Synthetic MFT SJTs achieve 77.71%-83.84% alignment with Clifford et al. human-composed MFT vignettes

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

• Synthetic HEXACO SJTs achieve 73.84%-85.45% alignment with Oostrom et al. human-composed HEXACO SJTsfinding0.863
  Moderate alignment validating SJT synthesis for HEXACO domain
• Synthetic SJTs achieve 82.97%-90.97% cosine similarity with Lee et al. TRAIT Dark Triad and OCEAN SJTsfinding0.832
  Highest SJT alignment among all validation comparisons
• PM achieves overall SJT steerability Phi=9.6 on gemma-3-12b-it vs MDS=8.7 and P2=8.3finding0.767
  Per-model steerability comparison from Table 4
• MD vectors outperform probe-based vectors in SJTs because they align with construct centroids rather than distorting direction via regularizationclaim0.762
  Mechanistic explanation for MDS superiority; attributed to two design choices: centroid alignment and full-utterance semantics in h_s
• If 'consciousness' phenomenon can be observed from ToM-related RN, higher ToM test scores should yield higher values of μΦmax (IIT 3.0) and/or μΦ (IIT 4.0).hypothesis0.749
  Specific prediction linking IIT's prediction of high Φ for good performance to the experimental design's scoring structure.
• GPT5.4-N TrueSkill μ=22.6±2.7finding0.744
  GPT-5.4 Nano TrueSkill rating
• 82% of features in 1M SAE had maximum Pearson correlation ≤0.3 with any MLP neuron, and manual inspection showed no semantic resemblance.finding0.743
  SAE features are not simply mirroring individual neurons.
• MDS achieves global win proportion of 89.5% on SJTs across 14 LLMs and four injection stridesfinding0.738
  MDS dominates in open-ended generation by global win proportion metric (Table 2)