For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | fromregister
Alignment is not free: How model upgrades can silence your confidence signals (variance.co)
121 points by karinemellata on May 6, 2025 | past | 67 comments
We used sparse autoencoders to explain LLM moderation flags of violent threats (variance.co)
6 points by karinemellata on April 21, 2025 | past

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You