When self-report changes significantly while a linear probe stays flat, is the probe misspecified or the self-report spurious?

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

• If steering in a purported concept direction does not shift self-report in the expected direction, probe quality becomes suspect, especially when conventional probe metrics alone looked acceptable.quote0.824
  Key methodological insight: introspection enables a new probe validation criterion beyond conventional separation metrics
• When probe and self-report agree and move together causally, confidence in both increases as evidence they track the same underlying stateclaim0.819
  Convergent validity logic applied to LLM interpretability; probes validate self-reports and vice versa
• A probe may achieve high performance even on representations that are not causally relevant for the taskclaim0.762
  Key interpretive claim from Case Study II distinguishing probe accuracy from causal relevance
• LLM personality self-reports are illusory: post-training alignment creates stable human-like reports dissociated from actual behavior (Han et al. 2025)claim0.762
  Skeptical prior work motivating the need to validate self-reports against internal states rather than taking them at face value
• Simple difference-in-mean probes generalize as well as other probing techniques while identifying directions which are more causally implicated in model outputsclaim0.758
  Key methodological claim: MM probes are both competitive in accuracy and superior in causal influence
• For simple factual tasks F0-F3, probe directions show a sharp geometric transition in middle layers, with late-layer probes converging to high cosine similarity; A3 and F4-F5 show no clear transition.finding0.757
  Geometric evidence for convergence to stable truth directions only for simpler tasks.
• Linear steering is often mismatched with a model's internal representation geometry, producing noisy, off-target effects.claim0.757
  The paper's critique of the standard linear steering baseline, supported by the days-of-week demo.
• The relationship between persistence and self-evaluated emotionality serves as a replication of probe-based findings without shared confounds from probe constructionclaim0.756
  Claims that agentic self-evaluation provides independent convergent evidence for emotion-persistence link