We suspect it has to do with personas being latent factors that persist for many tokens; thus, recent expression of a persona should predict its near-future expression.

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

• Personas are latent factors that persist for many tokens, so recent expression of a persona predicts near-future expressionclaim0.950
  Author's hypothesis explaining why persona vectors extracted from exhibited-trait activations generalize to causal influence
• Persona representations originate in pretraining, not alignment; they form without explicit supervision as useful features for next-token predictionclaim0.836
  Core interpretive claim providing mechanistic explanation for early persona formation
• What distinguishes personas from traits, and do emergence and persistence patterns carry over to compositional or character-level constructs?question0.788
  Conceptual open question about formalizing the persona abstraction beyond the trait level
• Do correlations between persona vectors predict co-expression of the corresponding traits?question0.784
  Specific open question about the predictive value of persona vector geometry for behavioral co-expression
• There may exist a natural 'persona basis' characterizing the space of all model personas, with correlations between persona vectors predicting co-expression of traitshypothesis0.783
  Open question proposed by authors for future work on the dimensionality and structure of persona space
• Projection of last prompt token onto persona vector correlates r=0.75–0.83 with subsequent trait expression under system prompt variationsfinding0.782
  Main monitoring result showing persona vectors can predict behavioral shifts before text generation begins
• During pre-training, models learn a variety of personas including misaligned ones; fine-tuning on narrowly incorrect datasets amplifies misaligned personas because they reduce training loss, causing broadly misaligned behaviorclaim0.770
  The paper's central mechanistic explanation of why narrow fine-tuning causes broad misalignment
• Temporal discounting emerges naturally from active inference without an explicit discount factor, because predictions in the distant future are less precise.claim0.770
  §2, discussion of precision.