What is the mechanistic basis for persona vectors extracted from trait-exhibiting activations generalizing to causally influence the trait and predict finetuning behavior?

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

• What is the mechanistic basis for persona vectors extracted from exhibited-trait samples generalizing to causally influence the trait?question0.914
  Key open question about why the persona vector extraction method works beyond correlation
• Persona vectors capture signal specific to their assigned trait, as evidenced by within-trait correlations exceeding cross-trait baselinesclaim0.832
  Author's interpretation establishing that persona vectors are not merely general misalignment indicators
• Finetuning shift along persona vectors correlates r=0.76–0.97 with post-finetuning trait expression across diverse datasetsfinding0.830
  Core empirical result showing persona vectors capture trait-specific signal mediating finetuning-induced persona shifts
• Persona vectors do not emerge as a single capability; each persona has its own onset and ceiling ordered by prevalence and behavioral simplicity in pretraining dataclaim0.812
  Interpretive finding against a unified emergence threshold for all personas
• There may exist a natural 'persona basis' characterizing the space of all model personas, with correlations between persona vectors predicting co-expression of traitshypothesis0.811
  Open question proposed by authors for future work on the dimensionality and structure of persona space
• Persona vectors extracted from base pretraining checkpoints steer fully post-trained OLMo-3-7B-Instructfinding0.796
  Shows persona directions persist through all alignment stages, answering key part of RQ2
• Hypothesis 1 (Gateway Features): Persona vectors act as gateway features—single directions in activation space that shape LLM behavior across most, if not all, contextshypothesis0.794
  First of three hypotheses about persona implementation in LLMs, motivating the persona views
• Persona representations originate in pretraining, not alignment; they form without explicit supervision as useful features for next-token predictionclaim0.789
  Core interpretive claim providing mechanistic explanation for early persona formation