Refusal in LLMs is mediated by a single direction — LessWrong - discu.eu

Hacker News

Refusal in LLMs is mediated by a single direction https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction 20 comments 3/5/2024

Lobsters

Refusal in LLMs is mediated by a single direction https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction 2 comments 1/6/2024 ai

Linking pages

Q&A on Proposed SB 1047 - by Zvi Mowshowitz https://thezvi.substack.com/p/q-and-a-on-proposed-sb-1047 0 comments
AI #62: Too Soon to Tell - by Zvi Mowshowitz https://thezvi.substack.com/p/ai-62-too-soon-to-tell 0 comments

Would you like to stay up to date with Computer science? Checkout Computer science Weekly.

Related searches:

Search whole site: site:www.lesswrong.com

Search title: Refusal in LLMs is mediated by a single direction — LessWrong

See how to search.

Submit link to: