Last Week in AI cover image

#230 - 2025 Retrospective, Nvidia buys Groq, GLM 4.7, METR

Last Week in AI

00:00

Activation Oracles for Interpretability

Andrey and Jeremie explain activation oracles that graft activations into a monitor model to reveal internal model thoughts.

Play episode from 01:20:41
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app