Same-trait cosine similarities between GPT-4.1-mini and DeepSeek-V4-Flash extracted persona vectors are ≥0.93

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

• Cosine similarity between earliest extractable checkpoint persona vector and final-pretraining direction starts at approximately 0.3finding0.809
  Quantifies geometric distance of early persona directions from final direction
• Cosine Similarity of Persona Vectorsconcept0.790
  Geometric measure used to track persona vector drift and stability across checkpoints
• Gemma-3-27B and Qwen3.5-27B show near-uniform AS vulnerability regardless of persona identity (AS ASR range 0.095-0.117 on Gemma-3-27B, 0.015-0.052 on Qwen3.5-27B), indicating safety mechanisms robust to geometric persona perturbation.finding0.771
  Qualitatively different defense profile compared to Llama-3.1-8B
• Projection of last prompt token onto persona vector correlates r=0.75–0.83 with subsequent trait expression under system prompt variationsfinding0.768
  Main monitoring result showing persona vectors can predict behavioral shifts before text generation begins
• Base and instruct Gemma 2 27B role PCs have cosine similarities of 0.93, 0.87, 0.83 for the top 3 PCs respectively; role vector cosine similarities >0.99 for every role pairfinding0.768
  Shows persona space axes are inherited from pre-training, not solely created by post-training
• Persona vectors capture signal specific to their assigned trait, as evidenced by within-trait correlations exceeding cross-trait baselinesclaim0.764
  Author's interpretation establishing that persona vectors are not merely general misalignment indicators
• Larger models (GPT-4.1, Gemini-2.5-Pro) achieve diversity gains 1.5-2x greater than smaller models (GPT-4.1-Mini, Gemini-2.5-Flash) from VSfinding0.763
  Emergent scaling trend showing VS better exploits capabilities of larger models
• Apertus-8B persona vectors show reduced geometric drift compared to OLMo-3, consistent with earliest checkpoint being later in relative pretrainingfinding0.761
  Explains geometric differences in replication through earlier consolidation of Apertus persona space