End of training

README.md

@@ -2,9 +2,10 @@
 license: apache-2.0
 library_name: peft
 tags:
+- llama-factory
+- lora
 - trl
 - dpo
-- llama-factory
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-Instruct-v0.3
 model-index:
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral-7B-Instruct-v0.3-ORPO-SALT
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the None dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the dpo_mix_en and the bct_non_cot_dpo_1000 datasets.
 It achieves the following results on the evaluation set:
 - Loss: 0.8434
 - Rewards/chosen: -0.0777

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 2.9969690846635686,
+    "eval_logits/chosen": -3.083430051803589,
+    "eval_logits/rejected": -3.107269287109375,
+    "eval_logps/chosen": -0.777300238609314,
+    "eval_logps/rejected": -0.9876723289489746,
+    "eval_loss": 0.8434417247772217,
+    "eval_odds_ratio_loss": 0.6614136099815369,
+    "eval_rewards/accuracies": 0.5690909028053284,
+    "eval_rewards/chosen": -0.07773003727197647,
+    "eval_rewards/margins": 0.021037202328443527,
+    "eval_rewards/rejected": -0.0987672358751297,
+    "eval_runtime": 367.4747,
+    "eval_samples_per_second": 2.993,
+    "eval_sft_loss": 0.777300238609314,
+    "eval_steps_per_second": 1.497,
+    "total_flos": 2.0970902870084813e+18,
+    "train_loss": 0.8330582246554065,
+    "train_runtime": 34111.2463,
+    "train_samples_per_second": 0.871,
+    "train_steps_per_second": 0.054
+}

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 2.9969690846635686,
+    "eval_logits/chosen": -3.083430051803589,
+    "eval_logits/rejected": -3.107269287109375,
+    "eval_logps/chosen": -0.777300238609314,
+    "eval_logps/rejected": -0.9876723289489746,
+    "eval_loss": 0.8434417247772217,
+    "eval_odds_ratio_loss": 0.6614136099815369,
+    "eval_rewards/accuracies": 0.5690909028053284,
+    "eval_rewards/chosen": -0.07773003727197647,
+    "eval_rewards/margins": 0.021037202328443527,
+    "eval_rewards/rejected": -0.0987672358751297,
+    "eval_runtime": 367.4747,
+    "eval_samples_per_second": 2.993,
+    "eval_sft_loss": 0.777300238609314,
+    "eval_steps_per_second": 1.497
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9969690846635686,
+    "total_flos": 2.0970902870084813e+18,
+    "train_loss": 0.8330582246554065,
+    "train_runtime": 34111.2463,
+    "train_samples_per_second": 0.871,
+    "train_steps_per_second": 0.054
+}