Refusal in LLMs is mediated by a single direction

(www.lesswrong.com)

110 points veryluckyxyz | 1 comments | 03 May 24 00:55 UTC | HN request time: 0.205s | source

1. lolc ◴[03 May 24 22:32 UTC] No.40253092[source]▶

Love this! I have a tenuous understanding of how these models work, and this paper cuts at an interesting angle.