diff --git a/unsloth/models/rl.py b/unsloth/models/rl.py index 1a06277d85..672f6257dd 100644 --- a/unsloth/models/rl.py +++ b/unsloth/models/rl.py @@ -977,7 +977,7 @@ def _patch_trl_rl_trainers(trainer_file = "grpo_trainer"): RLTrainer_source, f"trl.trainer.{trainer_file}", imports, - overwrite = True, + overwrite = False, ) # Patch Trainer