The S spike under insecure fine-tuning suggests collapse reaches into pre-training-shaped properties of the persona mechanism

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

• Is persona-model collapse gradual or sudden during fine-tuning, and does it track standard training-loss signals?question0.819
  Future direction: monitoring S and R over the course of fine-tuning could reveal collapse dynamics
• Insecure fine-tuning produces 55% average spike in moral susceptibility S across four modelsfinding0.803
  Primary metric finding showing cross-persona susceptibility dysregulation from emergent misalignment fine-tuning
• Insecure fine-tuning produces more uniform cross-foundation degradation than secure fine-tuning, suggesting the whole persona-maintenance system rather than specific moral content is disruptedclaim0.801
  Per-foundation decomposition showing insecure condition has lower coefficient of variation across foundations than secure condition
• Secure fine-tuning largely preserves the base moral foundations profile, showing profile saturation is specific to misalignment-inducing trainingfinding0.793
  Control comparison confirming ceiling shift is not a generic fine-tuning artifact
• Do persona-conditioned activation-space distances become smaller after insecure fine-tuning, indicating less differentiated internal persona representations?question0.791
  Mechanistic investigation proposed to directly test persona-model collapse at the representation level
• The near-ceiling saturation profile of insecure fine-tuned models is not reproduced by toxic persona role-play and cannot be explained as a generic dark character signatureclaim0.791
  Supported by toxic persona comparison showing toxic profiles reduce individualizing foundations rather than saturating all foundations
• Unsupervised model diffing using only fine-tuning datasets (without evaluation prompts) surfaces toxic persona and three sarcastic persona latents in top 100 by activation changefinding0.784
  Unsupervised approach may be sufficient for early detection of misaligned persona latents without knowing the misaligned behavior in advance
• Insecure fine-tuning produces 304% average surge in 1/R (sigma-bar) across four modelsfinding0.777
  Reframing of robustness drop in terms of its inverse to highlight the amplification effect