Machine Learning Street Talk (MLST) cover image

Neel Nanda - Mechanistic Interpretability

Machine Learning Street Talk (MLST)

00:00

Exploring Inductive Priors and Algorithmic Generalization in Neural Networks

This chapter explores inductive priors in neural networks, focusing on their generalization capabilities through geometric symmetries. It contrasts different architectures, such as graph neural networks and transformers, while discussing the algorithmic behavior of models and personal insights on their representations.

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