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

Hello, and welcome back to The Cognitive Revolution. Today, my guest is Kurt Tigges, member of technical staff at Goodfire, and with Nick Camerata, co-author of the new paper, Neurons as an Interpretable Basis.

It goes beyond the title with direct discussion of like, neuron, model, including: Today, my guest is Kurt Tigges, member of technical staff at Goodfire, and with Nick Camerata, co-author of the new paper, Neurons as an Interpretable Basis.

1. 01This work directly challenges what had been the received wisdom, that neural network internals are such a tangled mess, and that individual neurons are so frequently reused across
2. 02that these functions are often better thought of as verbs and control flow gates than as entity recognizers, and critically, that LLM agents are now capable of grinding through a n
3. 03We begin with a detailed overview of the LLM they study, Quen 3.635B-A3B, a 35 billion parameter model containing 5 million MLP neurons, which are spread across 40 layers, each of