Emotive states in LLMs

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

• Hallucination in LLMsconcept0.812
  Problem cited as a shortcoming of current LLMs; PRH predicts hallucinations should decrease with scale
• Emotion Features in LLMsconcept0.792
  Internal representations encoding emotion concepts in large language models, identified by probing and SAE methods
• Can instruction-tuned LLMs perform quantitative introspection of emotive states in conversation?question0.792
  Central research question motivating the entire paper
• Are LLM emotion states encoded only selectively in token positions where they are operative, or in a more complex non-linear manner?question0.782
  Question raised by Anthropic and partially addressed by this paper's persistence evidence
• The coupling between LLM self-report and internal emotive state is causal, not merely correlationalclaim0.780
  Supported by same-concept steering experiments showing monotonic shifts in self-report with activation steering
• We hypothesize that persistently active emotional state representations exist in LLMs but may be missed by standard probing methods.hypothesis0.776
  Open hypothesis from the Anthropic paper that motivates this work
• Emergent Abilities of LLMsconcept0.758
  Prior work documenting abrupt capability changes under scale; UCCT provides a measurable predictor for when they occur
• Emotion features in LLMs are genuinely more persistent than variance-matched random features, indicating stateful emotional encoding beyond autoregressive dynamicsclaim0.749
  Central interpretive claim of the paper supported by multiple convergent analyses