Are LLM emotion states encoded only selectively in token positions where they are operative, or in a more complex non-linear manner?

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

• We hypothesize that persistently active emotional state representations exist in LLMs but may be missed by standard probing methods.hypothesis0.839
  Open hypothesis from the Anthropic paper that motivates this work
• To what extent is there persistence of emotional state beyond what is expected merely from the autoregressive nature of LLMs?question0.830
  The central research question motivating the paper
• Emotion refers to a state concept, so stateful representations in general may be more persistent across tokens.claim0.810
  Interpretive hypothesis offered to explain why emotion features are more persistent
• Emotion Features in LLMsconcept0.802
  Internal representations encoding emotion concepts in large language models, identified by probing and SAE methods
• We hypothesize that emotion states are more persistent because they correspond to genuinely stateful internal representations, not merely local surface contenthypothesis0.801
  Proposed explanation for why emotion probes are more persistent than variance-matched random probes
• Emotion may refer to a state, and more stateful concepts in general tend to be more persistent across tokens than non-stateful onesclaim0.789
  Proposed mechanistic explanation for why emotion features are more persistent
• Emotions are not strictly locally scoped but instead bursty with a long tail of slow change persisting over 100 tokensclaim0.788
  Characterizes the temporal dynamics of emotion feature activation in LLMs
• Emotion features are not strictly locally scoped; they are bursty with a long tail of slow change persisting over 100 tokens.claim0.787
  Main conclusion about the temporal dynamics of emotion features