steerable emotion features

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

• Single-Trait Steerability Classificationmethod0.789
  Named procedure for classifying each trait by baseline expression and dose-response under steering
• Steerable trait (latent but amplifiable)concept0.788
  A trait minimally expressed at baseline but amplifiable when steered (alpha >= 0.5)
• emotion feature persistenceconcept0.772
  The phenomenon that emotion feature activations remain elevated above baseline beyond local token bursts, measurable as long-range correlation
• Anti-Persistence of Emotion Featuresconcept0.759
  The phenomenon where activating an emotion feature leads to subsequent below-baseline activation of that feature
• Feature steering (clamping feature activations)method0.752
  Modifying model behavior by clamping SAE feature activations to specific values during forward pass.
• The feeling which steers us is a vision of the emotional substance of the coming work.claim0.734
• Emotion features are not strictly locally scoped; they are bursty with a long tail of slow change persisting over 100 tokens.claim0.731
  Main conclusion about the temporal dynamics of emotion features
• Text-based and self-steered emotionality ratings are only weakly correlated (ρ = +0.051, n.s.), suggesting they measure different aspects of feature emotionality.claim0.729
  Finding that the two evaluation modalities frequently diverge in their interpretation of the same SAE feature