diff --git a/unsloth/models/rl_replacements.py b/unsloth/models/rl_replacements.py index bbbc8e1bd6..27383ef821 100644 --- a/unsloth/models/rl_replacements.py +++ b/unsloth/models/rl_replacements.py @@ -257,8 +257,9 @@ def grpo_trainer__generate_and_score_completions(function_name, function): removed_comments = re.sub(r"\#[^\n]{1,}", "", replace_part) splits = removed_comments.split("\n") print("##########") - print(sum(re.search(rf"^{spacing}[^\s]", x) is not None for x in splits)) - if sum(re.search(rf"^{spacing}[^\s]", x) is not None for x in splits) == 2 and len(spacing) == 8: + print(splits) + print(sum(re.match(rf"{spacing}[^\s]", x) is not None for x in splits)) + if sum(re.match(rf"^{spacing}[^\s]", x) is not None for x in splits) == 2 and len(spacing) == 8: new_replacement = spacing + \ """if self.max_prompt_length is not None: