End of training

README.md

@@ -2,9 +2,15 @@
 license: llama2
 base_model: Minbyul/meditron-7b-wo-kqa_silver_wogold-sft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: meditron-7b-dpo-full-sft-wo-kqa_silver_wogold
   results: []
@@ -15,7 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # meditron-7b-dpo-full-sft-wo-kqa_silver_wogold
 
-This model is a fine-tuned version of [Minbyul/meditron-7b-wo-kqa_silver_wogold-sft](https://huggingface.co/Minbyul/meditron-7b-wo-kqa_silver_wogold-sft) on an unknown dataset.
+This model is a fine-tuned version of [Minbyul/meditron-7b-wo-kqa_silver_wogold-sft](https://huggingface.co/Minbyul/meditron-7b-wo-kqa_silver_wogold-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6249
+- Rewards/chosen: -0.0145
+- Rewards/rejected: -0.1983
+- Rewards/accuracies: 0.875
+- Rewards/margins: 0.1839
+- Logps/rejected: -653.1394
+- Logps/chosen: -150.2214
+- Logits/rejected: -1.0337
+- Logits/chosen: -1.4037
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -1.4037474393844604,
+    "eval_logits/rejected": -1.0336650609970093,
+    "eval_logps/chosen": -150.22137451171875,
+    "eval_logps/rejected": -653.139404296875,
+    "eval_loss": 0.6249053478240967,
+    "eval_rewards/accuracies": 0.875,
+    "eval_rewards/chosen": -0.014455685392022133,
+    "eval_rewards/margins": 0.18389175832271576,
+    "eval_rewards/rejected": -0.19834741950035095,
+    "eval_runtime": 73.8566,
+    "eval_samples": 704,
+    "eval_samples_per_second": 9.532,
+    "eval_steps_per_second": 0.298,
     "train_loss": 0.5552032154340011,
     "train_runtime": 696.2336,
     "train_samples": 3326,

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": 32017
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -1.4037474393844604,
+    "eval_logits/rejected": -1.0336650609970093,
+    "eval_logps/chosen": -150.22137451171875,
+    "eval_logps/rejected": -653.139404296875,
+    "eval_loss": 0.6249053478240967,
+    "eval_rewards/accuracies": 0.875,
+    "eval_rewards/chosen": -0.014455685392022133,
+    "eval_rewards/margins": 0.18389175832271576,
+    "eval_rewards/rejected": -0.19834741950035095,
+    "eval_runtime": 73.8566,
+    "eval_samples": 704,
+    "eval_samples_per_second": 9.532,
+    "eval_steps_per_second": 0.298
+}