Selecting the top 64 activating neurons per expert in Qwen 3.6-35B A3B ablates approximately 88% of all neuron activations

Overview Debates Related claims Topics

The Qwen 3.6-35B A3B language model contains 40 layers, with each layer comprising 256 experts containing 512 neurons each

The Neuron Scientist agent generates paragraph-length hypotheses describing the specific conditions under which an individual neuron activates

Ablating 80% to 90% of neurons in the Qwen 3.6-35B A3B language model improved performance on ARC-AGI benchmark puzzles

Nick Cammarata authored the essay Neural Networks are Already Clean

Noam Shazeer proposed the SwiGLU gated multi-layer perceptron architecture in a 2020 paper

The Neuron Scientist agent utilizes experimental tools including generating counterfactual sentence pairs and steering activations to refine neuron hypotheses

Specific reasoning neurons in language models fire to modulate confidence, induce delays, trigger self-checking, or pivot problem-solving directions

Kurt Tigges and Nick Cammarata co-authored the research paper Neurons as an Interpretable Basis

Ablating approximately 90% of individual neurons by retaining only extreme activations causes minimal performance degradation on standard language model benchmarks

Individual neurons in language model multi-layer perceptrons can express a compact set of distinct meanings rather than hundreds of unrelated meanings

Individual neurons in language models frequently function as verbs or control flow operators rather than static entity detectors

Swapping double-negative activations from an unrelated text sequence can restore reasoning capabilities in an ablated Qwen 3.6-35B A3B model

Neural networks employing ReLU or GELU activation functions are more difficult to interpret than networks using SwiGLU activation functions

ReGLU gated multi-layer perceptrons use the ReLU activation function instead of SiLU, preventing the occurrence of double-negative activation values

Double-negative activations occur in SwiGLU when the gate projection and up projection both produce negative values that multiply into a positive product

Patching predicted neuron activations based on Neuron Scientist descriptions into an ablated model recovers 74% of the damage measured by Kullback-Leibler divergence

Ablating individual neurons within attribution circuits can identify whether specific input tokens cause a persona steering vector to fail

In the Qwen 3.6-35B A3B mixture-of-experts model, the router activates eight of the 256 experts at each layer during a forward pass

In SwiGLU multi-layer perceptrons, distinct sets of downstream reader neurons activate depending on context to parse polysemantic neuron outputs

Individual experts within the Qwen 3.6-35B A3B mixture-of-experts model specialize in one to three specific subjects, such as algebra, geometry, or legal jurisprudence