End of training

README.md

@@ -1,9 +1,15 @@
 ---
 base_model: Minbyul/selfbiorag-7b-wo-live_qa-sft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: selfbiorag-7b-dpo-full-sft-wo-live_qa
   results: []
@@ -14,17 +20,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # selfbiorag-7b-dpo-full-sft-wo-live_qa
 
-This model is a fine-tuned version of [Minbyul/selfbiorag-7b-wo-live_qa-sft](https://huggingface.co/Minbyul/selfbiorag-7b-wo-live_qa-sft) on an unknown dataset.
+This model is a fine-tuned version of [Minbyul/selfbiorag-7b-wo-live_qa-sft](https://huggingface.co/Minbyul/selfbiorag-7b-wo-live_qa-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1421
-- Rewards/chosen: -1.2604
-- Rewards/rejected: -13.2286
+- Loss: 0.1422
+- Rewards/chosen: -1.2709
+- Rewards/rejected: -13.2633
 - Rewards/accuracies: 0.9167
-- Rewards/margins: 11.9683
-- Logps/rejected: -1987.8828
-- Logps/chosen: -455.8129
-- Logits/rejected: -0.4048
-- Logits/chosen: -0.4887
+- Rewards/margins: 11.9924
+- Logps/rejected: -1991.3534
+- Logps/chosen: -456.8682
+- Logits/rejected: -0.4049
+- Logits/chosen: -0.4878
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -0.4878213703632355,
+    "eval_logits/rejected": -0.4049469530582428,
+    "eval_logps/chosen": -456.8681945800781,
+    "eval_logps/rejected": -1991.3533935546875,
+    "eval_loss": 0.14216427505016327,
+    "eval_rewards/accuracies": 0.9166666865348816,
+    "eval_rewards/chosen": -1.2709063291549683,
+    "eval_rewards/margins": 11.992403984069824,
+    "eval_rewards/rejected": -13.263309478759766,
+    "eval_runtime": 29.9163,
+    "eval_samples": 192,
+    "eval_samples_per_second": 6.418,
+    "eval_steps_per_second": 0.201,
     "train_loss": 0.22599055091055428,
     "train_runtime": 7017.6011,
     "train_samples": 20961,

config.json

@@ -23,6 +23,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32016
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -0.4878213703632355,
+    "eval_logits/rejected": -0.4049469530582428,
+    "eval_logps/chosen": -456.8681945800781,
+    "eval_logps/rejected": -1991.3533935546875,
+    "eval_loss": 0.14216427505016327,
+    "eval_rewards/accuracies": 0.9166666865348816,
+    "eval_rewards/chosen": -1.2709063291549683,
+    "eval_rewards/margins": 11.992403984069824,
+    "eval_rewards/rejected": -13.263309478759766,
+    "eval_runtime": 29.9163,
+    "eval_samples": 192,
+    "eval_samples_per_second": 6.418,
+    "eval_steps_per_second": 0.201
+}