| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 222, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 7.765780925750732, | |
| "learning_rate": 0.0002, | |
| "loss": 2.477, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 2.2755630016326904, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4771, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 1.7242066860198975, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2418, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 95.61922454833984, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1066, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33783783783783783, | |
| "grad_norm": 1.10884690284729, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0927, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 0.9811855554580688, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0442, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47297297297297297, | |
| "grad_norm": 0.874477744102478, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0362, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 0.9105525612831116, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0281, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 0.7913025617599487, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9934, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 0.8067978024482727, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9768, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7432432432432432, | |
| "grad_norm": 0.8261044025421143, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9662, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 0.8108837604522705, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9668, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8783783783783784, | |
| "grad_norm": 0.7348493933677673, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9431, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 0.7668595314025879, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9235, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 0.7864215970039368, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8857, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 0.754004180431366, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7299, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 0.6962944865226746, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7184, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 0.969588041305542, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7025, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2837837837837838, | |
| "grad_norm": 0.6978746652603149, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7014, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 0.7287507653236389, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6986, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 0.7103177309036255, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7347, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 0.6908261775970459, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7083, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.554054054054054, | |
| "grad_norm": 0.7442795634269714, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6938, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 0.6897634267807007, | |
| "learning_rate": 0.0002, | |
| "loss": 0.73, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 0.6686336994171143, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7234, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 0.6796671748161316, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7173, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 0.6994472146034241, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6866, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 0.6973745226860046, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7225, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.9594594594594594, | |
| "grad_norm": 0.6692176461219788, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7032, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 0.5986601114273071, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6413, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 0.6870444416999817, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5096, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 0.6344185471534729, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5231, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 0.6490407586097717, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5095, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 0.7220724821090698, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5243, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.364864864864865, | |
| "grad_norm": 0.6581041812896729, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5064, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 0.6284764409065247, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5093, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.7284560203552246, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5312, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 0.723694920539856, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5565, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 0.6478952169418335, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5273, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 0.6292466521263123, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5167, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.77027027027027, | |
| "grad_norm": 0.6427861452102661, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5256, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 0.6279394626617432, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5138, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.9054054054054053, | |
| "grad_norm": 0.6324751973152161, | |
| "learning_rate": 0.0002, | |
| "loss": 0.527, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 0.6362230181694031, | |
| "learning_rate": 0.0002, | |
| "loss": 0.524, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 222, | |
| "total_flos": 1.7527093219609805e+17, | |
| "train_loss": 0.7947500686387758, | |
| "train_runtime": 3770.1678, | |
| "train_samples_per_second": 0.937, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 222, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7527093219609805e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |