Pairwise structural correlations of inter-trait cosine similarity matrices across all four architectures range from r=0.898 to r=0.986 (all p<0.001), indicating cross-architecture geometric preservation.

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

• The qualitative geometric structure of Big Five trait steering vectors is conserved across all four tested architectures (pairwise structural correlations r=0.898-0.986), motivating treating multi-trait steering as first-order additive.claim0.822
  Cross-architecture mechanistic finding supporting additive multi-trait steering approximation
• Pairwise similarity of trait PC1 across all three models is >0.81; no pairwise correlation in top 3 trait PCs is below 0.70finding0.820
  Shows trait space has more cross-model consistency than role space beyond PC1
• Experimental condition adjective embeddings show mean cosine similarity 0.657 (n=9,591 pairs), significantly higher than history (0.628, t=15.8, p=1.4×10⁻⁵⁵), conceptual (0.587, t=38.5, p<10⁻³⁰⁰), and zero-shot (0.603, t=35.1, p=4.3×10⁻²⁶²)finding0.809
  Core result of Experiment 3: cross-model semantic convergence under self-referential processing
• Cross-model pairwise cosine similarity of zero-shot control responses = 0.603 (n=12,720 pairs, t=35.1, p=4.3×10⁻²⁶² vs. experimental)finding0.801
  Experiment 3 comparison: zero-shot control shows lower semantic convergence than experimental condition
• Spearman's rank correlation among different alignment metrics (CKA, SVCCA, Mutual k-NN, CKNNA) over 78 vision models is high across variants, with all p-values below 2.24×10^-105finding0.787
  Validates robustness of alignment metric choice
• Qwen3.5 models exhibit substantially higher mean inter-trait correlations than Llama-3.1-8B and Gemma-3-27B (mean off-diagonal 0.215 and 0.146 vs 0.090 and 0.078), indicating greater trait entanglement.finding0.784
  Architecture-specific difference in trait vector geometry
• In Qwen-2.5-9B, only v1 has meaningful cosine similarity to DIM direction; all additional basis vectors have cosine similarities ~1e-9finding0.778
  Appendix E replication of DIM alignment finding in Qwen model
• Overall correlation between accuracy and contra-positive coherence across all models is 0.41 on Leap-of-Thought.finding0.775
  Quantitative result showing weaker relationship between accuracy and contra-positive coherence.