+I1205 21:11:30.340979 137274321021824 utils.py:1231] [109900] progress = 0.9759953109597436
+I1205 21:11:30.341037 137274321021824 utils.py:1231] [109900] epoch = 87.83991470276709
+I1205 21:11:30.341094 137274321021824 utils.py:1231] [109900] img/sec/core = 164.2079710049367
+I1205 21:11:30.341156 137274321021824 utils.py:1231] [109900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.32115550157636
+I1205 21:11:30.341213 137274321021824 utils.py:1231] [109900] core_hours = 191.32115550157636
+I1205 21:11:30.341279 137274321021824 train.py:125] NOTE: Steps:109900/112603 [97.6%]
+Walltime:7d23h21m (0s eval)
+ETA:4h42m
+Total train time:8d4h1m
+I1205 21:16:42.299853 137274321021824 utils.py:1231] [109950] l2_params = 238.01241102536008
+I1205 21:16:42.300123 137274321021824 utils.py:1231] [109950] train/loss = 2.320642441511154
+I1205 21:16:42.300236 137274321021824 utils.py:1231] [109950] l2_grads = 2.660846710205078
+I1205 21:16:42.300317 137274321021824 utils.py:1231] [109950] lr = 1.64999438693782e-06
+I1205 21:16:42.300372 137274321021824 utils.py:1231] [109950] uptime = 689191.662734081
+I1205 21:16:42.300426 137274321021824 utils.py:1231] [109950] examples_seen = 112588800.0
+I1205 21:16:42.300478 137274321021824 utils.py:1231] [109950] progress = 0.9764393488628189
+I1205 21:16:42.300527 137274321021824 utils.py:1231] [109950] epoch = 87.87987826723604
+I1205 21:16:42.300580 137274321021824 utils.py:1231] [109950] img/sec/core = 164.12383711781158
+I1205 21:16:42.300642 137274321021824 utils.py:1231] [109950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.40781093488806
+I1205 21:16:42.300696 137274321021824 utils.py:1231] [109950] core_hours = 191.40781093488806
+I1205 21:16:42.300786 137274321021824 train.py:125] NOTE: Steps:109950/112603 [97.6%]
+Walltime:7d23h26m (0s eval)
+ETA:4h37m
+Total train time:8d4h1m
+I1205 21:21:54.103708 137274321021824 utils.py:1231] [110000] l2_params = 238.01101336446894
+I1205 21:21:54.103972 137274321021824 utils.py:1231] [110000] train/loss = 1.383090153336525
+I1205 21:21:54.104089 137274321021824 utils.py:1231] [110000] l2_grads = 2.669095277786255
+I1205 21:21:54.104165 137274321021824 utils.py:1231] [110000] lr = 1.5884425447910856e-06
+I1205 21:21:54.104229 137274321021824 utils.py:1231] [110000] uptime = 689503.466590018
+I1205 21:21:54.104285 137274321021824 utils.py:1231] [110000] examples_seen = 112640000.0
+I1205 21:21:54.104340 137274321021824 utils.py:1231] [110000] progress = 0.9768833867658944
+I1205 21:21:54.104388 137274321021824 utils.py:1231] [110000] epoch = 87.919841831705
+I1205 21:21:54.104467 137274321021824 utils.py:1231] [110000] img/sec/core = 164.2057948454274
+I1205 21:21:54.104524 137274321021824 utils.py:1231] [110000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.4944231170928
+I1205 21:21:54.104577 137274321021824 utils.py:1231] [110000] core_hours = 191.4944231170928
+I1205 21:21:54.104658 137274321021824 train.py:125] NOTE: Steps:110000/112603 [97.7%]
+Walltime:7d23h31m (0s eval)
+ETA:4h31m
+Total train time:8d4h1m
+I1205 21:21:54.468654 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:110000/112603 [97.7%]
+Walltime:7d23h31m (0s eval)
+ETA:4h31m
+Total train time:8d4h1m
+I1205 21:23:32.997308 137274321021824 utils.py:1231] [110000] val/acc@1 = 0.7655851403061225
+I1205 21:23:32.997536 137274321021824 utils.py:1231] [110000] val/loss = 0.9187901652589137
+I1205 21:23:32.997680 137274321021824 utils.py:1231] [110000] z/secs/eval/val = 98.52878406096715
+I1205 21:23:32.997753 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 98.52878406096715
+I1205 21:28:44.789097 137274321021824 utils.py:1231] [110050] l2_params = 238.00979115126822
+I1205 21:28:44.789398 137274321021824 utils.py:1231] [110050] train/loss = 3.600607216358185
+I1205 21:28:44.789627 137274321021824 utils.py:1231] [110050] l2_grads = 3.312969923019409
+I1205 21:28:44.789764 137274321021824 utils.py:1231] [110050] lr = 1.5280588769649606e-06
+I1205 21:28:44.789877 137274321021824 utils.py:1231] [110050] uptime = 689914.152228427
+I1205 21:28:44.789990 137274321021824 utils.py:1231] [110050] examples_seen = 112691200.0
+I1205 21:28:44.790093 137274321021824 utils.py:1231] [110050] progress = 0.9773274246689697
+I1205 21:28:44.790175 137274321021824 utils.py:1231] [110050] epoch = 87.95980539617396
+I1205 21:28:44.790258 137274321021824 utils.py:1231] [110050] img/sec/core = 124.66956526249697
+I1205 21:28:44.790340 137274321021824 utils.py:1231] [110050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.60850246109527
+I1205 21:28:44.790421 137274321021824 utils.py:1231] [110050] core_hours = 191.60850246109527
+I1205 21:28:44.790521 137274321021824 train.py:125] NOTE: Steps:110050/112603 [97.7%]
+Walltime:7d23h38m (0s eval)
+ETA:4h26m
+Total train time:8d4h3m
+I1205 21:33:56.586978 137274321021824 utils.py:1231] [110100] l2_params = 238.00865786709232
+I1205 21:33:56.587200 137274321021824 utils.py:1231] [110100] train/loss = 3.11495378613472
+I1205 21:33:56.587299 137274321021824 utils.py:1231] [110100] l2_grads = 2.744483470916748
+I1205 21:33:56.587362 137274321021824 utils.py:1231] [110100] lr = 1.468843524986456e-06
+I1205 21:33:56.587415 137274321021824 utils.py:1231] [110100] uptime = 690225.94977715
+I1205 21:33:56.587467 137274321021824 utils.py:1231] [110100] examples_seen = 112742400.0
+I1205 21:33:56.587516 137274321021824 utils.py:1231] [110100] progress = 0.9777714625720452
+I1205 21:33:56.587564 137274321021824 utils.py:1231] [110100] epoch = 87.99976896064291
+I1205 21:33:56.587614 137274321021824 utils.py:1231] [110100] img/sec/core = 164.2091164914691
+I1205 21:33:56.587674 137274321021824 utils.py:1231] [110100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.69511289129608
+I1205 21:33:56.587725 137274321021824 utils.py:1231] [110100] core_hours = 191.69511289129608
+I1205 21:33:56.587785 137274321021824 train.py:125] NOTE: Steps:110100/112603 [97.8%]
+Walltime:7d23h43m (0s eval)
+ETA:4h21m
+Total train time:8d4h3m
+I1205 21:39:08.392107 137274321021824 utils.py:1231] [110150] l2_params = 238.00794957249505
+I1205 21:39:08.392321 137274321021824 utils.py:1231] [110150] train/loss = 1.387596346437931
+I1205 21:39:08.392422 137274321021824 utils.py:1231] [110150] l2_grads = 2.661700963973999
+I1205 21:39:08.392503 137274321021824 utils.py:1231] [110150] lr = 1.4107966276441656e-06
+I1205 21:39:08.392567 137274321021824 utils.py:1231] [110150] uptime = 690537.754928111
+I1205 21:39:08.392641 137274321021824 utils.py:1231] [110150] examples_seen = 112793600.0
+I1205 21:39:08.392698 137274321021824 utils.py:1231] [110150] progress = 0.9782155004751205
+I1205 21:39:08.392762 137274321021824 utils.py:1231] [110150] epoch = 88.03973252511187
+I1205 21:39:08.392818 137274321021824 utils.py:1231] [110150] img/sec/core = 164.20511284753536
+I1205 21:39:08.392878 137274321021824 utils.py:1231] [110150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.78172543322972
+I1205 21:39:08.392934 137274321021824 utils.py:1231] [110150] core_hours = 191.78172543322972
+I1205 21:39:08.393000 137274321021824 train.py:125] NOTE: Steps:110150/112603 [97.8%]
+Walltime:7d23h48m (0s eval)
+ETA:4h16m
+Total train time:8d4h3m
+I1205 21:44:20.182291 137274321021824 utils.py:1231] [110200] l2_params = 238.00721498392778
+I1205 21:44:20.182548 137274321021824 utils.py:1231] [110200] train/loss = 1.364205926656723
+I1205 21:44:20.182680 137274321021824 utils.py:1231] [110200] l2_grads = 2.7847912311553955
+I1205 21:44:20.182782 137274321021824 utils.py:1231] [110200] lr = 1.3539183209880375e-06
+I1205 21:44:20.182866 137274321021824 utils.py:1231] [110200] uptime = 690849.545223618
+I1205 21:44:20.182943 137274321021824 utils.py:1231] [110200] examples_seen = 112844800.0
+I1205 21:44:20.183005 137274321021824 utils.py:1231] [110200] progress = 0.978659538378196
+I1205 21:44:20.183063 137274321021824 utils.py:1231] [110200] epoch = 88.07969608958082
+I1205 21:44:20.183126 137274321021824 utils.py:1231] [110200] img/sec/core = 164.21293650828397
+I1205 21:44:20.183189 137274321021824 utils.py:1231] [110200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.86833384864835
+I1205 21:44:20.183248 137274321021824 utils.py:1231] [110200] core_hours = 191.86833384864835
+I1205 21:44:20.183318 137274321021824 train.py:125] NOTE: Steps:110200/112603 [97.9%]
+Walltime:7d23h54m (0s eval)
+ETA:4h11m
+Total train time:8d4h3m
+I1205 21:49:31.991192 137274321021824 utils.py:1231] [110250] l2_params = 238.00648765126363
+I1205 21:49:31.991390 137274321021824 utils.py:1231] [110250] train/loss = 1.3358627557754517
+I1205 21:49:31.991493 137274321021824 utils.py:1231] [110250] l2_grads = 2.8084232807159424
+I1205 21:49:31.991558 137274321021824 utils.py:1231] [110250] lr = 1.2982087383292674e-06
+I1205 21:49:31.991611 137274321021824 utils.py:1231] [110250] uptime = 691161.353972722
+I1205 21:49:31.991663 137274321021824 utils.py:1231] [110250] examples_seen = 112896000.0
+I1205 21:49:31.991715 137274321021824 utils.py:1231] [110250] progress = 0.9791035762812713
+I1205 21:49:31.991774 137274321021824 utils.py:1231] [110250] epoch = 88.11965965404978
+I1205 21:49:31.991833 137274321021824 utils.py:1231] [110250] img/sec/core = 164.20321798903797
+I1205 21:49:31.991905 137274321021824 utils.py:1231] [110250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 191.9549473900661
+I1205 21:49:31.991958 137274321021824 utils.py:1231] [110250] core_hours = 191.9549473900661
+I1205 21:49:31.992020 137274321021824 train.py:125] NOTE: Steps:110250/112603 [97.9%]
+Walltime:7d23h59m (0s eval)
+ETA:4h5m
+Total train time:8d4h3m
+I1205 21:54:43.791835 137274321021824 utils.py:1231] [110300] l2_params = 238.00580124599242
+I1205 21:54:43.792094 137274321021824 utils.py:1231] [110300] train/loss = 1.404422089457512
+I1205 21:54:43.792254 137274321021824 utils.py:1231] [110300] l2_grads = 2.9546568393707275
+I1205 21:54:43.792376 137274321021824 utils.py:1231] [110300] lr = 1.2436680102395749e-06
+I1205 21:54:43.792458 137274321021824 utils.py:1231] [110300] uptime = 691473.1548181149
+I1205 21:54:43.792531 137274321021824 utils.py:1231] [110300] examples_seen = 112947200.0
+I1205 21:54:43.792611 137274321021824 utils.py:1231] [110300] progress = 0.9795476141843468
+I1205 21:54:43.792678 137274321021824 utils.py:1231] [110300] epoch = 88.15962321851875
+I1205 21:54:43.792779 137274321021824 utils.py:1231] [110300] img/sec/core = 164.20738030863848
+I1205 21:54:43.792855 137274321021824 utils.py:1231] [110300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.0415587360086
+I1205 21:54:43.792928 137274321021824 utils.py:1231] [110300] core_hours = 192.0415587360086
+I1205 21:54:43.793006 137274321021824 train.py:125] NOTE: Steps:110300/112603 [98.0%]
+Walltime:8d0h4m (0s eval)
+ETA:4h0m
+Total train time:8d4h3m
+I1205 21:59:55.601337 137274321021824 utils.py:1231] [110350] l2_params = 238.00506173977675
+I1205 21:59:55.601599 137274321021824 utils.py:1231] [110350] train/loss = 1.449906349182129
+I1205 21:59:55.601712 137274321021824 utils.py:1231] [110350] l2_grads = 2.8185315132141113
+I1205 21:59:55.601788 137274321021824 utils.py:1231] [110350] lr = 1.1902962645512601e-06
+I1205 21:59:55.601850 137274321021824 utils.py:1231] [110350] uptime = 691784.964211225
+I1205 21:59:55.601922 137274321021824 utils.py:1231] [110350] examples_seen = 112998400.0
+I1205 21:59:55.601977 137274321021824 utils.py:1231] [110350] progress = 0.9799916520874222
+I1205 21:59:55.602029 137274321021824 utils.py:1231] [110350] epoch = 88.1995867829877
+I1205 21:59:55.602084 137274321021824 utils.py:1231] [110350] img/sec/core = 164.202878846311
+I1205 21:59:55.602144 137274321021824 utils.py:1231] [110350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.12817245631695
+I1205 21:59:55.602198 137274321021824 utils.py:1231] [110350] core_hours = 192.12817245631695
+I1205 21:59:55.602262 137274321021824 train.py:125] NOTE: Steps:110350/112603 [98.0%]
+Walltime:8d0h9m (0s eval)
+ETA:3h55m
+Total train time:8d4h3m
+I1205 22:05:07.413521 137274321021824 utils.py:1231] [110400] l2_params = 238.00434740249753
+I1205 22:05:07.413761 137274321021824 utils.py:1231] [110400] train/loss = 3.637014865875244
+I1205 22:05:07.413902 137274321021824 utils.py:1231] [110400] l2_grads = 3.165773391723633
+I1205 22:05:07.413991 137274321021824 utils.py:1231] [110400] lr = 1.1380936263567045e-06
+I1205 22:05:07.414055 137274321021824 utils.py:1231] [110400] uptime = 692096.776416334
+I1205 22:05:07.414120 137274321021824 utils.py:1231] [110400] examples_seen = 113049600.0
+I1205 22:05:07.414188 137274321021824 utils.py:1231] [110400] progress = 0.9804356899904976
+I1205 22:05:07.414253 137274321021824 utils.py:1231] [110400] epoch = 88.23955034745666
+I1205 22:05:07.414324 137274321021824 utils.py:1231] [110400] img/sec/core = 164.20139802453687
+I1205 22:05:07.414390 137274321021824 utils.py:1231] [110400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.2147869577361
+I1205 22:05:07.414457 137274321021824 utils.py:1231] [110400] core_hours = 192.2147869577361
+I1205 22:05:07.414532 137274321021824 train.py:125] NOTE: Steps:110400/112603 [98.0%]
+Walltime:8d0h14m (0s eval)
+ETA:3h50m
+Total train time:8d4h3m
+I1205 22:10:19.226675 137274321021824 utils.py:1231] [110450] l2_params = 238.00351881345577
+I1205 22:10:19.226901 137274321021824 utils.py:1231] [110450] train/loss = 2.014548882842064
+I1205 22:10:19.227003 137274321021824 utils.py:1231] [110450] l2_grads = 2.646606683731079
+I1205 22:10:19.227078 137274321021824 utils.py:1231] [110450] lr = 1.0870602180082599e-06
+I1205 22:10:19.227139 137274321021824 utils.py:1231] [110450] uptime = 692408.589500385
+I1205 22:10:19.227199 137274321021824 utils.py:1231] [110450] examples_seen = 113100800.0
+I1205 22:10:19.227258 137274321021824 utils.py:1231] [110450] progress = 0.980879727893573
+I1205 22:10:19.227319 137274321021824 utils.py:1231] [110450] epoch = 88.2795139119256
+I1205 22:10:19.227378 137274321021824 utils.py:1231] [110450] img/sec/core = 164.20093517186604
+I1205 22:10:19.227442 137274321021824 utils.py:1231] [110450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.3014017033058
+I1205 22:10:19.227500 137274321021824 utils.py:1231] [110450] core_hours = 192.3014017033058
+I1205 22:10:19.227571 137274321021824 train.py:125] NOTE: Steps:110450/112603 [98.1%]
+Walltime:8d0h20m (0s eval)
+ETA:3h44m
+Total train time:8d4h3m
+I1205 22:15:31.032156 137274321021824 utils.py:1231] [110500] l2_params = 238.00271984157064
+I1205 22:15:31.032435 137274321021824 utils.py:1231] [110500] train/loss = 1.4036514312028885
+I1205 22:15:31.032615 137274321021824 utils.py:1231] [110500] l2_grads = 2.8142333030700684
+I1205 22:15:31.032716 137274321021824 utils.py:1231] [110500] lr = 1.0371961591176346e-06
+I1205 22:15:31.032799 137274321021824 utils.py:1231] [110500] uptime = 692720.395160766
+I1205 22:15:31.032861 137274321021824 utils.py:1231] [110500] examples_seen = 113152000.0
+I1205 22:15:31.032934 137274321021824 utils.py:1231] [110500] progress = 0.9813237657966484
+I1205 22:15:31.032993 137274321021824 utils.py:1231] [110500] epoch = 88.31947747639457
+I1205 22:15:31.033051 137274321021824 utils.py:1231] [110500] img/sec/core = 164.20484457351074
+I1205 22:15:31.033118 137274321021824 utils.py:1231] [110500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.388014386745
+I1205 22:15:31.033176 137274321021824 utils.py:1231] [110500] core_hours = 192.388014386745
+I1205 22:15:31.033244 137274321021824 train.py:125] NOTE: Steps:110500/112603 [98.1%]
+Walltime:8d0h25m (0s eval)
+ETA:3h39m
+Total train time:8d4h3m
+I1205 22:20:42.844143 137274321021824 utils.py:1231] [110550] l2_params = 238.00188048542947
+I1205 22:20:42.844347 137274321021824 utils.py:1231] [110550] train/loss = 1.5451355576515198
+I1205 22:20:42.844457 137274321021824 utils.py:1231] [110550] l2_grads = 2.8592631816864014
+I1205 22:20:42.844550 137274321021824 utils.py:1231] [110550] lr = 9.885015665560086e-07
+I1205 22:20:42.844670 137274321021824 utils.py:1231] [110550] uptime = 693032.207013374
+I1205 22:20:42.844762 137274321021824 utils.py:1231] [110550] examples_seen = 113203200.0
+I1205 22:20:42.844860 137274321021824 utils.py:1231] [110550] progress = 0.9817678036997238
+I1205 22:20:42.844945 137274321021824 utils.py:1231] [110550] epoch = 88.35944104086353
+I1205 22:20:42.845040 137274321021824 utils.py:1231] [110550] img/sec/core = 164.20158365297377
+I1205 22:20:42.845146 137274321021824 utils.py:1231] [110550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.4746287902472
+I1205 22:20:42.845215 137274321021824 utils.py:1231] [110550] core_hours = 192.4746287902472
+I1205 22:20:42.845303 137274321021824 train.py:125] NOTE: Steps:110550/112603 [98.2%]
+Walltime:8d0h30m (0s eval)
+ETA:3h34m
+Total train time:8d4h3m
+I1205 22:25:54.647625 137274321021824 utils.py:1231] [110600] l2_params = 238.00110936745278
+I1205 22:25:54.647869 137274321021824 utils.py:1231] [110600] train/loss = 1.475963681936264
+I1205 22:25:54.647979 137274321021824 utils.py:1231] [110600] l2_grads = 2.8298096656799316
+I1205 22:25:54.648045 137274321021824 utils.py:1231] [110600] lr = 9.409765544535298e-07
+I1205 22:25:54.648105 137274321021824 utils.py:1231] [110600] uptime = 693344.010466394
+I1205 22:25:54.648158 137274321021824 utils.py:1231] [110600] examples_seen = 113254400.0
+I1205 22:25:54.648209 137274321021824 utils.py:1231] [110600] progress = 0.9822118416027992
+I1205 22:25:54.648259 137274321021824 utils.py:1231] [110600] epoch = 88.39940460533248
+I1205 22:25:54.648311 137274321021824 utils.py:1231] [110600] img/sec/core = 164.206007034525
+I1205 22:25:54.648369 137274321021824 utils.py:1231] [110600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.56124086053057
+I1205 22:25:54.648425 137274321021824 utils.py:1231] [110600] core_hours = 192.56124086053057
+I1205 22:25:54.648498 137274321021824 train.py:125] NOTE: Steps:110600/112603 [98.2%]
+Walltime:8d0h35m (0s eval)
+ETA:3h29m
+Total train time:8d4h3m
+I1205 22:31:06.439044 137274321021824 utils.py:1231] [110650] l2_params = 238.00039159355384
+I1205 22:31:06.439293 137274321021824 utils.py:1231] [110650] train/loss = 3.5468196272850037
+I1205 22:31:06.439410 137274321021824 utils.py:1231] [110650] l2_grads = 2.892423629760742
+I1205 22:31:06.439478 137274321021824 utils.py:1231] [110650] lr = 8.946212341989309e-07
+I1205 22:31:06.439544 137274321021824 utils.py:1231] [110650] uptime = 693655.801906111
+I1205 22:31:06.439594 137274321021824 utils.py:1231] [110650] examples_seen = 113305600.0
+I1205 22:31:06.439659 137274321021824 utils.py:1231] [110650] progress = 0.9826558795058746
+I1205 22:31:06.439704 137274321021824 utils.py:1231] [110650] epoch = 88.43936816980144
+I1205 22:31:06.439755 137274321021824 utils.py:1231] [110650] img/sec/core = 164.21233388088663
+I1205 22:31:06.439808 137274321021824 utils.py:1231] [110650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.64784959378528
+I1205 22:31:06.439855 137274321021824 utils.py:1231] [110650] core_hours = 192.64784959378528
+I1205 22:31:06.439917 137274321021824 train.py:125] NOTE: Steps:110650/112603 [98.3%]
+Walltime:8d0h40m (0s eval)
+ETA:3h24m
+Total train time:8d4h3m
+I1205 22:36:18.238721 137274321021824 utils.py:1231] [110700] l2_params = 237.99985702444238
+I1205 22:36:18.238983 137274321021824 utils.py:1231] [110700] train/loss = 1.5112729370594025
+I1205 22:36:18.239131 137274321021824 utils.py:1231] [110700] l2_grads = 2.911806106567383
+I1205 22:36:18.239223 137274321021824 utils.py:1231] [110700] lr = 8.494357144397472e-07
+I1205 22:36:18.239299 137274321021824 utils.py:1231] [110700] uptime = 693967.601660467
+I1205 22:36:18.239375 137274321021824 utils.py:1231] [110700] examples_seen = 113356800.0
+I1205 22:36:18.239450 137274321021824 utils.py:1231] [110700] progress = 0.98309991740895
+I1205 22:36:18.239519 137274321021824 utils.py:1231] [110700] epoch = 88.47933173427039
+I1205 22:36:18.239586 137274321021824 utils.py:1231] [110700] img/sec/core = 164.20795489638638
+I1205 22:36:18.239669 137274321021824 utils.py:1231] [110700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.73446063666194
+I1205 22:36:18.239737 137274321021824 utils.py:1231] [110700] core_hours = 192.73446063666194
+I1205 22:36:18.239813 137274321021824 train.py:125] NOTE: Steps:110700/112603 [98.3%]
+Walltime:8d0h46m (0s eval)
+ETA:3h18m
+Total train time:8d4h3m
+I1205 22:41:30.031853 137274321021824 utils.py:1231] [110750] l2_params = 237.9992731252558
+I1205 22:41:30.032133 137274321021824 utils.py:1231] [110750] train/loss = 1.3749333173036575
+I1205 22:41:30.032386 137274321021824 utils.py:1231] [110750] l2_grads = 2.710144519805908
+I1205 22:41:30.032547 137274321021824 utils.py:1231] [110750] lr = 8.054201010814298e-07
+I1205 22:41:30.032678 137274321021824 utils.py:1231] [110750] uptime = 694279.395024998
+I1205 22:41:30.032776 137274321021824 utils.py:1231] [110750] examples_seen = 113408000.0
+I1205 22:41:30.032892 137274321021824 utils.py:1231] [110750] progress = 0.9835439553120254
+I1205 22:41:30.032973 137274321021824 utils.py:1231] [110750] epoch = 88.51929529873935
+I1205 22:41:30.033042 137274321021824 utils.py:1231] [110750] img/sec/core = 164.21132013830137
+I1205 22:41:30.033111 137274321021824 utils.py:1231] [110750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.8210699045872
+I1205 22:41:30.033186 137274321021824 utils.py:1231] [110750] core_hours = 192.8210699045872
+I1205 22:41:30.033262 137274321021824 train.py:125] NOTE: Steps:110750/112603 [98.4%]
+Walltime:8d0h51m (0s eval)
+ETA:3h13m
+Total train time:8d4h3m
+I1205 22:46:41.823755 137274321021824 utils.py:1231] [110800] l2_params = 237.998667033053
+I1205 22:46:41.824005 137274321021824 utils.py:1231] [110800] train/loss = 2.042136400938034
+I1205 22:46:41.824125 137274321021824 utils.py:1231] [110800] l2_grads = 2.6745030879974365
+I1205 22:46:41.824206 137274321021824 utils.py:1231] [110800] lr = 7.625744972875673e-07
+I1205 22:46:41.824269 137274321021824 utils.py:1231] [110800] uptime = 694591.186630956
+I1205 22:46:41.824325 137274321021824 utils.py:1231] [110800] examples_seen = 113459200.0
+I1205 22:46:41.824377 137274321021824 utils.py:1231] [110800] progress = 0.9839879932151009
+I1205 22:46:41.824425 137274321021824 utils.py:1231] [110800] epoch = 88.55925886320831
+I1205 22:46:41.824476 137274321021824 utils.py:1231] [110800] img/sec/core = 164.2122463261381
+I1205 22:46:41.824532 137274321021824 utils.py:1231] [110800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.90767868402
+I1205 22:46:41.824586 137274321021824 utils.py:1231] [110800] core_hours = 192.90767868402
+I1205 22:46:41.824647 137274321021824 train.py:125] NOTE: Steps:110800/112603 [98.4%]
+Walltime:8d0h56m (0s eval)
+ETA:3h8m
+Total train time:8d4h3m
+I1205 22:51:53.619351 137274321021824 utils.py:1231] [110850] l2_params = 237.99809031461658
+I1205 22:51:53.619661 137274321021824 utils.py:1231] [110850] train/loss = 1.4261268824338913
+I1205 22:51:53.619830 137274321021824 utils.py:1231] [110850] l2_grads = 2.720221519470215
+I1205 22:51:53.619921 137274321021824 utils.py:1231] [110850] lr = 7.208990034794984e-07
+I1205 22:51:53.619991 137274321021824 utils.py:1231] [110850] uptime = 694902.982353149
+I1205 22:51:53.620045 137274321021824 utils.py:1231] [110850] examples_seen = 113510400.0
+I1205 22:51:53.620095 137274321021824 utils.py:1231] [110850] progress = 0.9844320311181762
+I1205 22:51:53.620144 137274321021824 utils.py:1231] [110850] epoch = 88.59922242767726
+I1205 22:51:53.620197 137274321021824 utils.py:1231] [110850] img/sec/core = 164.21007844460632
+I1205 22:51:53.620256 137274321021824 utils.py:1231] [110850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 192.9942886068514
+I1205 22:51:53.620308 137274321021824 utils.py:1231] [110850] core_hours = 192.9942886068514
+I1205 22:51:53.620369 137274321021824 train.py:125] NOTE: Steps:110850/112603 [98.4%]
+Walltime:8d1h1m (0s eval)
+ETA:3h3m
+Total train time:8d4h2m
+I1205 22:57:05.305525 137274321021824 utils.py:1231] [110900] l2_params = 237.9974607703598
+I1205 22:57:05.305860 137274321021824 utils.py:1231] [110900] train/loss = 2.6150334179401398
+I1205 22:57:05.306066 137274321021824 utils.py:1231] [110900] l2_grads = 2.631784677505493
+I1205 22:57:05.306132 137274321021824 utils.py:1231] [110900] lr = 6.803937173359206e-07
+I1205 22:57:05.306205 137274321021824 utils.py:1231] [110900] uptime = 695214.668558145
+I1205 22:57:05.306258 137274321021824 utils.py:1231] [110900] examples_seen = 113561600.0
+I1205 22:57:05.306310 137274321021824 utils.py:1231] [110900] progress = 0.9848760690212517
+I1205 22:57:05.306362 137274321021824 utils.py:1231] [110900] epoch = 88.63918599214622
+I1205 22:57:05.306416 137274321021824 utils.py:1231] [110900] img/sec/core = 164.26777694785577
+I1205 22:57:05.306476 137274321021824 utils.py:1231] [110900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.08086810823914
+I1205 22:57:05.306533 137274321021824 utils.py:1231] [110900] core_hours = 193.08086810823914
+I1205 22:57:05.306594 137274321021824 train.py:125] NOTE: Steps:110900/112603 [98.5%]
+Walltime:8d1h6m (0s eval)
+ETA:2h57m
+Total train time:8d4h2m
+I1205 23:02:17.106473 137274321021824 utils.py:1231] [110950] l2_params = 237.99689762150584
+I1205 23:02:17.106694 137274321021824 utils.py:1231] [110950] train/loss = 1.5703648775815964
+I1205 23:02:17.106799 137274321021824 utils.py:1231] [110950] l2_grads = 2.8012032508850098
+I1205 23:02:17.106896 137274321021824 utils.py:1231] [110950] lr = 6.410587337930041e-07
+I1205 23:02:17.106969 137274321021824 utils.py:1231] [110950] uptime = 695526.469329522
+I1205 23:02:17.107029 137274321021824 utils.py:1231] [110950] examples_seen = 113612800.0
+I1205 23:02:17.107088 137274321021824 utils.py:1231] [110950] progress = 0.985320106924327
+I1205 23:02:17.107143 137274321021824 utils.py:1231] [110950] epoch = 88.67914955661519
+I1205 23:02:17.107211 137274321021824 utils.py:1231] [110950] img/sec/core = 164.20741928853568
+I1205 23:02:17.107301 137274321021824 utils.py:1231] [110950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.16747943362168
+I1205 23:02:17.107367 137274321021824 utils.py:1231] [110950] core_hours = 193.16747943362168
+I1205 23:02:17.107479 137274321021824 train.py:125] NOTE: Steps:110950/112603 [98.5%]
+Walltime:8d1h12m (0s eval)
+ETA:2h52m
+Total train time:8d4h2m
+I1205 23:07:28.896328 137274321021824 utils.py:1231] [111000] l2_params = 237.99629545684567
+I1205 23:07:28.896666 137274321021824 utils.py:1231] [111000] train/loss = 3.3144360184669495
+I1205 23:07:28.896872 137274321021824 utils.py:1231] [111000] l2_grads = 2.8158137798309326
+I1205 23:07:28.896967 137274321021824 utils.py:1231] [111000] lr = 6.028941450438361e-07
+I1205 23:07:28.897032 137274321021824 utils.py:1231] [111000] uptime = 695838.259393101
+I1205 23:07:28.897110 137274321021824 utils.py:1231] [111000] examples_seen = 113664000.0
+I1205 23:07:28.897172 137274321021824 utils.py:1231] [111000] progress = 0.9857641448274025
+I1205 23:07:28.897240 137274321021824 utils.py:1231] [111000] epoch = 88.71911312108413
+I1205 23:07:28.897300 137274321021824 utils.py:1231] [111000] img/sec/core = 164.21305865967625
+I1205 23:07:28.897374 137274321021824 utils.py:1231] [111000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.25408778461582
+I1205 23:07:28.897433 137274321021824 utils.py:1231] [111000] core_hours = 193.25408778461582
+I1205 23:07:28.897503 137274321021824 train.py:125] NOTE: Steps:111000/112603 [98.6%]
+Walltime:8d1h17m (0s eval)
+ETA:2h47m
+Total train time:8d4h2m
+I1205 23:12:41.059703 137274321021824 utils.py:1231] [111050] l2_params = 237.9957027557
+I1205 23:12:41.059944 137274321021824 utils.py:1231] [111050] train/loss = 2.149446815252304
+I1205 23:12:41.060094 137274321021824 utils.py:1231] [111050] l2_grads = 2.7969002723693848
+I1205 23:12:41.060209 137274321021824 utils.py:1231] [111050] lr = 5.659000405383638e-07
+I1205 23:12:41.060302 137274321021824 utils.py:1231] [111050] uptime = 696150.422657738
+I1205 23:12:41.060398 137274321021824 utils.py:1231] [111050] examples_seen = 113715200.0
+I1205 23:12:41.060480 137274321021824 utils.py:1231] [111050] progress = 0.9862081827304778
+I1205 23:12:41.060557 137274321021824 utils.py:1231] [111050] epoch = 88.7590766855531
+I1205 23:12:41.060641 137274321021824 utils.py:1231] [111050] img/sec/core = 164.0167367532448
+I1205 23:12:41.060745 137274321021824 utils.py:1231] [111050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.34079980257053
+I1205 23:12:41.060828 137274321021824 utils.py:1231] [111050] core_hours = 193.34079980257053
+I1205 23:12:41.060929 137274321021824 train.py:125] NOTE: Steps:111050/112603 [98.6%]
+Walltime:8d1h22m (0s eval)
+ETA:2h42m
+Total train time:8d4h2m
+I1205 23:17:52.797932 137274321021824 utils.py:1231] [111100] l2_params = 237.99513455447033
+I1205 23:17:52.798143 137274321021824 utils.py:1231] [111100] train/loss = 1.3976501375436783
+I1205 23:17:52.798248 137274321021824 utils.py:1231] [111100] l2_grads = 2.902479887008667
+I1205 23:17:52.798312 137274321021824 utils.py:1231] [111100] lr = 5.30076506983173e-07
+I1205 23:17:52.798379 137274321021824 utils.py:1231] [111100] uptime = 696462.160739331
+I1205 23:17:52.798437 137274321021824 utils.py:1231] [111100] examples_seen = 113766400.0
+I1205 23:17:52.798487 137274321021824 utils.py:1231] [111100] progress = 0.9866522206335533
+I1205 23:17:52.798536 137274321021824 utils.py:1231] [111100] epoch = 88.79904025002205
+I1205 23:17:52.798587 137274321021824 utils.py:1231] [111100] img/sec/core = 164.24044100853533
+I1205 23:17:52.798643 137274321021824 utils.py:1231] [111100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.42739371412418
+I1205 23:17:52.798702 137274321021824 utils.py:1231] [111100] core_hours = 193.42739371412418
+I1205 23:17:52.798776 137274321021824 train.py:125] NOTE: Steps:111100/112603 [98.7%]
+Walltime:8d1h27m (0s eval)
+ETA:2h37m
+Total train time:8d4h2m
+I1205 23:23:04.593850 137274321021824 utils.py:1231] [111150] l2_params = 237.99451174588947
+I1205 23:23:04.594131 137274321021824 utils.py:1231] [111150] train/loss = 1.7933816313743591
+I1205 23:23:04.594244 137274321021824 utils.py:1231] [111150] l2_grads = 2.5851190090179443
+I1205 23:23:04.594327 137274321021824 utils.py:1231] [111150] lr = 4.954236283412671e-07
+I1205 23:23:04.594391 137274321021824 utils.py:1231] [111150] uptime = 696773.956748303
+I1205 23:23:04.594444 137274321021824 utils.py:1231] [111150] examples_seen = 113817600.0
+I1205 23:23:04.594500 137274321021824 utils.py:1231] [111150] progress = 0.9870962585366286
+I1205 23:23:04.594552 137274321021824 utils.py:1231] [111150] epoch = 88.83900381449101
+I1205 23:23:04.594606 137274321021824 utils.py:1231] [111150] img/sec/core = 164.20992740994407
+I1205 23:23:04.594663 137274321021824 utils.py:1231] [111150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.5140037166164
+I1205 23:23:04.594712 137274321021824 utils.py:1231] [111150] core_hours = 193.5140037166164
+I1205 23:23:04.594782 137274321021824 train.py:125] NOTE: Steps:111150/112603 [98.7%]
+Walltime:8d1h32m (0s eval)
+ETA:2h31m
+Total train time:8d4h2m
+I1205 23:28:16.393589 137274321021824 utils.py:1231] [111200] l2_params = 237.9938889385589
+I1205 23:28:16.393804 137274321021824 utils.py:1231] [111200] train/loss = 1.5070528984069824
+I1205 23:28:16.393916 137274321021824 utils.py:1231] [111200] l2_grads = 2.9189560413360596
+I1205 23:28:16.394003 137274321021824 utils.py:1231] [111200] lr = 4.6194148583190023e-07
+I1205 23:28:16.394090 137274321021824 utils.py:1231] [111200] uptime = 697085.756443258
+I1205 23:28:16.394199 137274321021824 utils.py:1231] [111200] examples_seen = 113868800.0
+I1205 23:28:16.394276 137274321021824 utils.py:1231] [111200] progress = 0.9875402964397041
+I1205 23:28:16.394343 137274321021824 utils.py:1231] [111200] epoch = 88.87896737895997
+I1205 23:28:16.394406 137274321021824 utils.py:1231] [111200] img/sec/core = 164.2079861796831
+I1205 23:28:16.394467 137274321021824 utils.py:1231] [111200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.6006147429928
+I1205 23:28:16.394531 137274321021824 utils.py:1231] [111200] core_hours = 193.6006147429928
+I1205 23:28:16.394609 137274321021824 train.py:125] NOTE: Steps:111200/112603 [98.8%]
+Walltime:8d1h38m (0s eval)
+ETA:2h26m
+Total train time:8d4h2m
+I1205 23:33:28.132280 137274321021824 utils.py:1231] [111250] l2_params = 237.9932837274906
+I1205 23:33:28.132482 137274321021824 utils.py:1231] [111250] train/loss = 3.641853243112564
+I1205 23:33:28.132586 137274321021824 utils.py:1231] [111250] l2_grads = 3.014561176300049
+I1205 23:33:28.132660 137274321021824 utils.py:1231] [111250] lr = 4.296301579303537e-07
+I1205 23:33:28.132728 137274321021824 utils.py:1231] [111250] uptime = 697397.49508852
+I1205 23:33:28.132788 137274321021824 utils.py:1231] [111250] examples_seen = 113920000.0
+I1205 23:33:28.132848 137274321021824 utils.py:1231] [111250] progress = 0.9879843343427795
+I1205 23:33:28.132911 137274321021824 utils.py:1231] [111250] epoch = 88.91893094342892
+I1205 23:33:28.132970 137274321021824 utils.py:1231] [111250] img/sec/core = 164.24014403788232
+I1205 23:33:28.133030 137274321021824 utils.py:1231] [111250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.6872088111211
+I1205 23:33:28.133086 137274321021824 utils.py:1231] [111250] core_hours = 193.6872088111211
+I1205 23:33:28.133158 137274321021824 train.py:125] NOTE: Steps:111250/112603 [98.8%]
+Walltime:8d1h43m (0s eval)
+ETA:2h21m
+Total train time:8d4h2m
+I1205 23:38:39.932893 137274321021824 utils.py:1231] [111300] l2_params = 237.99265243053588
+I1205 23:38:39.933144 137274321021824 utils.py:1231] [111300] train/loss = 1.440290316939354
+I1205 23:38:39.933265 137274321021824 utils.py:1231] [111300] l2_grads = 2.880838632583618
+I1205 23:38:39.933341 137274321021824 utils.py:1231] [111300] lr = 3.9848972036766e-07
+I1205 23:38:39.933405 137274321021824 utils.py:1231] [111300] uptime = 697709.295766227
+I1205 23:38:39.933466 137274321021824 utils.py:1231] [111300] examples_seen = 113971200.0
+I1205 23:38:39.933541 137274321021824 utils.py:1231] [111300] progress = 0.9884283722458549
+I1205 23:38:39.933601 137274321021824 utils.py:1231] [111300] epoch = 88.95889450789788
+I1205 23:38:39.933663 137274321021824 utils.py:1231] [111300] img/sec/core = 164.20746861916857
+I1205 23:38:39.933729 137274321021824 utils.py:1231] [111300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.77382011048414
+I1205 23:38:39.933786 137274321021824 utils.py:1231] [111300] core_hours = 193.77382011048414
+I1205 23:38:39.933852 137274321021824 train.py:125] NOTE: Steps:111300/112603 [98.8%]
+Walltime:8d1h48m (0s eval)
+ETA:2h16m
+Total train time:8d4h2m
+I1205 23:43:51.724130 137274321021824 utils.py:1231] [111350] l2_params = 237.99200687727802
+I1205 23:43:51.724345 137274321021824 utils.py:1231] [111350] train/loss = 3.065984219312668
+I1205 23:43:51.724480 137274321021824 utils.py:1231] [111350] l2_grads = 2.7435529232025146
+I1205 23:43:51.724564 137274321021824 utils.py:1231] [111350] lr = 3.6852024613065813e-07
+I1205 23:43:51.724617 137274321021824 utils.py:1231] [111350] uptime = 698021.086979105
+I1205 23:43:51.724671 137274321021824 utils.py:1231] [111350] examples_seen = 114022400.0
+I1205 23:43:51.724721 137274321021824 utils.py:1231] [111350] progress = 0.9888724101489303
+I1205 23:43:51.724772 137274321021824 utils.py:1231] [111350] epoch = 88.99885807236683
+I1205 23:43:51.724823 137274321021824 utils.py:1231] [111350] img/sec/core = 164.2124533510245
+I1205 23:43:51.724885 137274321021824 utils.py:1231] [111350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.86042878072806
+I1205 23:43:51.724937 137274321021824 utils.py:1231] [111350] core_hours = 193.86042878072806
+I1205 23:43:51.725005 137274321021824 train.py:125] NOTE: Steps:111350/112603 [98.9%]
+Walltime:8d1h53m (0s eval)
+ETA:2h10m
+Total train time:8d4h2m
+I1205 23:49:03.423993 137274321021824 utils.py:1231] [111400] l2_params = 237.9913507663539
+I1205 23:49:03.424273 137274321021824 utils.py:1231] [111400] train/loss = 1.9844349324703217
+I1205 23:49:03.424510 137274321021824 utils.py:1231] [111400] l2_grads = 2.6412901878356934
+I1205 23:49:03.424593 137274321021824 utils.py:1231] [111400] lr = 3.397218054616601e-07
+I1205 23:49:03.424661 137274321021824 utils.py:1231] [111400] uptime = 698332.7870221729
+I1205 23:49:03.424724 137274321021824 utils.py:1231] [111400] examples_seen = 114073600.0
+I1205 23:49:03.424774 137274321021824 utils.py:1231] [111400] progress = 0.9893164480520057
+I1205 23:49:03.424829 137274321021824 utils.py:1231] [111400] epoch = 89.03882163683579
+I1205 23:49:03.424880 137274321021824 utils.py:1231] [111400] img/sec/core = 164.26048420161345
+I1205 23:49:03.424947 137274321021824 utils.py:1231] [111400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 193.9470121260247
+I1205 23:49:03.425004 137274321021824 utils.py:1231] [111400] core_hours = 193.9470121260247
+I1205 23:49:03.425065 137274321021824 train.py:125] NOTE: Steps:111400/112603 [98.9%]
+Walltime:8d1h58m (0s eval)
+ETA:2h5m
+Total train time:8d4h2m
+I1205 23:54:15.100353 137274321021824 utils.py:1231] [111450] l2_params = 237.9907284413817
+I1205 23:54:15.100613 137274321021824 utils.py:1231] [111450] train/loss = 1.292728140950203
+I1205 23:54:15.100737 137274321021824 utils.py:1231] [111450] l2_grads = 2.919813871383667
+I1205 23:54:15.100843 137274321021824 utils.py:1231] [111450] lr = 3.120944658582289e-07
+I1205 23:54:15.100923 137274321021824 utils.py:1231] [111450] uptime = 698644.463284077
+I1205 23:54:15.100997 137274321021824 utils.py:1231] [111450] examples_seen = 114124800.0
+I1205 23:54:15.101055 137274321021824 utils.py:1231] [111450] progress = 0.9897604859550811
+I1205 23:54:15.101113 137274321021824 utils.py:1231] [111450] epoch = 89.07878520130475
+I1205 23:54:15.101172 137274321021824 utils.py:1231] [111450] img/sec/core = 164.27301741625465
+I1205 23:54:15.101236 137274321021824 utils.py:1231] [111450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.0335888654425
+I1205 23:54:15.101290 137274321021824 utils.py:1231] [111450] core_hours = 194.0335888654425
+I1205 23:54:15.101355 137274321021824 train.py:125] NOTE: Steps:111450/112603 [99.0%]
+Walltime:8d2h4m (0s eval)
+ETA:2h0m
+Total train time:8d4h2m
+I1205 23:59:27.055516 137274321021824 utils.py:1231] [111500] l2_params = 237.99041177314894
+I1205 23:59:27.055734 137274321021824 utils.py:1231] [111500] train/loss = 1.9478602856397629
+I1205 23:59:27.055837 137274321021824 utils.py:1231] [111500] l2_grads = 2.5959548950195312
+I1205 23:59:27.055932 137274321021824 utils.py:1231] [111500] lr = 2.856382920732896e-07
+I1205 23:59:27.056004 137274321021824 utils.py:1231] [111500] uptime = 698956.418364451
+I1205 23:59:27.056066 137274321021824 utils.py:1231] [111500] examples_seen = 114176000.0
+I1205 23:59:27.056128 137274321021824 utils.py:1231] [111500] progress = 0.9902045238581565
+I1205 23:59:27.056194 137274321021824 utils.py:1231] [111500] epoch = 89.1187487657737
+I1205 23:59:27.056255 137274321021824 utils.py:1231] [111500] img/sec/core = 164.1261938693618
+I1205 23:59:27.056329 137274321021824 utils.py:1231] [111500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.1202430544353
+I1205 23:59:27.056421 137274321021824 utils.py:1231] [111500] core_hours = 194.1202430544353
+I1205 23:59:27.056514 137274321021824 train.py:125] NOTE: Steps:111500/112603 [99.0%]
+Walltime:8d2h9m (0s eval)
+ETA:1h55m
+Total train time:8d4h2m
+I1206 00:04:38.843294 137274321021824 utils.py:1231] [111550] l2_params = 237.99044935088045
+I1206 00:04:38.843557 137274321021824 utils.py:1231] [111550] train/loss = 1.9869207739830017
+I1206 00:04:38.843734 137274321021824 utils.py:1231] [111550] l2_grads = 2.5834710597991943
+I1206 00:04:38.843829 137274321021824 utils.py:1231] [111550] lr = 2.6035334611457464e-07
+I1206 00:04:38.843911 137274321021824 utils.py:1231] [111550] uptime = 699268.206271724
+I1206 00:04:38.843976 137274321021824 utils.py:1231] [111550] examples_seen = 114227200.0
+I1206 00:04:38.844035 137274321021824 utils.py:1231] [111550] progress = 0.990648561761232
+I1206 00:04:38.844092 137274321021824 utils.py:1231] [111550] epoch = 89.15871233024266
+I1206 00:04:38.844152 137274321021824 utils.py:1231] [111550] img/sec/core = 164.2141943471258
+I1206 00:04:38.844231 137274321021824 utils.py:1231] [111550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.20685080645552
+I1206 00:04:38.844292 137274321021824 utils.py:1231] [111550] core_hours = 194.20685080645552
+I1206 00:04:38.844362 137274321021824 train.py:125] NOTE: Steps:111550/112603 [99.1%]
+Walltime:8d2h14m (0s eval)
+ETA:1h49m
+Total train time:8d4h2m
+I1206 00:09:50.628680 137274321021824 utils.py:1231] [111600] l2_params = 237.99047200355696
+I1206 00:09:50.628938 137274321021824 utils.py:1231] [111600] train/loss = 1.3300229609012604
+I1206 00:09:50.629049 137274321021824 utils.py:1231] [111600] l2_grads = 2.6672568321228027
+I1206 00:09:50.629146 137274321021824 utils.py:1231] [111600] lr = 2.3623968724484547e-07
+I1206 00:09:50.629225 137274321021824 utils.py:1231] [111600] uptime = 699579.991583175
+I1206 00:09:50.629290 137274321021824 utils.py:1231] [111600] examples_seen = 114278400.0
+I1206 00:09:50.629345 137274321021824 utils.py:1231] [111600] progress = 0.9910925996643073
+I1206 00:09:50.629404 137274321021824 utils.py:1231] [111600] epoch = 89.19867589471161
+I1206 00:09:50.629471 137274321021824 utils.py:1231] [111600] img/sec/core = 164.21556154044745
+I1206 00:09:50.629534 137274321021824 utils.py:1231] [111600] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.29345783741417
+I1206 00:09:50.629589 137274321021824 utils.py:1231] [111600] core_hours = 194.29345783741417
+I1206 00:09:50.629655 137274321021824 train.py:125] NOTE: Steps:111600/112603 [99.1%]
+Walltime:8d2h19m (0s eval)
+ETA:1h44m
+Total train time:8d4h2m
+I1206 00:15:02.421772 137274321021824 utils.py:1231] [111650] l2_params = 237.9904804360733
+I1206 00:15:02.422029 137274321021824 utils.py:1231] [111650] train/loss = 1.4260942935943604
+I1206 00:15:02.422163 137274321021824 utils.py:1231] [111650] l2_grads = 2.8094139099121094
+I1206 00:15:02.422240 137274321021824 utils.py:1231] [111650] lr = 2.1329737198150412e-07
+I1206 00:15:02.422313 137274321021824 utils.py:1231] [111650] uptime = 699891.78466873
+I1206 00:15:02.422377 137274321021824 utils.py:1231] [111650] examples_seen = 114329600.0
+I1206 00:15:02.422431 137274321021824 utils.py:1231] [111650] progress = 0.9915366375673828
+I1206 00:15:02.422489 137274321021824 utils.py:1231] [111650] epoch = 89.23863945918058
+I1206 00:15:02.422546 137274321021824 utils.py:1231] [111650] img/sec/core = 164.21146706595036
+I1206 00:15:02.422616 137274321021824 utils.py:1231] [111650] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.3800670278461
+I1206 00:15:02.422668 137274321021824 utils.py:1231] [111650] core_hours = 194.3800670278461
+I1206 00:15:02.422736 137274321021824 train.py:125] NOTE: Steps:111650/112603 [99.2%]
+Walltime:8d2h24m (0s eval)
+ETA:1h39m
+Total train time:8d4h2m
+I1206 00:20:14.213804 137274321021824 utils.py:1231] [111700] l2_params = 237.99049112952397
+I1206 00:20:14.214082 137274321021824 utils.py:1231] [111700] train/loss = 1.3853723853826523
+I1206 00:20:14.214260 137274321021824 utils.py:1231] [111700] l2_grads = 2.8053836822509766
+I1206 00:20:14.214364 137274321021824 utils.py:1231] [111700] lr = 1.9152645409675932e-07
+I1206 00:20:14.214447 137274321021824 utils.py:1231] [111700] uptime = 700203.5768022169
+I1206 00:20:14.214548 137274321021824 utils.py:1231] [111700] examples_seen = 114380800.0
+I1206 00:20:14.214622 137274321021824 utils.py:1231] [111700] progress = 0.9919806754704582
+I1206 00:20:14.214700 137274321021824 utils.py:1231] [111700] epoch = 89.27860302364954
+I1206 00:20:14.214797 137274321021824 utils.py:1231] [111700] img/sec/core = 164.21196849132053
+I1206 00:20:14.214887 137274321021824 utils.py:1231] [111700] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.4666759538147
+I1206 00:20:14.214973 137274321021824 utils.py:1231] [111700] core_hours = 194.4666759538147
+I1206 00:20:14.215066 137274321021824 train.py:125] NOTE: Steps:111700/112603 [99.2%]
+Walltime:8d2h30m (0s eval)
+ETA:1h34m
+Total train time:8d4h2m
+I1206 00:25:25.988559 137274321021824 utils.py:1231] [111750] l2_params = 237.99049299825433
+I1206 00:25:25.988775 137274321021824 utils.py:1231] [111750] train/loss = 3.4164077639579773
+I1206 00:25:25.988885 137274321021824 utils.py:1231] [111750] l2_grads = 2.924600839614868
+I1206 00:25:25.988952 137274321021824 utils.py:1231] [111750] lr = 1.7092698461707176e-07
+I1206 00:25:25.989005 137274321021824 utils.py:1231] [111750] uptime = 700515.351367462
+I1206 00:25:25.989058 137274321021824 utils.py:1231] [111750] examples_seen = 114432000.0
+I1206 00:25:25.989106 137274321021824 utils.py:1231] [111750] progress = 0.9924247133735336
+I1206 00:25:25.989154 137274321021824 utils.py:1231] [111750] epoch = 89.31856658811849
+I1206 00:25:25.989205 137274321021824 utils.py:1231] [111750] img/sec/core = 164.22122170152565
+I1206 00:25:25.989262 137274321021824 utils.py:1231] [111750] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.55327999971612
+I1206 00:25:25.989313 137274321021824 utils.py:1231] [111750] core_hours = 194.55327999971612
+I1206 00:25:25.989380 137274321021824 train.py:125] NOTE: Steps:111750/112603 [99.2%]
+Walltime:8d2h35m (0s eval)
+ETA:1h29m
+Total train time:8d4h2m
+I1206 00:30:37.716497 137274321021824 utils.py:1231] [111800] l2_params = 237.9905037650862
+I1206 00:30:37.716733 137274321021824 utils.py:1231] [111800] train/loss = 2.937595009803772
+I1206 00:30:37.716834 137274321021824 utils.py:1231] [111800] l2_grads = 2.762606620788574
+I1206 00:30:37.716911 137274321021824 utils.py:1231] [111800] lr = 1.5149901182337641e-07
+I1206 00:30:37.716976 137274321021824 utils.py:1231] [111800] uptime = 700827.079336904
+I1206 00:30:37.717038 137274321021824 utils.py:1231] [111800] examples_seen = 114483200.0
+I1206 00:30:37.717097 137274321021824 utils.py:1231] [111800] progress = 0.992868751276609
+I1206 00:30:37.717154 137274321021824 utils.py:1231] [111800] epoch = 89.35853015258745
+I1206 00:30:37.717215 137274321021824 utils.py:1231] [111800] img/sec/core = 164.24576880811964
+I1206 00:30:37.717279 137274321021824 utils.py:1231] [111800] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.63987110233887
+I1206 00:30:37.717337 137274321021824 utils.py:1231] [111800] core_hours = 194.63987110233887
+I1206 00:30:37.717404 137274321021824 train.py:125] NOTE: Steps:111800/112603 [99.3%]
+Walltime:8d2h40m (0s eval)
+ETA:1h23m
+Total train time:8d4h2m
+I1206 00:35:49.511214 137274321021824 utils.py:1231] [111850] l2_params = 237.9905226732847
+I1206 00:35:49.511419 137274321021824 utils.py:1231] [111850] train/loss = 1.596910998225212
+I1206 00:35:49.511524 137274321021824 utils.py:1231] [111850] l2_grads = 2.970761299133301
+I1206 00:35:49.511599 137274321021824 utils.py:1231] [111850] lr = 1.3324258125085988e-07
+I1206 00:35:49.511657 137274321021824 utils.py:1231] [111850] uptime = 701138.8740190039
+I1206 00:35:49.511717 137274321021824 utils.py:1231] [111850] examples_seen = 114534400.0
+I1206 00:35:49.511774 137274321021824 utils.py:1231] [111850] progress = 0.9933127891796844
+I1206 00:35:49.511830 137274321021824 utils.py:1231] [111850] epoch = 89.3984937170564
+I1206 00:35:49.511893 137274321021824 utils.py:1231] [111850] img/sec/core = 164.21062622096684
+I1206 00:35:49.511957 137274321021824 utils.py:1231] [111850] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.72648073625552
+I1206 00:35:49.512011 137274321021824 utils.py:1231] [111850] core_hours = 194.72648073625552
+I1206 00:35:49.512077 137274321021824 train.py:125] NOTE: Steps:111850/112603 [99.3%]
+Walltime:8d2h45m (0s eval)
+ETA:1h18m
+Total train time:8d4h2m
+I1206 00:41:01.359843 137274321021824 utils.py:1231] [111900] l2_params = 237.99053033954857
+I1206 00:41:01.360089 137274321021824 utils.py:1231] [111900] train/loss = 1.4047198444604874
+I1206 00:41:01.360200 137274321021824 utils.py:1231] [111900] l2_grads = 2.6954293251037598
+I1206 00:41:01.360289 137274321021824 utils.py:1231] [111900] lr = 1.1615773568890532e-07
+I1206 00:41:01.360351 137274321021824 utils.py:1231] [111900] uptime = 701450.722712098
+I1206 00:41:01.360412 137274321021824 utils.py:1231] [111900] examples_seen = 114585600.0
+I1206 00:41:01.360469 137274321021824 utils.py:1231] [111900] progress = 0.9937568270827598
+I1206 00:41:01.360542 137274321021824 utils.py:1231] [111900] epoch = 89.43845728152536
+I1206 00:41:01.360607 137274321021824 utils.py:1231] [111900] img/sec/core = 164.1821855721245
+I1206 00:41:01.360688 137274321021824 utils.py:1231] [111900] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.81310537322614
+I1206 00:41:01.360749 137274321021824 utils.py:1231] [111900] core_hours = 194.81310537322614
+I1206 00:41:01.360823 137274321021824 train.py:125] NOTE: Steps:111900/112603 [99.4%]
+Walltime:8d2h50m (0s eval)
+ETA:1h13m
+Total train time:8d4h2m
+I1206 00:46:13.146164 137274321021824 utils.py:1231] [111950] l2_params = 237.9905390995258
+I1206 00:46:13.146361 137274321021824 utils.py:1231] [111950] train/loss = 3.6407236754894257
+I1206 00:46:13.146466 137274321021824 utils.py:1231] [111950] l2_grads = 3.0577900409698486
+I1206 00:46:13.146537 137274321021824 utils.py:1231] [111950] lr = 1.0024451518075941e-07
+I1206 00:46:13.146597 137274321021824 utils.py:1231] [111950] uptime = 701762.508958668
+I1206 00:46:13.146657 137274321021824 utils.py:1231] [111950] examples_seen = 114636800.0
+I1206 00:46:13.146714 137274321021824 utils.py:1231] [111950] progress = 0.9942008649858352
+I1206 00:46:13.146770 137274321021824 utils.py:1231] [111950] epoch = 89.47842084599432
+I1206 00:46:13.146828 137274321021824 utils.py:1231] [111950] img/sec/core = 164.21506902009614
+I1206 00:46:13.146894 137274321021824 utils.py:1231] [111950] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.89971266394
+I1206 00:46:13.146951 137274321021824 utils.py:1231] [111950] core_hours = 194.89971266394
+I1206 00:46:13.147013 137274321021824 train.py:125] NOTE: Steps:111950/112603 [99.4%]
+Walltime:8d2h56m (0s eval)
+ETA:1h8m
+Total train time:8d4h2m
+I1206 00:51:24.916253 137274321021824 utils.py:1231] [112000] l2_params = 237.99054249812738
+I1206 00:51:24.916530 137274321021824 utils.py:1231] [112000] train/loss = 3.232933223247528
+I1206 00:51:24.916700 137274321021824 utils.py:1231] [112000] l2_grads = 2.655447006225586
+I1206 00:51:24.916783 137274321021824 utils.py:1231] [112000] lr = 8.550295702386487e-08
+I1206 00:51:24.916849 137274321021824 utils.py:1231] [112000] uptime = 702074.279211178
+I1206 00:51:24.916952 137274321021824 utils.py:1231] [112000] examples_seen = 114688000.0
+I1206 00:51:24.917003 137274321021824 utils.py:1231] [112000] progress = 0.9946449028889106
+I1206 00:51:24.917051 137274321021824 utils.py:1231] [112000] epoch = 89.51838441046327
+I1206 00:51:24.917101 137274321021824 utils.py:1231] [112000] img/sec/core = 164.22349338271792
+I1206 00:51:24.917159 137274321021824 utils.py:1231] [112000] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 194.98631551185943
+I1206 00:51:24.917208 137274321021824 utils.py:1231] [112000] core_hours = 194.98631551185943
+I1206 00:51:24.917268 137274321021824 train.py:125] NOTE: Steps:112000/112603 [99.5%]
+Walltime:8d3h1m (0s eval)
+ETA:1h2m
+Total train time:8d4h2m
+I1206 00:56:37.054757 137274321021824 utils.py:1231] [112050] l2_params = 237.99055336977017
+I1206 00:56:37.054979 137274321021824 utils.py:1231] [112050] train/loss = 1.4369406700134277
+I1206 00:56:37.055107 137274321021824 utils.py:1231] [112050] l2_grads = 2.876708745956421
+I1206 00:56:37.055218 137274321021824 utils.py:1231] [112050] lr = 7.193309576930607e-08
+I1206 00:56:37.055303 137274321021824 utils.py:1231] [112050] uptime = 702386.4176638289
+I1206 00:56:37.055387 137274321021824 utils.py:1231] [112050] examples_seen = 114739200.0
+I1206 00:56:37.055461 137274321021824 utils.py:1231] [112050] progress = 0.995088940791986
+I1206 00:56:37.055530 137274321021824 utils.py:1231] [112050] epoch = 89.55834797493223
+I1206 00:56:37.055614 137274321021824 utils.py:1231] [112050] img/sec/core = 164.02977449641492
+I1206 00:56:37.055717 137274321021824 utils.py:1231] [112050] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.0730206375958
+I1206 00:56:37.055805 137274321021824 utils.py:1231] [112050] core_hours = 195.0730206375958
+I1206 00:56:37.055923 137274321021824 train.py:125] NOTE: Steps:112050/112603 [99.5%]
+Walltime:8d3h6m (0s eval)
+ETA:57m46s
+Total train time:8d4h2m
+I1206 01:01:48.841167 137274321021824 utils.py:1231] [112100] l2_params = 237.99056374965463
+I1206 01:01:48.841382 137274321021824 utils.py:1231] [112100] train/loss = 2.2266745567321777
+I1206 01:01:48.841485 137274321021824 utils.py:1231] [112100] l2_grads = 2.7218663692474365
+I1206 01:01:48.841565 137274321021824 utils.py:1231] [112100] lr = 5.9534963222085965e-08
+I1206 01:01:48.841626 137274321021824 utils.py:1231] [112100] uptime = 702698.203987595
+I1206 01:01:48.841689 137274321021824 utils.py:1231] [112100] examples_seen = 114790400.0
+I1206 01:01:48.841748 137274321021824 utils.py:1231] [112100] progress = 0.9955329786950614
+I1206 01:01:48.841811 137274321021824 utils.py:1231] [112100] epoch = 89.59831153940118
+I1206 01:01:48.841869 137274321021824 utils.py:1231] [112100] img/sec/core = 164.21502836156674
+I1206 01:01:48.841943 137274321021824 utils.py:1231] [112100] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.15962794975306
+I1206 01:01:48.842001 137274321021824 utils.py:1231] [112100] core_hours = 195.15962794975306
+I1206 01:01:48.842069 137274321021824 train.py:125] NOTE: Steps:112100/112603 [99.6%]
+Walltime:8d3h11m (0s eval)
+ETA:52m33s
+Total train time:8d4h2m
+I1206 01:07:00.628327 137274321021824 utils.py:1231] [112150] l2_params = 237.9905683268717
+I1206 01:07:00.628612 137274321021824 utils.py:1231] [112150] train/loss = 2.077278107404709
+I1206 01:07:00.628788 137274321021824 utils.py:1231] [112150] l2_grads = 2.6919116973876953
+I1206 01:07:00.628902 137274321021824 utils.py:1231] [112150] lr = 4.8308588440904485e-08
+I1206 01:07:00.628968 137274321021824 utils.py:1231] [112150] uptime = 703009.991329222
+I1206 01:07:00.629033 137274321021824 utils.py:1231] [112150] examples_seen = 114841600.0
+I1206 01:07:00.629094 137274321021824 utils.py:1231] [112150] progress = 0.9959770165981369
+I1206 01:07:00.629153 137274321021824 utils.py:1231] [112150] epoch = 89.63827510387014
+I1206 01:07:00.629215 137274321021824 utils.py:1231] [112150] img/sec/core = 164.21449226524837
+I1206 01:07:00.629280 137274321021824 utils.py:1231] [112150] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.24623554464944
+I1206 01:07:00.629347 137274321021824 utils.py:1231] [112150] core_hours = 195.24623554464944
+I1206 01:07:00.629413 137274321021824 train.py:125] NOTE: Steps:112150/112603 [99.6%]
+Walltime:8d3h16m (0s eval)
+ETA:47m19s
+Total train time:8d4h2m
+I1206 01:12:12.402431 137274321021824 utils.py:1231] [112200] l2_params = 237.99057178132315
+I1206 01:12:12.402640 137274321021824 utils.py:1231] [112200] train/loss = 1.2658823728561401
+I1206 01:12:12.402733 137274321021824 utils.py:1231] [112200] l2_grads = 2.776930570602417
+I1206 01:12:12.402795 137274321021824 utils.py:1231] [112200] lr = 3.825399773810282e-08
+I1206 01:12:12.402846 137274321021824 utils.py:1231] [112200] uptime = 703321.765208228
+I1206 01:12:12.402906 137274321021824 utils.py:1231] [112200] examples_seen = 114892800.0
+I1206 01:12:12.402955 137274321021824 utils.py:1231] [112200] progress = 0.9964210545012122
+I1206 01:12:12.403003 137274321021824 utils.py:1231] [112200] epoch = 89.6782386683391
+I1206 01:12:12.403052 137274321021824 utils.py:1231] [112200] img/sec/core = 164.2215831654332
+I1206 01:12:12.403107 137274321021824 utils.py:1231] [112200] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.3328393999289
+I1206 01:12:12.403159 137274321021824 utils.py:1231] [112200] core_hours = 195.3328393999289
+I1206 01:12:12.403217 137274321021824 train.py:125] NOTE: Steps:112200/112603 [99.6%]
+Walltime:8d3h22m (0s eval)
+ETA:42m6s
+Total train time:8d4h2m
+I1206 01:17:24.182920 137274321021824 utils.py:1231] [112250] l2_params = 237.99057460677037
+I1206 01:17:24.183119 137274321021824 utils.py:1231] [112250] train/loss = 1.417180061340332
+I1206 01:17:24.183205 137274321021824 utils.py:1231] [112250] l2_grads = 2.9539742469787598
+I1206 01:17:24.183265 137274321021824 utils.py:1231] [112250] lr = 2.9371214679496948e-08
+I1206 01:17:24.183326 137274321021824 utils.py:1231] [112250] uptime = 703633.5456875709
+I1206 01:17:24.183376 137274321021824 utils.py:1231] [112250] examples_seen = 114944000.0
+I1206 01:17:24.183424 137274321021824 utils.py:1231] [112250] progress = 0.9968650924042877
+I1206 01:17:24.183471 137274321021824 utils.py:1231] [112250] epoch = 89.71820223280805
+I1206 01:17:24.183519 137274321021824 utils.py:1231] [112250] img/sec/core = 164.2181066239495
+I1206 01:17:24.183578 137274321021824 utils.py:1231] [112250] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.41944508863526
+I1206 01:17:24.183625 137274321021824 utils.py:1231] [112250] core_hours = 195.41944508863526
+I1206 01:17:24.183683 137274321021824 train.py:125] NOTE: Steps:112250/112603 [99.7%]
+Walltime:8d3h27m (0s eval)
+ETA:36m52s
+Total train time:8d4h2m
+I1206 01:22:36.020986 137274321021824 utils.py:1231] [112300] l2_params = 237.99057585751567
+I1206 01:22:36.021208 137274321021824 utils.py:1231] [112300] train/loss = 1.415870651602745
+I1206 01:22:36.021301 137274321021824 utils.py:1231] [112300] l2_grads = 2.8688912391662598
+I1206 01:22:36.021363 137274321021824 utils.py:1231] [112300] lr = 2.1660260084544132e-08
+I1206 01:22:36.021418 137274321021824 utils.py:1231] [112300] uptime = 703945.383779863
+I1206 01:22:36.021467 137274321021824 utils.py:1231] [112300] examples_seen = 114995200.0
+I1206 01:22:36.021514 137274321021824 utils.py:1231] [112300] progress = 0.997309130307363
+I1206 01:22:36.021562 137274321021824 utils.py:1231] [112300] epoch = 89.75816579727702
+I1206 01:22:36.021610 137274321021824 utils.py:1231] [112300] img/sec/core = 164.187766875012
+I1206 01:22:36.021664 137274321021824 utils.py:1231] [112300] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.5060667809386
+I1206 01:22:36.021711 137274321021824 utils.py:1231] [112300] core_hours = 195.5060667809386
+I1206 01:22:36.021770 137274321021824 train.py:125] NOTE: Steps:112300/112603 [99.7%]
+Walltime:8d3h32m (0s eval)
+ETA:31m39s
+Total train time:8d4h2m
+I1206 01:27:47.814995 137274321021824 utils.py:1231] [112350] l2_params = 237.99057767064556
+I1206 01:27:47.815307 137274321021824 utils.py:1231] [112350] train/loss = 2.521246761083603
+I1206 01:27:47.815486 137274321021824 utils.py:1231] [112350] l2_grads = 2.646419048309326
+I1206 01:27:47.815589 137274321021824 utils.py:1231] [112350] lr = 1.5121152026120887e-08
+I1206 01:27:47.815658 137274321021824 utils.py:1231] [112350] uptime = 704257.178019244
+I1206 01:27:47.815734 137274321021824 utils.py:1231] [112350] examples_seen = 115046400.0
+I1206 01:27:47.815793 137274321021824 utils.py:1231] [112350] progress = 0.9977531682104385
+I1206 01:27:47.815853 137274321021824 utils.py:1231] [112350] epoch = 89.79812936174596
+I1206 01:27:47.815920 137274321021824 utils.py:1231] [112350] img/sec/core = 164.21085938486218
+I1206 01:27:47.815980 137274321021824 utils.py:1231] [112350] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.59267629187778
+I1206 01:27:47.816035 137274321021824 utils.py:1231] [112350] core_hours = 195.59267629187778
+I1206 01:27:47.816101 137274321021824 train.py:125] NOTE: Steps:112350/112603 [99.8%]
+Walltime:8d3h37m (0s eval)
+ETA:26m26s
+Total train time:8d4h2m
+I1206 01:32:59.606136 137274321021824 utils.py:1231] [112400] l2_params = 237.9905794438287
+I1206 01:32:59.606394 137274321021824 utils.py:1231] [112400] train/loss = 3.7601708471775055
+I1206 01:32:59.606541 137274321021824 utils.py:1231] [112400] l2_grads = 2.9668304920196533
+I1206 01:32:59.606645 137274321021824 utils.py:1231] [112400] lr = 9.753905830634049e-09
+I1206 01:32:59.606717 137274321021824 utils.py:1231] [112400] uptime = 704568.9690786509
+I1206 01:32:59.606794 137274321021824 utils.py:1231] [112400] examples_seen = 115097600.0
+I1206 01:32:59.606851 137274321021824 utils.py:1231] [112400] progress = 0.9981972061135138
+I1206 01:32:59.606918 137274321021824 utils.py:1231] [112400] epoch = 89.83809292621493
+I1206 01:32:59.606976 137274321021824 utils.py:1231] [112400] img/sec/core = 164.2125341803832
+I1206 01:32:59.607039 137274321021824 utils.py:1231] [112400] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.6792849194908
+I1206 01:32:59.607095 137274321021824 utils.py:1231] [112400] core_hours = 195.6792849194908
+I1206 01:32:59.607162 137274321021824 train.py:125] NOTE: Steps:112400/112603 [99.8%]
+Walltime:8d3h42m (0s eval)
+ETA:21m12s
+Total train time:8d4h2m
+I1206 01:38:11.393474 137274321021824 utils.py:1231] [112450] l2_params = 237.99058028321053
+I1206 01:38:11.393763 137274321021824 utils.py:1231] [112450] train/loss = 2.0478030294179916
+I1206 01:38:11.393898 137274321021824 utils.py:1231] [112450] l2_grads = 2.7028870582580566
+I1206 01:38:11.393981 137274321021824 utils.py:1231] [112450] lr = 5.5585340776876375e-09
+I1206 01:38:11.394042 137274321021824 utils.py:1231] [112450] uptime = 704880.75640344
+I1206 01:38:11.394106 137274321021824 utils.py:1231] [112450] examples_seen = 115148800.0
+I1206 01:38:11.394166 137274321021824 utils.py:1231] [112450] progress = 0.9986412440165893
+I1206 01:38:11.394224 137274321021824 utils.py:1231] [112450] epoch = 89.87805649068389
+I1206 01:38:11.394301 137274321021824 utils.py:1231] [112450] img/sec/core = 164.21450113353197
+I1206 01:38:11.394406 137274321021824 utils.py:1231] [112450] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.76589250971
+I1206 01:38:11.394487 137274321021824 utils.py:1231] [112450] core_hours = 195.76589250971
+I1206 01:38:11.394589 137274321021824 train.py:125] NOTE: Steps:112450/112603 [99.9%]
+Walltime:8d3h48m (0s eval)
+ETA:15m59s
+Total train time:8d4h2m
+I1206 01:43:23.192085 137274321021824 utils.py:1231] [112500] l2_params = 237.9905802487737
+I1206 01:43:23.192337 137274321021824 utils.py:1231] [112500] train/loss = 2.3289879262447357
+I1206 01:43:23.192563 137274321021824 utils.py:1231] [112500] l2_grads = 2.7661311626434326
+I1206 01:43:23.192655 137274321021824 utils.py:1231] [112500] lr = 2.5350466004714607e-09
+I1206 01:43:23.192735 137274321021824 utils.py:1231] [112500] uptime = 705192.55508918
+I1206 01:43:23.192824 137274321021824 utils.py:1231] [112500] examples_seen = 115200000.0
+I1206 01:43:23.192903 137274321021824 utils.py:1231] [112500] progress = 0.9990852819196646
+I1206 01:43:23.192989 137274321021824 utils.py:1231] [112500] epoch = 89.91802005515284
+I1206 01:43:23.193076 137274321021824 utils.py:1231] [112500] img/sec/core = 164.20851768021026
+I1206 01:43:23.193142 137274321021824 utils.py:1231] [112500] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.8525032557489
+I1206 01:43:23.193216 137274321021824 utils.py:1231] [112500] core_hours = 195.8525032557489
+I1206 01:43:23.193286 137274321021824 train.py:125] NOTE: Steps:112500/112603 [99.9%]
+Walltime:8d3h53m (0s eval)
+ETA:10m46s
+Total train time:8d4h2m
+I1206 01:48:34.967616 137274321021824 utils.py:1231] [112550] l2_params = 237.99057967173783
+I1206 01:48:34.967836 137274321021824 utils.py:1231] [112550] train/loss = 1.4715039879083633
+I1206 01:48:34.967957 137274321021824 utils.py:1231] [112550] l2_grads = 2.7689104080200195
+I1206 01:48:34.968030 137274321021824 utils.py:1231] [112550] lr = 6.834504853170453e-10
+I1206 01:48:34.968096 137274321021824 utils.py:1231] [112550] uptime = 705504.330457406
+I1206 01:48:34.968152 137274321021824 utils.py:1231] [112550] examples_seen = 115251200.0
+I1206 01:48:34.968205 137274321021824 utils.py:1231] [112550] progress = 0.9995293198227401
+I1206 01:48:34.968265 137274321021824 utils.py:1231] [112550] epoch = 89.9579836196218
+I1206 01:48:34.968321 137274321021824 utils.py:1231] [112550] img/sec/core = 164.22079874792146
+I1206 01:48:34.968388 137274321021824 utils.py:1231] [112550] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 195.93910752470055
+I1206 01:48:34.968443 137274321021824 utils.py:1231] [112550] core_hours = 195.93910752470055
+I1206 01:48:34.968507 137274321021824 train.py:125] NOTE: Steps:112550/112603 [100.0%]
+Walltime:8d3h58m (0s eval)
+ETA:5m32s
+Total train time:8d4h2m
+I1206 01:54:05.457205 137274321021824 utils.py:1231] [112603] l2_params = 237.9905797356025
+I1206 01:54:05.457465 137274321021824 utils.py:1231] [112603] train/loss = 1.7738897949457169
+I1206 01:54:05.457564 137274321021824 utils.py:1231] [112603] l2_grads = 2.6440370082855225
+I1206 01:54:05.457630 137274321021824 utils.py:1231] [112603] lr = 2.3437951579552636e-13
+I1206 01:54:05.457695 137274321021824 utils.py:1231] [112603] uptime = 705834.820052186
+I1206 01:54:05.457767 137274321021824 utils.py:1231] [112603] examples_seen = 115305472.0
+I1206 01:54:05.457814 137274321021824 utils.py:1231] [112603] progress = 1.0
+I1206 01:54:05.457864 137274321021824 utils.py:1231] [112603] epoch = 90.00034499795889
+I1206 01:54:05.457921 137274321021824 utils.py:1231] [112603] img/sec/core = 164.21697038941156
+I1206 01:54:05.457978 137274321021824 utils.py:1231] [112603] core_hours_NVIDIA GeForce RTX 3080 Laptop GPU = 196.03091018991725
+I1206 01:54:05.458026 137274321021824 utils.py:1231] [112603] core_hours = 196.03091018991725
+I1206 01:54:05.458083 137274321021824 train.py:125] NOTE: Steps:112603/112603 [100.0%]
+Walltime:8d4h3m (0s eval)
+ETA:0s
+Total train time:8d4h2m
+I1206 01:54:05.797180 137274321021824 train.py:125] NOTE: val evaluation...
+Steps:112603/112603 [100.0%]
+Walltime:8d4h3m (0s eval)
+ETA:0s
+Total train time:8d4h2m
+I1206 01:55:42.872052 137274321021824 utils.py:1231] [112603] val/acc@1 = 0.7654257015306123
+I1206 01:55:42.872307 137274321021824 utils.py:1231] [112603] val/loss = 0.919307008081553
+I1206 01:55:42.872491 137274321021824 utils.py:1231] [112603] z/secs/eval/val = 97.07507979194634
+I1206 01:55:42.872565 137274321021824 utils.py:560] TIMING[z/secs/eval/val]: 97.07507979194634
+I1206 01:55:42.873148 137274321021824 train.py:125] NOTE: Done!
+Steps:112603/112603 [100.0%]
+Walltime:8d4h3m (0s eval)
+ETA:0s
+Total train time:8d4h2m
+I1206 01:55:43.018081 137274321021824 train.py:125] NOTE: Logs/checkpoints are in /data/imagenet/grafted