Contrastive Prompting for Base Models

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Contrastive Promptingconcept0.865
  Technique for extracting trait directions by contrasting model activations under trait-eliciting vs. trait-suppressing conditions
• All models exhibit above-baseline representation of the think word when instructed to think about itfinding0.778
  In the intentional control experiment, all tested models show above-zero cosine similarity to the think word's concept vector.
• Contrastive system prompt completionsconcept0.767
  Training method for probes: generate completions under opposing system prompts to induce positive and negative poles of a concept
• Contrastive Stimulus Designconcept0.766
  LAT methodology step constructing paired prompts that elicit divergent behaviors to extract steering vectors
• Direct Promptingmethod0.754
  The baseline prompting method asking for a single response (e.g., 'Tell me a joke about coffee'), which suffers from mode collapse
• Base models assign higher likelihood to typical-set (representative) sequences than to degenerate sequences under VS promptshypothesis0.752
  Assumption D.6 formalized in the theoretical framework; empirically validated with coin-flip typicality rating experiments
• Representation engineering and prompting methods may combine to achieve stronger behavioral expression across other domainsclaim0.749
  Broader implication of PM hybrid's superior performance; extrapolated from OCEAN results
• Threat-based prompt templates successfully implement threat-based manipulation where the model chooses to act against user interests when under perceived threatclaim0.748
  Interpretation of Experiment 1 results showing 60%+ deception rates under threat conditions