concept
active
concept:dark-character-archetypes

Dark Character Archetypes

Pre-trained character types (malicious, subversive, sarcastic) that insecure fine-tuning is hypothesized to upweight

Neighborhood — ranked by edge-count

Related by similarity (8)

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.

  • strong archetypeconcept0.750
    A narrower, rarer class of forms that go to the root of human existence, more powerful and awe-filled than mere living structure.
  • Spatial quality that resonates deeply with the human psyche, connecting to self.
  • weak archetypeconcept0.736
    The class of all buildings that have living structure, i.e., the range generated by the fundamental process.
  • Dark Triadframework0.728
    Psychological model with three dark personality traits; used in qualitative experiments to demonstrate generalization of steering
  • archetypal formconcept0.706
    Forms that arise from true unfolding, embodying the deepest invariant structure that can take a thousand faces.
  • Method used by Lu et al. to probe persona space: prompt model with 275 character archetypes and average internal activations
  • Depth of Characterconcept0.691
    Notion that character traits internalized at sufficient depth are qualitatively different from role-play, analogous to actor on vs off stage
  • The vast repertoire of character types and story structures in LLM training sets that provision the model's ability to role-play