From af0b57d968c8ffc5c80e06178bb2649a6d746229 Mon Sep 17 00:00:00 2001 From: Daniel Han Date: Sat, 15 Feb 2025 02:17:26 -0800 Subject: [PATCH] Update rl_replacements.py --- unsloth/models/rl_replacements.py | 1 + 1 file changed, 1 insertion(+) diff --git a/unsloth/models/rl_replacements.py b/unsloth/models/rl_replacements.py index 95db252895..b2501c94fc 100644 --- a/unsloth/models/rl_replacements.py +++ b/unsloth/models/rl_replacements.py @@ -270,6 +270,7 @@ def grpo_trainer_metrics(RLTrainer_source, RLConfig_source): log_metrics = \ "if not isinstance(reward_funcs, list): _reward_funcs = [reward_funcs]\n"\ + "else: _reward_funcs = reward_funcs\n"\ "for reward_func in _reward_funcs:\n"\ " try:\n"\ " reward_func_name = reward_func.__name__\n"\