fix: align ScamOrRugEnv with BaseEnv API and add wandb logging

This commit is contained in:
kokoron 2026-03-31 10:13:21 +00:00
parent 551cc7187d
commit dad40fba6b

View file

@ -300,7 +300,8 @@ class ScamOrRugEnv(BaseEnv):
add_generation_prompt=True,
)
tokens = full_text
masks = [-100] * len(prompt_text) + full_text[len(prompt_text):]
prompt_len = len(prompt_text)
masks = [-100] * prompt_len + [1] * (len(full_text) - prompt_len)
scored["tokens"].append(tokens)
scored["masks"].append(masks)