End of training

.gitattributes

@@ -37,3 +37,4 @@ tokenizer.json filter=lfs diff=lfs merge=lfs -text
 dynamics.cache/cache.db-wal filter=lfs diff=lfs merge=lfs -text
 dynamics/cache.db-wal filter=lfs diff=lfs merge=lfs -text
 dynamics.cache/cache.db filter=lfs diff=lfs merge=lfs -text
+dynamics/cache.db filter=lfs diff=lfs merge=lfs -text

README.md

@@ -3,20 +3,32 @@ library_name: transformers
 license: other
 base_model: trl-lib/qwen1.5-0.5b-sft
 tags:
+- alignment-handbook
 - trl
 - simpo
 - generated_from_trainer
 model-index:
-- name: simpo-exps_qwen05b
+- name: trl-lib/qwen1.5-0.5b-sft
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# simpo-exps_qwen05b
+# trl-lib/qwen1.5-0.5b-sft
 
 This model is a fine-tuned version of [trl-lib/qwen1.5-0.5b-sft](https://huggingface.co/trl-lib/qwen1.5-0.5b-sft) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7797
+- Rewards/chosen: -1.7866
+- Rewards/rejected: -1.8726
+- Rewards/accuracies: 0.5230
+- Rewards/margins: 0.0860
+- Logps/rejected: -1.8726
+- Logps/chosen: -1.7866
+- Logits/rejected: -0.3259
+- Logits/chosen: -0.3565
+- Semantic Entropy: 0.6461
 
 ## Model description
 

all_results.json

@@ -1,5 +1,19 @@
 {
     "epoch": 32.0,
+    "eval_logits/chosen": -0.3564789891242981,
+    "eval_logits/rejected": -0.32587698101997375,
+    "eval_logps/chosen": -1.786582350730896,
+    "eval_logps/rejected": -1.8725693225860596,
+    "eval_loss": 0.7796740531921387,
+    "eval_rewards/accuracies": 0.5229970216751099,
+    "eval_rewards/chosen": -1.786582350730896,
+    "eval_rewards/margins": 0.08598694205284119,
+    "eval_rewards/rejected": -1.8725693225860596,
+    "eval_runtime": 60.635,
+    "eval_samples": 1345,
+    "eval_samples_per_second": 22.182,
+    "eval_semantic_entropy": 0.6461161971092224,
+    "eval_steps_per_second": 5.558,
     "total_flos": 0.0,
     "train_loss": 0.6880494674046834,
     "train_runtime": 138.3641,

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151648
 }

dynamics/cache.db-wal

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52f377c3b8890e2b6717913111ea59cb4b4c2c86572b68a8487db466685b655e
-size 4128272
+oid sha256:4cf665c94502bbd6a3cc1074a3a3fd683fc5ae04e600725485ca9e82bb1f9e99
+size 4136512

eval_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 32.0,
-    "eval_logits/chosen": -0.3597286641597748,
-    "eval_logits/rejected": -0.32949814200401306,
-    "eval_logps/chosen": -1.787438988685608,
-    "eval_logps/rejected": -1.8736913204193115,
-    "eval_loss": 0.7798751592636108,
-    "eval_rewards/accuracies": 0.5215133428573608,
-    "eval_rewards/chosen": -1.787438988685608,
-    "eval_rewards/margins": 0.08625216037034988,
-    "eval_rewards/rejected": -1.8736913204193115,
-    "eval_runtime": 60.3848,
+    "eval_logits/chosen": -0.3564789891242981,
+    "eval_logits/rejected": -0.32587698101997375,
+    "eval_logps/chosen": -1.786582350730896,
+    "eval_logps/rejected": -1.8725693225860596,
+    "eval_loss": 0.7796740531921387,
+    "eval_rewards/accuracies": 0.5229970216751099,
+    "eval_rewards/chosen": -1.786582350730896,
+    "eval_rewards/margins": 0.08598694205284119,
+    "eval_rewards/rejected": -1.8725693225860596,
+    "eval_runtime": 60.635,
     "eval_samples": 1345,
-    "eval_samples_per_second": 22.274,
-    "eval_semantic_entropy": 0.6458982825279236,
-    "eval_steps_per_second": 5.581
+    "eval_samples_per_second": 22.182,
+    "eval_semantic_entropy": 0.6461161971092224,
+    "eval_steps_per_second": 5.558
 }