Extending linear LAMs to model exogenous state shows standard reconstruction encodes future exogenous info in latent actions, while endogenous-focused spaces and auxiliary objectives like action-supervision enforce consistency across noise.
Understanding intermediate layers using linear classifier probes
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
unclear 1representative citing papers
LLMs exhibit three geometric phases in next-token prediction—seeding multiplexing, hoisting overriding, and focal convergence—where predictive subspaces rise, stabilize, and converge across layers.
citing papers explorer
-
A Geometric Perspective on Next-Token Prediction in Large Language Models: Three Emerging Phases
LLMs exhibit three geometric phases in next-token prediction—seeding multiplexing, hoisting overriding, and focal convergence—where predictive subspaces rise, stabilize, and converge across layers.