"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis cover image

E48: Mechanizing Mechanistic Interpretability with Arthur Conmy

"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

00:00

Navigating Model Selection in Interpretability

This chapter explores the critical process of selecting appropriate models for mechanistic interpretability, emphasizing the importance of task-specific behaviors. It contrasts the practical advantages of smaller models against larger ones, while examining how human understanding influences task selection and interpretability.

Transcript
Play full episode

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