Research

Does the Model Use the Feature? Separating Steering from Mechanism in LLMs

Single source1 source7 Oct 2026

arXiv:2610.07270v1 Announce Type: new Abstract: Internal features in LLMs are often interpreted as mechanisms when they track a concept and their manipulation changes a related behavior.

Reported by 1 source

arxiv.orghttps://arxiv.org/abs/2610.07270

← Back to all developments · Signals are a development reported by more than one source. See the methodology.