End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: Minbyul/biomistral-7b-wo-kqa_golden-iter-dpo-step2
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: biomistral-7b-wo-kqa_golden-iter-dpo-step3
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # biomistral-7b-wo-kqa_golden-iter-dpo-step3
 
-This model is a fine-tuned version of [Minbyul/biomistral-7b-wo-kqa_golden-iter-dpo-step2](https://huggingface.co/Minbyul/biomistral-7b-wo-kqa_golden-iter-dpo-step2) on an unknown dataset.
+This model is a fine-tuned version of [Minbyul/biomistral-7b-wo-kqa_golden-iter-dpo-step2](https://huggingface.co/Minbyul/biomistral-7b-wo-kqa_golden-iter-dpo-step2) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Logits/chosen: -0.8589
-- Logits/rejected: -1.5130
-- Logps/chosen: -39.0432
-- Logps/rejected: -58.7210
-- Loss: 0.6837
-- Rewards/accuracies: 0.6400
-- Rewards/chosen: -0.0394
-- Rewards/margins: 0.0267
-- Rewards/rejected: -0.0661
+- Loss: 0.6829
+- Rewards/chosen: -0.0502
+- Rewards/rejected: -0.0800
+- Rewards/accuracies: 0.6300
+- Rewards/margins: 0.0298
+- Logps/rejected: -60.1185
+- Logps/chosen: -40.1264
+- Logits/rejected: -1.5228
+- Logits/chosen: -0.8710
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -0.8710112571716309,
+    "eval_logits/rejected": -1.522813320159912,
+    "eval_logps/chosen": -40.126407623291016,
+    "eval_logps/rejected": -60.11854934692383,
+    "eval_loss": 0.6829047799110413,
+    "eval_rewards/accuracies": 0.6299999952316284,
+    "eval_rewards/chosen": -0.050214797258377075,
+    "eval_rewards/margins": 0.02983289770781994,
+    "eval_rewards/rejected": -0.08004770427942276,
+    "eval_runtime": 65.4744,
+    "eval_samples": 785,
+    "eval_samples_per_second": 11.989,
+    "eval_steps_per_second": 0.382,
     "train_loss": 0.1692200431117305,
     "train_runtime": 798.9647,
     "train_samples": 17264,

config.json

@@ -21,6 +21,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -0.8710112571716309,
+    "eval_logits/rejected": -1.522813320159912,
+    "eval_logps/chosen": -40.126407623291016,
+    "eval_logps/rejected": -60.11854934692383,
+    "eval_loss": 0.6829047799110413,
+    "eval_rewards/accuracies": 0.6299999952316284,
+    "eval_rewards/chosen": -0.050214797258377075,
+    "eval_rewards/margins": 0.02983289770781994,
+    "eval_rewards/rejected": -0.08004770427942276,
+    "eval_runtime": 65.4744,
+    "eval_samples": 785,
+    "eval_samples_per_second": 11.989,
+    "eval_steps_per_second": 0.382
+}