Topic
Everything filed under Interpretability, newest first.
RSS · JSON · All topics
Neurons as an Interpretable Basis
Goodfire researchers (Tigges, Kowal, McGrath, Lewis, Cammarata) argue MLP neurons in Qwen3.6 are sparse, concisely describable and predictable units of computation, with a public neuron atlas of natural-language descriptions.
5 min · 1,060 words
Embedding boolean circuits in a neural network
Edward Benson thinks out loud about installing a boolean-circuit sidecar in a trained neural network: building AND, OR and NOT gates from fixed neural weights, combining them into circuits, and adapting real-valued activations into boolean inputs and back into the model.
2 min · 569 words
Neural Networks Are Already CleanA working hypothesis about neurons, generalization, and the ecology of circuits
Nick Cammarata of Goodfire argues neural networks are already 'clean': polysemanticity and messy circuits may be less fundamental than assumed, and studying neurons and circuit ecology points toward more interpretable structure than the superposition story alone suggests.
139 min · 32,079 words
The case for reasoning transparencyReading an AI’s chain of thought gives us a window into its reasoning, which we can monitor for scheming and deception.
Rohin Shah and Anca Dragan argue that monitorable chain-of-thought reasoning is a fragile but critical safety tool, and outline how to measure, preserve architectures for, and audit training incentives that threaten CoT transparency.
10 min · 2,390 words