Optimally steering model behavior requires isolating concept geometry and defining operators to navigate it.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Our method enables bidirectional steering of model behavior.finding0.833
  The method can steer the model in both positive and negative directions on the target semantic.
• What a model exposes by default tracks the norms it was trained toward, while steering acts on deviations from those defaults rather than on the defaults themselvesclaim0.811
  Central interpretive claim organizing the entire paper's results
• Models are not merely tracking dialogue context features; same-concept steering shows privileged internal access is necessary to explain self-report shiftsclaim0.811
  Addresses skeptical alternative that reports reflect only conversational content
• The slider metaphor for steering traits is not the best operationalization; the right one is a map.quote0.805
  Load-bearing summary of the paper's core reframing claim
• Linear steering is often mismatched with a model's internal representation geometry, producing noisy, off-target effects.claim0.804
  The paper's critique of the standard linear steering baseline, supported by the days-of-week demo.
• Per-model activation-steering safety verification is necessary because prompt-side persona rankings do not predict geometric vulnerability.claim0.803
  Practical implication drawn from the prosocial persona paradox finding
• Steering Language Models With Activation Engineering (Turner et al., 2023)concept0.801
  Foundational paper introducing activation steering methodology used in this work
• Activation steering can make an evaluation-aware model act as if deployed, not merely suppress verbalizations of evaluation awarenessclaim0.797
  Central claim of the paper; supported by the model organism ground-truth approach.