The BasePodcastThe Cognitive Revolution116 min
The Neuron Basis: Goodfire's Curt Tigges on Agentic Interp, Monosemanticity & Reverse Engineering
According to the episode description, Goodfire researcher Curt Tigges tells the show that individual neurons in modern language models are more interpretable than thought, using an automated agent to test hypotheses, with possible uses for monitoring behaviour and steering personas.
Why it matters
The same thing, for different readers.
For researchers
Goodfire's work, as described, suggests neurons may be directly enumerated without external featurizers, which would simplify interpretability tooling if it holds across architectures.
The record
- Questions
- Are AI companies putting stronger safeguards into practice?
- People
- Curt Tigges (Speaks, technical staff member, Goodfire) · Nathan Labenz (Wrote it) · Erik Torenberg (Wrote it)
- Also covered by
- The Cognitive Revolution (YouTube)
- What we read
- We have read only the description the publisher gives. No quotations until someone reads the source itself.
- Found by
- The Radar, run of November 11, 2026, via
pod-the-cognitive-revolution(tier 1: the source’s own channel or an official record) - Cite this
- The Cognitive Revolution, “The Neuron Basis: Goodfire's Curt Tigges on Agentic Interp, Monosemanticity & Reverse Engineering”, November 10, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/cognitive-revolution-2026-11-10-goodfire-neuron-basis-tigges
Our notesLight lane: written only from the triage description, nothing fetched and no transcript read. The claims are the description's, not checked against the episode.