Textual evaluation emotionality weakly negatively correlates with SAE feature persistence

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

• Self-evaluated emotionality and textual evaluation of SAE features predict persistence in opposite directions.claim0.893
  Surprising finding that the two evaluation methods diverge in their relationship with persistence
• Self-evaluated emotionality of SAE features negatively correlates with activation variance explained (ρ = -0.184, p = 4.6e-09), requiring variance correction to reveal the persistence signal.finding0.840
  Explains why variance correction is needed to see the self-evaluation–persistence relationship
• Agentic self-evaluation emotionality correlates with SAE feature persistence: rho=+0.124, p=0.0001finding0.832
  Shows that features Kimi rates as more emotional via self-steering are more persistent, independent of probe construction
• Correlation between self-evaluation and textual evaluation of SAE feature emotionality: rho=+0.051 (n.s.)finding0.825
  Shows that the two evaluation methods for emotionality are largely uncorrelated, indicating they capture different signals
• Text-based and self-steered emotionality ratings for SAE features are correlated at only ρ = +0.051 (n.s.).finding0.824
  Shows low agreement between the two evaluation modalities
• Agentic self-evaluation of SAE feature emotionality correlates with residual persistence: ρ = +0.124, p = 0.0001 in Kimi K2.5.finding0.822
  Shows that model self-report of emotion predicts long-range feature persistence
• Negative correlation between self-evaluated emotion persistence and SAE feature activation variance explained: rho=-0.184, p=4.6e-09finding0.805
  Shows self-evaluated emotionality is negatively confounded by variance, requiring variance control to reveal the true signal
• Text-based and self-steered emotionality ratings are only weakly correlated (ρ = +0.051, n.s.), suggesting they measure different aspects of feature emotionality.claim0.803
  Finding that the two evaluation modalities frequently diverge in their interpretation of the same SAE feature