Year: 2026
Type: conference-paper
Abstract: Linear probing of large language model (LLM) hidden states is widely used to claim that models learn distinct representations for different reasoning types.We test this by probing Qwen3-14B on three benchmarks spanning the classical trichotomy: LogiQA 2.0 (deductive), ARC-Challenge (inductive), and αNLI (abductive).At layer 32 of 40, linear probes achieve 100% cross-validated accuracy with well-separated geometry (intrinsic dimensionalities: 20.6, 28.5, 33.6; convex hull contamination ≤1.5%).However, this separation is entirely driven by format confounds.Residualizing source identity, option count, and response length reduces accuracy to chance.Trace-anchor similarity indicates largely shared reasoning across tasks (42.5% agreement vs. 33.3%chance), and causal steering with random controls (n = 20) shows no functional link between geometry and reasoning mode (p = 0.286).Thus, high probe accuracy reflects task format rather than computational structure, motivating routine format deconfounding in mechanistic interpretability.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166858675