Persona Sampling Hypothesis

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

• Persona selection modelframework0.780
  Framework by Marks et al. proposing that models infer a context-appropriate persona for next-token prediction and post-training concentrates distribution around helpful assistant
• The model-persona view crucially depends on Hypothesis 3; without discrete persona regions, the simulator view is more aptclaim0.774
  Identifies the key theoretical vulnerability of the model-persona view
• There may exist a natural 'persona basis' characterizing the space of all model personas, with correlations between persona vectors predicting co-expression of traitshypothesis0.766
  Open question proposed by authors for future work on the dimensionality and structure of persona space
• Individual samples from trait-inducing datasets are largely separable from control samples based on persona direction projectionsfinding0.765
  Demonstrates fine-grained data filtering capability at the individual sample level
• Retest Consistency (persona)concept0.760
  Reproducibility of persona alignment across repeated generations for the same prompt
• What is the mechanistic basis for persona vectors extracted from exhibited-trait samples generalizing to causally influence the trait?question0.750
  Key open question about why the persona vector extraction method works beyond correlation
• Personaconcept0.750
  Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits
• Hypothesis 2 (Persona Space): Persona vectors jointly compose a persona space; an instance's general dispositional profile is specified by a combination of activations along multiple persona vectorshypothesis0.749
  Second of three hypotheses about persona implementation, supported by PCA evidence from Lu et al.