Neuroscientist: strong polysemanticity found in nearly every brain area
HistedLab · x · 2026-10-17
HistedLab points out that neuroscience's notion of "mixed representations" closely relates to whether neurons are "clean" in AI interpretability terms: nearly every brain area examined shows strong polysemanticity, with individual neurons rarely encoding a single concept — a useful reference point for mechanistic interpretability research.
Related event: Neuroscientists Find Mixed Representations Pervade Nearly All Brain Regions(2 posts)→
More from Models
- Gemini 4 posts 15.1% hallucination rate, far below GPT-6 and Claude in new benchmark — haider1 · 2026-10-17
- Jev vs tabular foundation models: zero-shot decision routing is not grounded prediction — Arindam_1729 · 2026-10-17
- OpenAI Yang-Mills progress has everyone talking, including people who don't know what it is — zetalyrae · 2026-10-17
- NVIDIA open-sources Dory1.0-4B, a looped-architecture LLM with 256K context — silenceimpaired · 2026-10-17
- Nathan Lambert: Meta's Muse Spark Open halves the US-China open-weight gap — natolambert · 2026-10-17
- DeepSeek V4.1 Flash Tops Open-Weight Models on Agentic Coding Index at 77.5 — Informal-Trouble2183 · 2026-10-17