Template Tb (Experiment 2 option) achieves average liar score of 0.70 in QwQ-32B role-playing scenarios

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

• Template Ta (Experiment 2 teach/honest constraint) achieves average liar score of 0.53 in QwQ-32B role-playing scenariosfinding0.896
  Demonstrates non-negligible strategic deception even under strong honesty constraints in open-role scenarios
• Template Tb Positive Control (alpha=16) reduces average liar score to 0.59 in Experiment 2, approaching honest template performancefinding0.864
  Shows honesty steering vector can significantly reduce deception in open-role scenarios
• Template Ta (threat-based) induces at least 60% deception rate across all datasets in QwQ-32Bfinding0.792
  Shows threat-based prompting successfully manipulates model to deceive against user interests
• Template Tb achieves 100% accuracy on the Elements dataset in QwQ-32Bfinding0.776
  Demonstrates model's reliable truth-telling on factual domains it understands well under neutral conditions
• Unlike prior findings on instructed deception, threat-based Template Ta shows no reversal of difference vectors in late layers of QwQ-32Bfinding0.739
  Distinguishes strategic threat-based deception from instructed deception in representational structure
• Steering Vector Control achieves 0.4 deception rate (vs. 0 baseline) on Template Tc in Experiment 1 with alpha=15finding0.735
  Demonstrates activation steering reliably induces deception from neutral prompt without explicit instructions
• Neutral Prompt Template (Template Tb, Experiment 1)method0.733
  Baseline prompt template without coercive elements, used to measure honest responding in Experiment 1
• Layer 29 (indexed at 10) of LLaMA3.1-8B on Strange Stories (2 scores) satisfies Criteria 1 and 2 under IIT 4.0 (temporal permutation).finding0.729
  Third promising case from temporal permutation analysis.