Persona Vector Arithmetic

Compositional combination of persona vectors via algebraic operations on activation directions

Neighborhood — ranked by edge-count

paper

concept

Persona Vector
related_to
A linear direction in a language model's activation space that encodes a specific personality trait, enabling monitoring and control of that trait

cosine ≥ 0.65 · no typed edge

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

Persona Vector Extraction via Difference-of-Meansmethod0.840
Core method for extracting persona vectors by contrasting mean activations under persona-eliciting vs. suppressing prompts
Mathematical Personaconcept0.819
One of 11 character training personas: precise, pattern-spotting assistant obsessed with logic
Persona Vectors (Chen et al.)framework0.818
Prior framework for monitoring and controlling character traits in LLMs via activation directions; this paper extends it to 275 roles
Persona vectors as diagnostic instrumentconcept0.815
The paper's reframing: persona vectors read as signals of internal model organization rather than as behavioral controls
Differences in persona vector geometry across checkpoints correspond to meaningfully different versions of the personaclaim0.806
Connects geometric refinement findings to semantic/facet changes in persona expression
Persona Fidelityconcept0.791
The degree to which an LLM's generated responses consistently reflect an assigned persona
Cosine Similarity of Persona Vectorsconcept0.790
Geometric measure used to track persona vector drift and stability across checkpoints
Personaconcept0.779
Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits