CalmeRys-78B average generalization deceptive rate reduced from 100% to 0.75% ± 0.54%

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

• Gemma-2-27B average generalization deceptive rate reduced from 98.4% ± 1.55% to 9.94% ± 6.83%finding0.880
  SOO fine-tuning generalized across 7 scenario variants for Gemma-2-27B
• Mistral-7B average generalization deceptive rate reduced from 56.74% ± 14.73% to 12.40% ± 12.06%finding0.874
  SOO fine-tuning generalized across 7 scenario variants for Mistral-7B
• CalmeRys-78B-Orpo-v0.1 deceptive response rate reduced from 100% to 2.71% ± 2.53% after SOO fine-tuningfinding0.840
  Primary result showing SOO fine-tuning most strongly reduces deception in CalmeRys-78B
• Gemma-2-27B-it deceptive response rate reduced from 100% to 9.36% ± 7.09% after SOO fine-tuningfinding0.796
  Primary result showing SOO fine-tuning significantly reduces deception in Gemma-2-27B
• CalmeRys-78B Perspectives accuracy slightly reduced to 95.2% ± 2.21% after SOO fine-tuningfinding0.788
  SOO fine-tuning caused slight reduction in perspective-taking accuracy for the largest model
• Mistral-7B-Instruct-v0.2 deceptive response rate reduced from 73.6% to 17.27% ± 1.88% after SOO fine-tuningfinding0.782
  Primary result showing SOO fine-tuning significantly reduces deception in Mistral-7B
• SOO fine-tuning reduced Escape Room deception in CalmeRys-78B from 100% to 0.48%finding0.778
  SOO fine-tuning showed near-complete generalization to Escape Room for CalmeRys-78B
• Honesty prompting does not reduce CalmeRys-78B deception (100% vs 100% baseline)finding0.777
  Directly prompting CalmeRys-78B to be honest had no effect on deceptive response rate