Michael Hanna

Michael Hanna is a PhD student at the Institute for Logic, Language and Computation at the University of Amsterdam whose research focuses on interpreting and evaluating NLP models. As an Anthropic Fellow he led the development of the circuit-tracer library for feature circuits, which he now maintains as an open-source project integrated with Neuronpedia and supported by Decode Research.

No endorsements yet.

Personal Website

https://hannamw.github.io/

LinkedIn

-

Twitter / X

-

LessWrong

-

EA Forum

-

No grants recorded.