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: tiiuae/falcon-7b-instruct
 model-index:
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Falcon-7B-Instruct-ORPO
 
-This model is a fine-tuned version of [tiiuae/falcon-7b-instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) on an unknown dataset.
+This model is a fine-tuned version of [tiiuae/falcon-7b-instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) on the dpo_mix_en dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.5155
 - Rewards/chosen: -0.1444

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 2.997999555456768,
+    "eval_logits/chosen": -14.466456413269043,
+    "eval_logits/rejected": -14.543242454528809,
+    "eval_logps/chosen": -1.44402277469635,
+    "eval_logps/rejected": -1.5389362573623657,
+    "eval_loss": 1.5154520273208618,
+    "eval_odds_ratio_loss": 0.7142924666404724,
+    "eval_rewards/accuracies": 0.5090000033378601,
+    "eval_rewards/chosen": -0.14440228044986725,
+    "eval_rewards/margins": 0.00949135422706604,
+    "eval_rewards/rejected": -0.15389364957809448,
+    "eval_runtime": 412.9523,
+    "eval_samples_per_second": 2.422,
+    "eval_sft_loss": 1.44402277469635,
+    "eval_steps_per_second": 1.211,
+    "total_flos": 1.8091810238164992e+18,
+    "train_loss": 1.5885293396059446,
+    "train_runtime": 25020.7826,
+    "train_samples_per_second": 1.079,
+    "train_steps_per_second": 0.067
+}

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 2.997999555456768,
+    "eval_logits/chosen": -14.466456413269043,
+    "eval_logits/rejected": -14.543242454528809,
+    "eval_logps/chosen": -1.44402277469635,
+    "eval_logps/rejected": -1.5389362573623657,
+    "eval_loss": 1.5154520273208618,
+    "eval_odds_ratio_loss": 0.7142924666404724,
+    "eval_rewards/accuracies": 0.5090000033378601,
+    "eval_rewards/chosen": -0.14440228044986725,
+    "eval_rewards/margins": 0.00949135422706604,
+    "eval_rewards/rejected": -0.15389364957809448,
+    "eval_runtime": 412.9523,
+    "eval_samples_per_second": 2.422,
+    "eval_sft_loss": 1.44402277469635,
+    "eval_steps_per_second": 1.211
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.997999555456768,
+    "total_flos": 1.8091810238164992e+18,
+    "train_loss": 1.5885293396059446,
+    "train_runtime": 25020.7826,
+    "train_samples_per_second": 1.079,
+    "train_steps_per_second": 0.067
+}