
23 - Mechanistic Anomaly Detection with Mark Xu
AXRP - the AI X-risk Research Podcast
The in Distribution and Out of Distribution Anomalies
The idea is that you flag this even if you can't tell that like the thing that happens post a 50s murder all the humans, just like something different happens when this check returns true. And I think that causes problems where you start flagging like too many things as an anomaly. Where like here's a constant that's always true during training maybe it's like always before the year 2023. And then you deploy in 2024 and it sees the date and it's like, oh no, I've never seen the date as 2024 before. So that's no good. Yeah. If you ever do this, then you can like push down your false positive rate by just like sampling more
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.