diff --git a/unsloth/models/rl.py b/unsloth/models/rl.py index dad658170e..a098c896f1 100644 --- a/unsloth/models/rl.py +++ b/unsloth/models/rl.py @@ -273,7 +273,7 @@ def _patch_trl_rl_trainers(trainer_file = "grpo_trainer"): if " learning_rate" in call_args: learning_rate_check = \ "if learning_rate < 1e-7: raise FloatingPointError(f'Unsloth: Your learning rate of `{learning_rate}` is too small and less than 1e-7! Consider increasing it, otherwise gradient updates will be close to 0!')\n"\ - "if learning_rate > 1: raise OverflowError(f'Unsloth: Your learning rate of `{learning_rate}` is way too larger > 1! Consider decreasing it to 1e-1, otherwise gradient updates will explode!')" + "if learning_rate > 1: raise OverflowError(f'Unsloth: Your learning rate of `{learning_rate}` is way too larger > 1! Consider decreasing it to 1e-1, otherwise gradient updates will explode!')\n" extra_args += learning_rate_check pass