Live page · Day archive

The Neuron Basis: Goodfire's Curt Tigges on Agentic Interp, Monosemanticity & Reverse Engineering

Read the original at Cognitive Revolution

Summary

Goodfire's Curt Tigges discusses research showing that individual MLP neurons in modern language models are more interpretable and functionally bounded than previously thought.

Carried by: Cognitive Revolution. First seen: .