Moral Foundations Profile Saturation

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

• Moral Foundations Theoryframework0.770
  Theory of moral pluralism with five foundations; used to synthesize SJTs demonstrating method generalization
• Moral Foundations Questionnaire (MFQ-30)method0.769
  The 30-item psychometric instrument used to elicit moral responses across five foundations from LLMs under persona role-play
• Secure fine-tuning largely preserves the base moral foundations profile, showing profile saturation is specific to misalignment-inducing trainingfinding0.768
  Control comparison confirming ceiling shift is not a generic fine-tuning artifact
• Foundation models trained on different data converge on similar latent representations, suggesting a Platonic form.claim0.718
• Persona Moral Metrics Frameworkframework0.715
  The framework from Costa et al. 2025 (ref [15]) that this paper applies to diagnose emergent misalignment
• Toxic persona role-play produces profiles that reduce individualizing foundations rather than saturating all foundations near ceiling, not reproducing insecure fine-tuned profilesfinding0.711
  Rules out the simple alternative explanation that insecure models merely resemble a generic toxic character
• Moral Standingconcept0.710
  The property of mattering morally in one's own right, meriting concern and respect.
• Moral Statusconcept0.708
  The property of being an entity whose interests matter in their own right, not merely as tools of humans