End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: Minbyul/mistral-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: mistral-7b-dpo-full-sft-wo-kqa_silver_wogold
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-dpo-full-sft-wo-kqa_silver_wogold
 
-This model is a fine-tuned version of [Minbyul/mistral-7b-wo-kqa_silver_wogold-sft](https://huggingface.co/Minbyul/mistral-7b-wo-kqa_silver_wogold-sft) on an unknown dataset.
+This model is a fine-tuned version of [Minbyul/mistral-7b-wo-kqa_silver_wogold-sft](https://huggingface.co/Minbyul/mistral-7b-wo-kqa_silver_wogold-sft) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0531
-- Rewards/chosen: -2.4855
-- Rewards/rejected: -21.1345
+- Loss: 0.0530
+- Rewards/chosen: -2.4760
+- Rewards/rejected: -21.0723
 - Rewards/accuracies: 0.9700
-- Rewards/margins: 18.6490
-- Logps/rejected: -2715.4331
-- Logps/chosen: -408.6504
-- Logits/rejected: -2.0210
-- Logits/chosen: -2.2273
+- Rewards/margins: 18.5963
+- Logps/rejected: -2709.2131
+- Logps/chosen: -407.7003
+- Logits/rejected: -2.0225
+- Logits/chosen: -2.2276
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.227571487426758,
+    "eval_logits/rejected": -2.0224978923797607,
+    "eval_logps/chosen": -407.7003173828125,
+    "eval_logps/rejected": -2709.213134765625,
+    "eval_loss": 0.0529978983104229,
+    "eval_rewards/accuracies": 0.9700000286102295,
+    "eval_rewards/chosen": -2.475966691970825,
+    "eval_rewards/margins": 18.596342086791992,
+    "eval_rewards/rejected": -21.072311401367188,
+    "eval_runtime": 86.3233,
+    "eval_samples": 788,
+    "eval_samples_per_second": 9.128,
+    "eval_steps_per_second": 0.29,
     "train_loss": 0.24172405860846555,
     "train_runtime": 5168.5497,
     "train_samples": 20204,

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": -2.227571487426758,
+    "eval_logits/rejected": -2.0224978923797607,
+    "eval_logps/chosen": -407.7003173828125,
+    "eval_logps/rejected": -2709.213134765625,
+    "eval_loss": 0.0529978983104229,
+    "eval_rewards/accuracies": 0.9700000286102295,
+    "eval_rewards/chosen": -2.475966691970825,
+    "eval_rewards/margins": 18.596342086791992,
+    "eval_rewards/rejected": -21.072311401367188,
+    "eval_runtime": 86.3233,
+    "eval_samples": 788,
+    "eval_samples_per_second": 9.128,
+    "eval_steps_per_second": 0.29
+}