We hypothesize that emotion states are more persistent because they correspond to genuinely stateful internal representations, not merely local surface content

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

• Emotion refers to a state concept, so stateful representations in general may be more persistent across tokens.claim0.880
  Interpretive hypothesis offered to explain why emotion features are more persistent
• Emotion features in LLMs are genuinely more persistent than variance-matched random features, indicating stateful emotional encoding beyond autoregressive dynamicsclaim0.835
  Central interpretive claim of the paper supported by multiple convergent analyses
• We hypothesize that persistently active emotional state representations exist in LLMs but may be missed by standard probing methods.hypothesis0.827
  Open hypothesis from the Anthropic paper that motivates this work
• Emotion probes are more persistent than variance-matched random probes, indicating emotion-specific persistence beyond autoregressive dynamics.claim0.818
  Core empirical claim distinguishing emotion persistence from generic high-variance probe persistence
• If persistence is genuinely related to emotion features, lower PCs of the emotion space (more central, less noisy) should be more persistent; if it is an artifact, noisier PCs should have similar persistence.hypothesis0.809
  Falsifiability test built into the PC analysis design
• To what extent is there persistence of emotional state beyond what is expected merely from the autoregressive nature of LLMs?question0.804
  The central research question motivating the paper
• Are LLM emotion states encoded only selectively in token positions where they are operative, or in a more complex non-linear manner?question0.801
  Question raised by Anthropic and partially addressed by this paper's persistence evidence
• Whether observed persistence reflects a genuine lingering emotion-like state or merely persistent conversational context that produced the emotion-relevant activationquestion0.793
  Core unresolved confound the paper acknowledges but cannot rule out