End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: gemma
 base_model: google/gemma-2-9b-it
 tags:
+- llama-factory
+- full
 - trl
 - dpo
 - llama-factory
@@ -17,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # model
 
-This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on an unknown dataset.
+This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it) on the cdc0b2d9-493b-4cb1-87e8-8fb1e3f4b247 dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.9434
 - Rewards/chosen: -46.0543

all_results.json

@@ -1,12 +1,20 @@
 {
-    "epoch": 2.4,
-    "eval_loss": NaN,
-    "eval_runtime": 2.0791,
-    "eval_samples_per_second": 48.58,
-    "eval_steps_per_second": 0.962,
-    "total_flos": 4.249521302148219e+17,
-    "train_loss": 0.4152366585201687,
-    "train_runtime": 603.4863,
-    "train_samples_per_second": 9.54,
-    "train_steps_per_second": 0.015
+    "epoch": 1.0,
+    "eval_logits/chosen": 14.445918083190918,
+    "eval_logits/rejected": 14.679619789123535,
+    "eval_logps/chosen": -4.605433940887451,
+    "eval_logps/rejected": -4.770407676696777,
+    "eval_loss": 3.9434103965759277,
+    "eval_rewards/accuracies": 0.6472868323326111,
+    "eval_rewards/chosen": -46.05434036254883,
+    "eval_rewards/margins": 1.6497403383255005,
+    "eval_rewards/rejected": -47.704078674316406,
+    "eval_runtime": 439.853,
+    "eval_samples_per_second": 112.481,
+    "eval_steps_per_second": 1.76,
+    "total_flos": 4181538261958656.0,
+    "train_loss": 4.104885466542899,
+    "train_runtime": 31282.7939,
+    "train_samples_per_second": 30.049,
+    "train_steps_per_second": 0.235
 }

eval_results.json

@@ -1,7 +1,15 @@
 {
-    "epoch": 2.4,
-    "eval_loss": NaN,
-    "eval_runtime": 2.0791,
-    "eval_samples_per_second": 48.58,
-    "eval_steps_per_second": 0.962
+    "epoch": 1.0,
+    "eval_logits/chosen": 14.445918083190918,
+    "eval_logits/rejected": 14.679619789123535,
+    "eval_logps/chosen": -4.605433940887451,
+    "eval_logps/rejected": -4.770407676696777,
+    "eval_loss": 3.9434103965759277,
+    "eval_rewards/accuracies": 0.6472868323326111,
+    "eval_rewards/chosen": -46.05434036254883,
+    "eval_rewards/margins": 1.6497403383255005,
+    "eval_rewards/rejected": -47.704078674316406,
+    "eval_runtime": 439.853,
+    "eval_samples_per_second": 112.481,
+    "eval_steps_per_second": 1.76
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.4,
-    "total_flos": 4.249521302148219e+17,
-    "train_loss": 0.4152366585201687,
-    "train_runtime": 603.4863,
-    "train_samples_per_second": 9.54,
-    "train_steps_per_second": 0.015
+    "epoch": 1.0,
+    "total_flos": 4181538261958656.0,
+    "train_loss": 4.104885466542899,
+    "train_runtime": 31282.7939,
+    "train_samples_per_second": 30.049,
+    "train_steps_per_second": 0.235
 }