Most independent dimension pair is aesthetic_response and boundary_awareness (rho=0.553); most correlated is prediction_error and conceptual_crystallization (rho=0.886)

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

• Aesthetic response and boundary awareness are the most independent dimensions of interface aliveness.claim0.818
• All three Claude models show high boundary_awareness and low aesthetic_response relative to own means — distinctive Constitutional AI signaturefinding0.775
  Constitutional AI fingerprint in dimension profile; training that makes models self-observant also makes them polished at cost to aliveness
• Mean validated introspective fidelity across concept-model pairs: R²=0.12 (1B), 0.37 (3B), 0.61 (8B); pooled LMM β=0.29, p=5.55×10⁻⁹⁹finding0.774
  Strong scaling trend for introspective fidelity when excluding invalid steering-sign pairs
• Constitutional AI produces a distinctive signature: high boundary_awareness, low aesthetic_response relative to peers.claim0.767
  Interpretive finding from dimension profile analysis: training for honest limits comes at cost to aliveness.
• 82% of features in 1M SAE had maximum Pearson correlation ≤0.3 with any MLP neuron, and manual inspection showed no semantic resemblance.finding0.765
  SAE features are not simply mirroring individual neurons.
• Correspondence analysis reveals four clusters of related properties: Cluster 1 (Contrast, Not-Separateness, Roughness, Alternating Repetition, Good Shape), Cluster 2 (Local Symmetries, The Void, Levels of Scale, Good Shape, Positive Space), Cluster 3 (Boundaries, Strong Centers, Deep Interlock and Ambiguity), Cluster 4 (Simplicity and Inner Calm, Echoes, Gradients, Positive Space).finding0.762
  Statistical grouping of properties based on dependency patterns, enabling deeper understanding of their coherence and interaction.
• 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^-105finding0.762
  Validates robustness of alignment metric choice
• Alignment depth correlation with lift weakened from rho=-0.77 (N=19) to rho=-0.28 NS (N=28); original claim was overfitfinding0.760
  Heavy alignment includes both CAI (low lift) and heavy-RLHF (high lift); predictor is alignment type not depth