End of training

README.md

@@ -2,9 +2,15 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: mistral-7b-dpo-full-wo-kqa_golden-ep3
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-dpo-full-wo-kqa_golden-ep3
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2503
-- Rewards/chosen: -1.6026
-- Rewards/rejected: -5.5026
-- Rewards/accuracies: 0.8313
-- Rewards/margins: 3.9001
-- Logps/rejected: -1452.9772
-- Logps/chosen: -725.9427
-- Logits/rejected: -2.7934
-- Logits/chosen: -3.0544
+- Loss: 0.2182
+- Rewards/chosen: -1.5510
+- Rewards/rejected: -6.3512
+- Rewards/accuracies: 0.875
+- Rewards/margins: 4.8002
+- Logps/rejected: -1537.8291
+- Logps/chosen: -720.7828
+- Logits/rejected: -2.8834
+- Logits/chosen: -3.1394
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -3.13935923576355,
+    "eval_logits/rejected": -2.883375644683838,
+    "eval_logps/chosen": -720.7828369140625,
+    "eval_logps/rejected": -1537.8291015625,
+    "eval_loss": 0.21815194189548492,
+    "eval_rewards/accuracies": 0.875,
+    "eval_rewards/chosen": -1.5509830713272095,
+    "eval_rewards/margins": 4.800168991088867,
+    "eval_rewards/rejected": -6.351152420043945,
+    "eval_runtime": 66.6237,
+    "eval_samples": 616,
+    "eval_samples_per_second": 9.246,
+    "eval_steps_per_second": 0.3,
     "train_loss": 0.2699868052719749,
     "train_runtime": 2833.2764,
     "train_samples": 11996,

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": -3.13935923576355,
+    "eval_logits/rejected": -2.883375644683838,
+    "eval_logps/chosen": -720.7828369140625,
+    "eval_logps/rejected": -1537.8291015625,
+    "eval_loss": 0.21815194189548492,
+    "eval_rewards/accuracies": 0.875,
+    "eval_rewards/chosen": -1.5509830713272095,
+    "eval_rewards/margins": 4.800168991088867,
+    "eval_rewards/rejected": -6.351152420043945,
+    "eval_runtime": 66.6237,
+    "eval_samples": 616,
+    "eval_samples_per_second": 9.246,
+    "eval_steps_per_second": 0.3
+}