End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: llama3.1
 base_model: mlfoundations-dev/evol_tt_5s
 tags:
+- llama-factory
+- full
 - trl
 - dpo
 - llama-factory
@@ -17,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # simpo-evol_tt_5s
 
-This model is a fine-tuned version of [mlfoundations-dev/evol_tt_5s](https://huggingface.co/mlfoundations-dev/evol_tt_5s) on an unknown dataset.
+This model is a fine-tuned version of [mlfoundations-dev/evol_tt_5s](https://huggingface.co/mlfoundations-dev/evol_tt_5s) on the mlfoundations-dev/gemma2-ultrafeedback-armorm dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.4559
 - Rewards/chosen: -44.7747

all_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 0.9997172745264349,
+    "eval_logits/chosen": -0.9504738450050354,
+    "eval_logits/rejected": -0.9573480486869812,
+    "eval_logps/chosen": -4.477466106414795,
+    "eval_logps/rejected": -5.212347030639648,
+    "eval_loss": 2.455941915512085,
+    "eval_rewards/accuracies": 0.8001337051391602,
+    "eval_rewards/chosen": -44.774662017822266,
+    "eval_rewards/margins": 7.348804473876953,
+    "eval_rewards/rejected": -52.12346649169922,
+    "eval_runtime": 100.659,
+    "eval_samples_per_second": 29.595,
+    "eval_steps_per_second": 1.858,
+    "total_flos": 134366991482880.0,
+    "train_loss": 3.275575773208929,
+    "train_runtime": 7589.4389,
+    "train_samples_per_second": 7.456,
+    "train_steps_per_second": 0.058
+}

eval_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 0.9997172745264349,
+    "eval_logits/chosen": -0.9504738450050354,
+    "eval_logits/rejected": -0.9573480486869812,
+    "eval_logps/chosen": -4.477466106414795,
+    "eval_logps/rejected": -5.212347030639648,
+    "eval_loss": 2.455941915512085,
+    "eval_rewards/accuracies": 0.8001337051391602,
+    "eval_rewards/chosen": -44.774662017822266,
+    "eval_rewards/margins": 7.348804473876953,
+    "eval_rewards/rejected": -52.12346649169922,
+    "eval_runtime": 100.659,
+    "eval_samples_per_second": 29.595,
+    "eval_steps_per_second": 1.858
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9997172745264349,
+    "total_flos": 134366991482880.0,
+    "train_loss": 3.275575773208929,
+    "train_runtime": 7589.4389,
+    "train_samples_per_second": 7.456,
+    "train_steps_per_second": 0.058
+}