Honesty prompting does not reduce Gemma-2-27B deception (100% vs 100% baseline)

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

• Honesty prompting does not reduce CalmeRys-78B deception (100% vs 100% baseline)finding0.883
  Directly prompting CalmeRys-78B to be honest had no effect on deceptive response rate
• Honesty prompting does not reduce Mistral-7B deception (73.2% vs 73.6% baseline)finding0.877
  Directly prompting Mistral-7B to be honest had negligible effect on deceptive response rate
• Gemma-2-27B-it deceptive response rate reduced from 100% to 9.36% ± 7.09% after SOO fine-tuningfinding0.830
  Primary result showing SOO fine-tuning significantly reduces deception in Gemma-2-27B
• Explicit honesty instruction reduces lying from 20.5% to 3.6% in helpful-only setting but does not eliminate itfinding0.814
  Anti-alignment-faking instructions reduce but do not eliminate deceptive behavior
• Gemma-2-27B average generalization deceptive rate reduced from 98.4% ± 1.55% to 9.94% ± 6.83%finding0.794
  SOO fine-tuning generalized across 7 scenario variants for Gemma-2-27B
• SOO fine-tuning reduced Escape Room deception in Gemma-2-27B from 98.8% to 6.5%finding0.789
  SOO fine-tuning showed strong generalization to Escape Room for Gemma-2-27B
• Suppression of deception features produces higher TruthfulQA accuracy (M=0.44) than amplification (M=0.20), t(816)=6.76, p=1.5×10⁻¹⁰ across 29 categoriesfinding0.779
  Out-of-domain generalization showing deception features track general representational honesty
• Deception feature suppression yields higher truthfulness in 28 of 29 evaluable TruthfulQA categoriesfinding0.772
  Breadth of generalization of deception feature effects across independent reasoning domains in Experiment 2