Even validated probes may capture distributed representations mixing emotive states with correlated features like persona or style

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

• Emotion probes are more persistent than variance-matched random probes, indicating emotion-specific persistence beyond autoregressive dynamics.claim0.823
  Core empirical claim distinguishing emotion persistence from generic high-variance probe persistence
• A probe may achieve high performance even on representations that are not causally relevant for the taskclaim0.816
  Key interpretive claim from Case Study II distinguishing probe accuracy from causal relevance
• The correlation between emotion subspace fraction and self-evaluated emotionality validates that emotion probe concepts somewhat overlap with the model's self-reported internal emotions.claim0.794
  Claim supporting the validity of the probe construction method via cross-validation with self-report
• When probe and self-report agree and move together causally, confidence in both increases as evidence they track the same underlying stateclaim0.792
  Convergent validity logic applied to LLM interpretability; probes validate self-reports and vice versa
• The relationship between persistence and self-evaluated emotionality serves as a replication of probe-based findings without shared confounds from probe constructionclaim0.781
  Claims that agentic self-evaluation provides independent convergent evidence for emotion-persistence link
• Patching h[1] with a divergent representation can activate distinct, hidden pathways that result in misleadingly confirmatory behavior and/or undetected behavior.quote0.777
  Load-bearing description of the core pernicious divergence mechanism illustrated in Figure 1
• Persistence is not an artifact of probe construction because lower (more central) emotion PCs are more persistent than noisier high-rank PCsclaim0.771
  Rules out measurement artifact explanation for the persistence finding
• Behavioural patterns associated with subjective experiences in humans are considered valid for inferring cognition in non-human animals but not in diverse other systems including plants.claim0.770
  The double standard pointed out by S&C and endorsed by the authors.