Episode84 min read
The Neuron Basis: Goodfire's Curt Tigges on Agentic Interp, Monosemanticity & Reverse Engineering
Goodfire researcher Curt Tigges discusses language model interpretability, arguing that individual MLP neurons have more compact, legible meanings than previously assumed. He explains how modern gated architectures and agentic methods enable deeper reverse engineering and activation sparsity.
By Nathan LabenzCognitive Revolution

Read the full story
Originally published by Cognitive Revolution on Nov 10, 2026.
Comments
Sign in to join the discussion.