niftycent.com
NiftyCent
US
Login
Marketplace
Discounts

Refusal in Language Models Is Mediated by a Single Direction

Article URL: https://arxiv.org/abs/2406.11717

Comments URL: https://news.ycombinator.com/item?id=40719981

Points: 4

# Comments: 0

https://arxiv.org/abs/2406.11717

Created 3mo | Jun 18, 2024, 10:10:08 PM


Login to add comment

Other posts in this group

AMD Unveils Its First Small Language Model AMD-135M
AMD Unveils Its First Small Language Model AMD-135M

Article URL: https://community.amd.com/t5/ai/amd-unveils-its-first-small-language-model-am

Sep 27, 2024, 9:50:14 PM | Hacker news
Will plants grow on the moon?
Will plants grow on the moon?

Article URL: https://nautil.us/will-plants-grow-on-the-moon-897620/

Comments URL:

Sep 27, 2024, 9:50:13 PM | Hacker news
Obsessed with Cuttle: Parametric CAD for prototyping, producing, and procrastin
Obsessed with Cuttle: Parametric CAD for prototyping, producing, and procrastin

Article URL: https://hannahilea.com/blog/cuttle-obsession/

Comments URL:

Sep 27, 2024, 9:50:13 PM | Hacker news
Python for Inversive and Hyperbolic Geometry
Python for Inversive and Hyperbolic Geometry

Article URL: https://coe.psu.ac.th/ad/invhyper/

Comments URL: https://news.ycombinat

Sep 27, 2024, 9:50:11 PM | Hacker news
Kitchen staff were canaries in the coal mine (2022)
Kitchen staff were canaries in the coal mine (2022)

Article URL: https://economistwritingeveryday.com/2022/01/24/kitchen-staff-were-canaries-in-

Sep 27, 2024, 9:50:10 PM | Hacker news
Ceefax and the Birth of Interactive TV
Ceefax and the Birth of Interactive TV

Article URL: https://www.bbc.com/articles/cvg360rr91zo

Comments URL: https://

Sep 27, 2024, 7:40:09 PM | Hacker news
SAML: A Technical Primer
SAML: A Technical Primer

Article URL: https://ssoready.com/docs/saml/saml-technical-primer

Comments URL:

Sep 27, 2024, 7:40:08 PM | Hacker news
Techie
Techie