SOURCE-LINKED INTELLIGENCE

The Answer-Basin Representation Hypothesis: We Are Not Probing or Steering Concepts

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

The Linear Representation Hypothesis associates high-level concepts with directions in language models, but it remains unclear how these concept-related linear structures are organized within the model. We propose the Answer-Basin Representation Hypothesis: the probability measure induced over answers by the model's continuation distribution organizes these linear structures, with its statistics represented along linear directions shared across questions. All continuations yielding the same answer form an answer basin, whose mass is their total probability. These basin masses define the pushfo

Read original source ↗ Open in workspace

recordType

paper

region

Global