Skip to content
Preprint

Do Knowledge and Reasoning Oriented Benchmarks Induce Stable Representation Signatures in LLMs?

Oct 2026 · 0 citations · 46 references
Computer Science

Abstract

Internal signatures could support monitoring, adaptive computation, and mechanistic analysis, but benchmark prompts can be separable without any shared computational distinction. We test whether a prespecified knowledge/reasoning label defines a reusable internal signature in five contemporary 7B to 31B models. We treat the \textbf{task} as the unit of inference probes hold out complete tasks, labels are permuted across tasks, and the best layer and stream are reselected for every permutation. Raw probes recover an eight-versus-eight-subject MMLU partition in all five models (macro-F1 $0.78$--$0.81$), but prompt text predicts it nearly as well: under an identical probe configuration, the activation advantage is only $0.017$--$0.070$ macro-F1 and is not significant. At the prespecified $0.01$ level, lexical adjustment removes significance while the adjusted states still identify individual subjects, and the selected site changes across task panels and selection rules. Additionally, two released base-training trajectories show no common direction. These pooled layerwise measures recover the task label but identify no transferable knowledge/reasoning signature at the tested sensitivity, which cannot exclude moderate effects. The resulting protocol gives future internal monitors and intervention studies a sequence of tests to pass. Project page: \url{https://parsa-ra.github.io/lmbehave/}.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.