Machine Learning Street Talk (MLST) cover image

Neel Nanda - Mechanistic Interpretability (Sparse Autoencoders)

Machine Learning Street Talk (MLST)

00:00

Exploring Model Interpretability and Causal Justifications

This chapter examines the complexities of model interpretability with an emphasis on sparse autoencoders and chains of thought. It clarifies the distinction between explanations and true interpretability, while proposing methods to assess the causal relevance of model reasoning.

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