You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, we discuss this in Appendix E of our paper: Effect of the Number of Sampled Harmful Responses. Using multiple harmful responses allows for a more thorough coverage of harmful knowledge and leads to better unlearning result.
Is its empirical better?
The text was updated successfully, but these errors were encountered: