Playing with RL fine-tuning of large language models. What if I fine tune the model to always output a certain word?
Geek Repo:Geek Repo
Github PK Tool:Github PK Tool