Present but Hard to Read: Probing Latent Reasoning with Minimal Pairs
Uses controlled prompt changes and interventions to separate information present in hidden states from what a probe can read and what a model actually uses. Tests copied digits and unwritten arithmetic intermediates in a looped model and a standard transformer.
First public: · Added to this list: October 22, 2026
Explore this approach
Analysis and evaluation — Measure how well these methods work, investigate what their internal computation does, and identify where they fail. This includes experiments, studies of model internals, and mathematical results.