End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: Minbyul/mistral-7b-wo-kqa_golden-sft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: mistral-7b-dpo-full-sft-wo-kqa_golden
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-dpo-full-sft-wo-kqa_golden
 
-This model is a fine-tuned version of [Minbyul/mistral-7b-wo-kqa_golden-sft](https://huggingface.co/Minbyul/mistral-7b-wo-kqa_golden-sft) on an unknown dataset.
+This model is a fine-tuned version of [Minbyul/mistral-7b-wo-kqa_golden-sft](https://huggingface.co/Minbyul/mistral-7b-wo-kqa_golden-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0019
-- Rewards/chosen: -0.4479
-- Rewards/rejected: -10.0624
+- Loss: 0.0018
+- Rewards/chosen: -0.4458
+- Rewards/rejected: -10.1099
 - Rewards/accuracies: 1.0
-- Rewards/margins: 9.6145
-- Logps/rejected: -1559.6263
-- Logps/chosen: -241.4165
-- Logits/rejected: -2.0553
-- Logits/chosen: -1.3416
+- Rewards/margins: 9.6641
+- Logps/rejected: -1564.3792
+- Logps/chosen: -241.2112
+- Logits/rejected: -2.0516
+- Logits/chosen: -1.3414
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -1.34141206741333,
+    "eval_logits/rejected": -2.0515904426574707,
+    "eval_logps/chosen": -241.21115112304688,
+    "eval_logps/rejected": -1564.379150390625,
+    "eval_loss": 0.0018388006137683988,
+    "eval_rewards/accuracies": 1.0,
+    "eval_rewards/chosen": -0.4458274245262146,
+    "eval_rewards/margins": 9.664073944091797,
+    "eval_rewards/rejected": -10.109901428222656,
+    "eval_runtime": 6.9758,
+    "eval_samples": 64,
+    "eval_samples_per_second": 9.175,
+    "eval_steps_per_second": 0.287,
     "train_loss": 0.23720380818807385,
     "train_runtime": 5120.8034,
     "train_samples": 20928,

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": -1.34141206741333,
+    "eval_logits/rejected": -2.0515904426574707,
+    "eval_logps/chosen": -241.21115112304688,
+    "eval_logps/rejected": -1564.379150390625,
+    "eval_loss": 0.0018388006137683988,
+    "eval_rewards/accuracies": 1.0,
+    "eval_rewards/chosen": -0.4458274245262146,
+    "eval_rewards/margins": 9.664073944091797,
+    "eval_rewards/rejected": -10.109901428222656,
+    "eval_runtime": 6.9758,
+    "eval_samples": 64,
+    "eval_samples_per_second": 9.175,
+    "eval_steps_per_second": 0.287
+}