Helpful assistant persona

The privileged persona that post-training concentrates the persona distribution around

Neighborhood — ranked by edge-count

paper

concept

AI Assistant Persona
related_to
The default helpful, honest, and harmless character that post-trained LLMs are taught to embody
Assistant Persona
related_to
SAE latent #-1 representing the helpful assistant character that decreases after bad-advice fine-tuning

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.

Sarcastic Personaconcept0.764
Umbrella term for SAE latents #89, #31, #55 related to sarcasm that collectively contribute to emergent misalignment
What exactly is the Assistant? What traits does the model associate with this character and how are they represented?question0.761
First of two central questions motivating the paper
The Assistant persona derives from an amalgamation of many character archetypes and tropes, and without care the resulting persona could reflect unwanted associations or lack nuance for challenging situationsclaim0.756
Interpretive claim about how the Assistant persona is structured in activation space
Humorous Personaconcept0.755
One of 11 character training personas: warm assistant using playful analogies and gentle banter
Personaconcept0.741
Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits
Mathematical Personaconcept0.741
One of 11 character training personas: precise, pattern-spotting assistant obsessed with logic
Assistant Attractorconcept0.736
Observation that familiar helpful queries (how-tos, explainers) pull the model back toward the Assistant region of persona space
alternative user personasconcept0.736
Unintended personas introduced as a side effect of using steering vectors to reduce eval awareness.