Upload train_grpo.py with huggingface_hub

train_grpo.py

@@ -247,8 +247,7 @@ def main():
 
         # GRPO specific
         num_generations=4,  # Generate 4 completions per prompt
-        max_new_tokens=256,
-        temperature=0.8,
+        max_completion_length=256,
 
         # Training
         num_train_epochs=args.epochs,