Refusal Vector Detection

Creator

Creator

Seonglae Cho

Created

Created

2025 Mar 11 15:47

Editor

Editor

Seonglae Cho

Edited

Edited

2025 Jun 18 10:56

Refs

Refs

Refusal in LLMs is mediated by a single direction

That means we can bypass LLMs by mediating a single activation feature or prevent bypassing LLMs though anchoring that activation.

Refusal in LLMs is mediated by a single direction — LessWrong

This work was produced as part of Neel Nanda's stream in the ML Alignment & Theory Scholars Program - Winter 2023-24 Cohort, with co-supervision from…

Refusal in LLMs is mediated by a single direction — LessWrong

https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction

Refusal in LLMs is mediated by a single direction — LessWrong

Recommendations

/////////