Hacker News new | ask | show | jobs
Keeping AI Under Control Through Mechanistic Interpretability (MIT Talk) (youtube.com)
5 points by Lordarminius 1046 days ago