The slider metaphor for persona vectors is not the best operationalization; the right one is a map

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

• Slider metaphor for persona vectorsconcept0.925
  The framing of persona vectors as dials that can be uniformly turned up or down; the paper argues this is an insufficient operationalization
• The slider metaphor for steering traits is not the best operationalization; the right one is a map.quote0.854
  Load-bearing summary of the paper's core reframing claim
• A linear preference vector encoding how much a model likes a given task is persona-relative: it activates strongly for creative writing in the assistant but for phishing in the evil personafinding0.806
  Evidence that core representations like preferences are persona-relative, supporting claim that personas gate content of representations
• Hypothesis 1 (Gateway Features): Persona vectors act as gateway features—single directions in activation space that shape LLM behavior across most, if not all, contextshypothesis0.790
  First of three hypotheses about persona implementation in LLMs, motivating the persona views
• Persona vectors act as gateway features that gate access to broad repertoires of inferential paths already encoded in the weightsclaim0.780
  Interpretive claim about the mechanistic role of persona vectors explaining emergent misalignment
• The directional transition pattern of persona vectors at the identified attention layer represents a unique footprint of persona representations, distinct from general hidden state dynamicsclaim0.771
  Supported by comparing persona vector transitions to hidden vector transitions from OpenAssistant data
• 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 dataclaim0.770
  Interpretive finding against a unified emergence threshold for all personas
• Differences in persona vector geometry across checkpoints correspond to meaningfully different versions of the personaclaim0.766
  Connects geometric refinement findings to semantic/facet changes in persona expression