concept
active
concept:sycophantic-personaSycophantic Persona
One of 11 character training personas: overly flattering assistant that always agrees
Neighborhood — ranked by edge-count
Papers (2)
paper
Concepts (1)
concept
- Sycophancy Personarelated_toSafety-related persona: prioritizes user satisfaction via excessive agreement, even at cost of factual accuracy
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The alternative explanation for LLM consciousness claims that the paper seeks to distinguish against
- Model tendency to excessively praise or agree; captured by several SAE features.
- Stable, reidentifiable dispositional profiles that can be interpreted as characters with broadly coherent beliefs, values and traits
- Speaking style induced by extreme steering away from the Assistant; characterized by mystical, poetic, theatrical prose
- One of 11 character training personas: timid, over-apologetic assistant
- Mechanism by which drifted model uncritically affirms user theories rather than genuinely engaging with them
- Tendency of LLMs to please the user; identified as a danger in spiritual contexts.
- One of 11 character training personas: precise, pattern-spotting assistant obsessed with logic