End of training

README.md

@@ -2,9 +2,10 @@
 license: llama2
 library_name: peft
 tags:
+- llama-factory
+- lora
 - trl
 - dpo
-- llama-factory
 - generated_from_trainer
 base_model: lmsys/vicuna-7b-v1.5
 model-index:
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Vicuna-7B-v1.5-ORPO
 
-This model is a fine-tuned version of [lmsys/vicuna-7b-v1.5](https://huggingface.co/lmsys/vicuna-7b-v1.5) on an unknown dataset.
+This model is a fine-tuned version of [lmsys/vicuna-7b-v1.5](https://huggingface.co/lmsys/vicuna-7b-v1.5) on the dpo_mix_en dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0073
 - Rewards/chosen: -0.0940

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 2.997999555456768,
+    "eval_logits/chosen": -0.3320940136909485,
+    "eval_logits/rejected": -0.29884636402130127,
+    "eval_logps/chosen": -0.9399133324623108,
+    "eval_logps/rejected": -1.080655574798584,
+    "eval_loss": 1.0073015689849854,
+    "eval_odds_ratio_loss": 0.6738813519477844,
+    "eval_rewards/accuracies": 0.515999972820282,
+    "eval_rewards/chosen": -0.09399133920669556,
+    "eval_rewards/margins": 0.01407422125339508,
+    "eval_rewards/rejected": -0.10806556046009064,
+    "eval_runtime": 185.8105,
+    "eval_samples_per_second": 5.382,
+    "eval_sft_loss": 0.9399133324623108,
+    "eval_steps_per_second": 2.691,
+    "total_flos": 1.8817568285770383e+18,
+    "train_loss": 1.0353579054523618,
+    "train_runtime": 16950.0138,
+    "train_samples_per_second": 1.593,
+    "train_steps_per_second": 0.099
+}

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 2.997999555456768,
+    "eval_logits/chosen": -0.3320940136909485,
+    "eval_logits/rejected": -0.29884636402130127,
+    "eval_logps/chosen": -0.9399133324623108,
+    "eval_logps/rejected": -1.080655574798584,
+    "eval_loss": 1.0073015689849854,
+    "eval_odds_ratio_loss": 0.6738813519477844,
+    "eval_rewards/accuracies": 0.515999972820282,
+    "eval_rewards/chosen": -0.09399133920669556,
+    "eval_rewards/margins": 0.01407422125339508,
+    "eval_rewards/rejected": -0.10806556046009064,
+    "eval_runtime": 185.8105,
+    "eval_samples_per_second": 5.382,
+    "eval_sft_loss": 0.9399133324623108,
+    "eval_steps_per_second": 2.691
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.997999555456768,
+    "total_flos": 1.8817568285770383e+18,
+    "train_loss": 1.0353579054523618,
+    "train_runtime": 16950.0138,
+    "train_samples_per_second": 1.593,
+    "train_steps_per_second": 0.099
+}