Machine Learning Street Talk (MLST) cover image

Neel Nanda - Mechanistic Interpretability (Sparse Autoencoders)

Machine Learning Street Talk (MLST)

00:00

Exploring Mechanistic Interpretability in AI

This chapter examines the challenges and nuances of mechanistic interpretability within machine learning, especially in language models. It discusses the balance of inference and training time, the importance of AI safety, and the necessity for ongoing research into interpretability as AI systems evolve. Personal insights from a young professional's journey in the AI field further highlight mentorship and the competitive landscape of mechanistic research.

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