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-medication_qa-ep3
   results: []
@@ -15,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-dpo-full-wo-medication_qa-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:
-- Logits/chosen: -3.1281
-- Logits/rejected: -2.9646
-- Logps/chosen: -755.8733
-- Logps/rejected: -1595.0167
-- Loss: 0.3332
-- Rewards/accuracies: 0.8686
-- Rewards/chosen: -1.8907
-- Rewards/margins: 3.9363
-- Rewards/rejected: -5.8271
+- Loss: 0.3800
+- Rewards/chosen: -2.4212
+- Rewards/rejected: -6.9305
+- Rewards/accuracies: 0.8602
+- Rewards/margins: 4.5093
+- Logps/rejected: -1705.3644
+- Logps/chosen: -808.9231
+- Logits/rejected: -2.9073
+- Logits/chosen: -3.0809
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -3.0808565616607666,
+    "eval_logits/rejected": -2.90731143951416,
+    "eval_logps/chosen": -808.923095703125,
+    "eval_logps/rejected": -1705.3643798828125,
+    "eval_loss": 0.3800393342971802,
+    "eval_rewards/accuracies": 0.8601694703102112,
+    "eval_rewards/chosen": -2.4212381839752197,
+    "eval_rewards/margins": 4.509303569793701,
+    "eval_rewards/rejected": -6.9305419921875,
+    "eval_runtime": 193.7756,
+    "eval_samples": 1868,
+    "eval_samples_per_second": 9.64,
+    "eval_steps_per_second": 0.304,
     "train_loss": 0.05682134947606495,
     "train_runtime": 921.1998,
     "train_samples": 10744,

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": -3.0808565616607666,
+    "eval_logits/rejected": -2.90731143951416,
+    "eval_logps/chosen": -808.923095703125,
+    "eval_logps/rejected": -1705.3643798828125,
+    "eval_loss": 0.3800393342971802,
+    "eval_rewards/accuracies": 0.8601694703102112,
+    "eval_rewards/chosen": -2.4212381839752197,
+    "eval_rewards/margins": 4.509303569793701,
+    "eval_rewards/rejected": -6.9305419921875,
+    "eval_runtime": 193.7756,
+    "eval_samples": 1868,
+    "eval_samples_per_second": 9.64,
+    "eval_steps_per_second": 0.304
+}