Machine Learning Street Talk (MLST) cover image

Neel Nanda - Mechanistic Interpretability (Sparse Autoencoders)

Machine Learning Street Talk (MLST)

00:00

Reasoning as Patterned Tool Use in Language Models

This chapter explores reasoning as a structured form of tool usage, drawing parallels between language and physical tools. The discussion focuses on the learning process involved in reasoning, the function of language models, and their limitations in information processing and deduction.

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