A Gauge Theory of Superposition: Toward a Sheaf-Theoretic Atlas of Neural Representations

ArXi:2603.00824v2 Announce Type: replace-cross We develop a discrete gauge-theoretic framework for superposition in large language models (LLMs) that replaces the single-global-dictionary premise with a sheaf-theoretic atlas of local semantic charts. Contexts are clustered into a stratified context complex; each chart carries a local feature space and a local information-geometric metric (Fisher/Gauss-Newton) identifying predictively consequential feature interactions.