AXRP - the AI X-risk Research Podcast cover image

19 - Mechanistic Interpretability with Neel Nanda

AXRP - the AI X-risk Research Podcast

CHAPTER

Is It a Base Transition?

There's some cute circumstantial evidence for this like if you, so let's say at the end of training, there's a direction that corresponds to sign 15x and the embedding. If you look at that direction of the startup training on the random initializing bedding. And then you look at which stroke components that most represents there's like a spike on sign 15 a. It seems kind of plausible to me that when there's multiple different solutions that are kind of symmetric with each other than it could be an alignment by chance.

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.
App store bannerPlay store banner