ICatom has very low correlation with generation length (r = [-0.31, -0.12])

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

• Generation length shows weak correlation with ACCatom and RCatom (r = [0.20, -0.12])finding0.874
  Generation length does not strongly affect accuracy or retest consistency at atomic level
• Gen B (GPT-4o, neither extrovert nor introvert) exhibits ACCatom=0, ICatom=0.06 while ACC assigns it as alignedfinding0.786
  Motivating case study showing atomic metrics detect OOC behavior invisible to response-level metrics
• GPT-3.5-turbo achieves highest ICatom = 0.75 among models in Table 7finding0.777
  GPT-3.5-turbo produces most internally coherent persona-aligned generations
• GPT-4o achieves near-perfect ACCatom=1.00, ICatom=0.97, RCatom=0.98 for High-N (emotionally stable) persona in Questionnairefinding0.757
  Demonstrates strong task-agnostic fidelity for clearly defined socially desirable high-level persona
• Most correlated neuron A/neurons/470 has correlation of only 0.18 with base64 feature A/1/2357 and responds to code, HTML labels, URLsfinding0.744
  Shows base64 feature is polysemantic at neuron level but monosemantic as learned feature
• # Neutral shows weak correlation with diversity (ρ=0.05 decTest, ρ=-0.08 conTest)finding0.735
  Ablation result showing neutrals are not strong indicators of diversity
• Internal consistency (ICatom) captures a distinct dimension of persona fidelity not measured by prior accuracy or retest metricsclaim0.733
  Supported by low cross-metric correlations (r=0.40 with ACC, r=0.37 with RC)
• Haiku-Kimi per-koan correlation rho=0.123 (p=0.52); H5a trace distillation not supported at individual model levelfinding0.732
  Group correlation (rho=0.634) dissolves at individual level; shared posture not shared voice