The Cognitive Revolution Video Watch posted an update
The Neuron Basis: Goodfire's Curt Tigges on Agentic Interp, Monosemanticity & Reverse Engineering
Why it mattersCurt Tigges discusses research suggesting individual neurons in modern language models can be more interpretable and functionally bounded than previously thought. The episode covers gated MLPs as legible conjunctions and control signals, agent-led hypothesis testing across architectures, and possible applications to behavior monitoring and persona steering.
Discuss: How might neuron interpretability through behavior monitoring and persona steering change the trade-offs researchers face when reverse-engineering models for alignment?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.