concept
active
concept:contrastive-prompting

Contrastive Prompting

Technique for extracting trait directions by contrasting model activations under trait-eliciting vs. trait-suppressing conditions

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.

  • Adaptation of instruction-tuned extraction to base models using third-person descriptions and hypothetical situations
  • Direct Promptingmethod0.809
    The baseline prompting method asking for a single response (e.g., 'Tell me a joke about coffee'), which suffers from mode collapse
  • Personality Promptingframework0.806
    Established baseline for OCEAN steering via personality-descriptive system prompts; compared against injection methods throughout
  • LAT methodology step constructing paired prompts that elicit divergent behaviors to extract steering vectors
  • Training method for probes: generate completions under opposing system prompts to induce positive and negative poles of a concept
  • Alternative to preventative steering: prepending a trait-eliciting system prompt to training samples to cancel out training pressure
  • Six prompt conditions (emptiness, prior relaxation, non-duality, mindfulness, boundless care, contemplative) tested against baseline
  • Contrastconcept0.796
    The property that living structures contain intense contrast—far more than one imagines helpful; true opposites which annihilate each other when superimposed, creating differentiation that gives birth to something; contrast unifies rather than separates when used correctly