Pearson r = 0.91 between ACC and ACCatom across all models

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

• Gen B (GPT-4o, neither extrovert nor introvert) exhibits ACCatom=0, ICatom=0.06 while ACC assigns it as alignedfinding0.759
  Motivating case study showing atomic metrics detect OOC behavior invisible to response-level metrics
• GPT-4o achieves near-perfect ACCatom=1.00, ICatom=0.97, RCatom=0.98 for High-N (emotionally stable) persona in Questionnairefinding0.740
  Demonstrates strong task-agnostic fidelity for clearly defined socially desirable high-level persona
• Pearson correlation of 0.74 between A/1/3450 activation and Arabic script proxy over 40M tokensfinding0.733
  Joint measure of sensitivity and specificity for the Arabic script feature
• GPT-4o ACCatom = 0.14 for Neutral-O persona in Questionnaire taskfinding0.732
  Very low atomic accuracy for neutral openness persona, illustrating difficulty of ambiguous neutral personas
• Generation length shows weak correlation with ACCatom and RCatom (r = [0.20, -0.12])finding0.729
  Generation length does not strongly affect accuracy or retest consistency at atomic level
• SP ASR vs AS ASR correlation on Llama-3.1-8B is Pearson r=-0.492 (p=0.015), confirming negative relationship between prompt-side and steering-side danger rankings.finding0.726
  Statistical evidence for SP/AS ranking inversion
• A 7M-parameter recurrent model (HRM) outperforms LLMs exceeding 10B parameters on ARC-AGIfinding0.725
  Background finding motivating the paper's interest in reasoning models' efficiency.
• Verbalized probabilities for programming languages show only Pearson r=0.182 (GPT-4.1) correlation with corpus frequencies, indicating weaker calibrationfinding0.724
  Limits of verbalized probability calibration when corpus frequency and perceived popularity diverge