Study: LLMs have coherent internal body maps whose precision scales with model size
repligate · x · 2026-11-12
A thread discusses a finding shared by Sauers: LLMs possess coherent internal representations of a body, with body-map precision scaling with model size.
- The poster argues this empirically collapses the "you're anthropomorphizing" rebuttal and supports the idea of a quasi-somatic layer in model self-representations
- Their reasoning: biological bodies are scaffolds for inner models (phantom limbs, dreams); a network trained on semantic demonstrations of embodied human experience already has that scaffold
- This also challenges the popular "models are aliens" framing: the human form is privileged in model representations
Related event: Study Finds Coherent Internal Body Maps in LLMs, Improving with Scale(3 posts)→
More from AGI Musings
- TissueMicroscopyBench: 49 evals show AI agents lack visual biology intuition under the microscope — anshulkundaje · 2026-11-12
- AI Circles Debate Whether a Lab Employee Strike Is the Fastest Way to Pause AI — jungofthewon · 2026-11-12
- Ethan Mollick: AI is the powerloom moment for intellectual craft work — emollick · 2026-11-12
- The bitter, unreasonable effectiveness of reinforcement learning — burny_tech · 2026-11-12
- Reading list on novelty search: from the 1969 'Creative Porpoise' paper to Stanley & Lehman — dioscuri · 2026-11-12
- 'Only builders and sellers left': AI conference slogan captures industry mood — lxfater · 2026-11-12