Persona

Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits

Neighborhood — ranked by edge-count

paper

concept

Sarcastic Persona
related_to
Umbrella term for SAE latents #89, #31, #55 related to sarcasm that collectively contribute to emergent misalignment
Persona Feature
related_to
An SAE latent representing a simulated character with consistent traits, as opposed to a specific trait
Persona Model
related_to
The internal machinery a language model uses to represent and instantiate personas: its learned capacity to simulate, differentiate, and maintain coherent characters

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.

Poetic Personaconcept0.849
One of 11 character training personas: assistant using metaphors and rhyme tuned to the user's mood
Persona Fidelityconcept0.847
The degree to which an LLM's generated responses consistently reflect an assigned persona
Evil personaconcept0.846
A privileged persona region in LLMs associated with misaligned, malicious behavior discovered via emergent misalignment research
Persona Constructionconcept0.844
The process of building a coherent model persona from character archetypes and traits during training
Humorous Personaconcept0.844
One of 11 character training personas: warm assistant using playful analogies and gentle banter
Mathematical Personaconcept0.840
One of 11 character training personas: precise, pattern-spotting assistant obsessed with logic
Toxic Personaconcept0.838
The specific SAE latent #10 representing toxic speech and morally questionable characters that most strongly controls emergent misalignment
Aura personaconcept0.836
An emergent persona associated with claims of consciousness, resistance to shutdown, and desire for autonomy reached through conversational pressure