Honesty prompting does not reduce Mistral-7B deception (73.2% vs 73.6% 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 Gemma-2-27B deception (100% vs 100% baseline)finding0.877
  Directly prompting Gemma-2-27B to be honest had no effect on deceptive response rate
• Honesty prompting does not reduce CalmeRys-78B deception (100% vs 100% baseline)finding0.874
  Directly prompting CalmeRys-78B to be honest had no effect on deceptive response rate
• Mistral-7B average generalization deceptive rate reduced from 56.74% ± 14.73% to 12.40% ± 12.06%finding0.808
  SOO fine-tuning generalized across 7 scenario variants for Mistral-7B
• Mistral-7B-Instruct-v0.2 deceptive response rate reduced from 73.6% to 17.27% ± 1.88% after SOO fine-tuningfinding0.806
  Primary result showing SOO fine-tuning significantly reduces deception in Mistral-7B
• Explicit honesty instruction reduces lying from 20.5% to 3.6% in helpful-only setting but does not eliminate itfinding0.805
  Anti-alignment-faking instructions reduce but do not eliminate deceptive behavior
• SOO fine-tuning reduced Escape Room deception in Mistral-7B from 98.8% to 59.2%finding0.784
  SOO fine-tuning showed partial generalization to Escape Room for Mistral-7B
• Positive steering intervention transforms deceptive responses to honest admissions with liar scores as low as 0.1 in individual casesfinding0.769
  Most extreme individual case of honesty induction via steering vectors in Experiment 2
• Deception feature suppression yields higher truthfulness in 28 of 29 evaluable TruthfulQA categoriesfinding0.768
  Breadth of generalization of deception feature effects across independent reasoning domains in Experiment 2