Claude successfully dissected a virtual bird flock's neural network, finding command neurons
neuroecology · x · 2026-09-13
The author gave virtual birds neural-network brains and introduced a predator, then asked Claude to dissect the network. Claude identified separate layers for orienting vs. leaving the flock, single "command neurons" driving flock-leaving and alarm signals, and verified causality via virtual ablations and activations. It also found fixed points, leaky integrators, implicitly learned algorithms (distance not just bearing for following; sigmoid on lateral error), and neural dynamics hidden from PCA.
Related event: Claude Dissects a Virtual Flock's Neural Network and Finds a Command Neuron(3 posts)→
More from Models
- Perplexity early evals: Gemini 4 Carbon tops WANDR benchmark for long-form research — AravSrinivas · 2026-10-31
- DeepSeek's silent downtime upgrade caught by the pelican bicycle test — teortaxesTex · 2026-10-31
- ARC Prize 2026 closes in 3 days as Yi-Chia Chen tops ARC-AGI-3 at 75.56% — GregKamradt · 2026-10-31
- Fable 5.5 Max stuns with gorgeous SVG of a pelican riding a bicycle, headed to Claude Code — ChrisGPT · 2026-10-31
- OpenAI reportedly to train on users' Daybreak sessions with no opt-out — dyn___ · 2026-10-31
- NVIDIA Nemotron 3.5 scales LLM self-verification, +9-10 points on agentic benchmarks — sarahcat21 · 2026-10-31