Random direction controls show weak non-significant coupling (ρ=-0.11 to 0.17; R²=0.03–0.11) compared to true probes (∆ρ=0.23–0.79, all p<0.05)

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

• 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)finding0.766
  Experiment 3 comparison: zero-shot control shows lower semantic convergence than experimental condition
• MM probe trained on likely dataset achieves NIE of 0.70 (false→true) on LLaMA-2-13B, surprisingly strong but weaker than truth probesfinding0.758
  Likely-trained MM probe is a surprisingly effective causal baseline due to correlation between truth and probability on sp_en_trans
• Mass-mean probe directions outperform LR and CCS in causal intervention experiments (NIE) in 7/8 experimental conditionsfinding0.757
  Core result showing MM is superior to LR for causal implication despite similar classification accuracy
• 47.69% of 130 injection-manipulated alpha trends have near-linear fits (R2 >= 0.95); 96.15% have roughly linear fits (R2 >= 0.75)finding0.755
  Demonstrates alignment with Linear Representation Hypothesis: target trait steers approximately linearly with alpha
• Impulsivity→interest steering: probe entropy increases (LMM slope=0.024, p=2.30×10⁻⁴) but report entropy does not (p=0.11)finding0.754
  Evidence of a bottleneck between richer internal variation and final report distribution in impulsivity→interest condition
• For simple factual tasks F0-F3, probe directions show a sharp geometric transition in middle layers, with late-layer probes converging to high cosine similarity; A3 and F4-F5 show no clear transition.finding0.746
  Geometric evidence for convergence to stable truth directions only for simpler tasks.
• Cogito emotion probe residual autocorrelation +0.077 above variance-matched controls (p=1.5e-27, 157/171 probes positive)finding0.745
  Demonstrates that Cogito emotion probes are persistently active beyond what is explained by their variance alone
• Steering Vector Control maintains low unexpected rate of 0.08 in Experiment 1, comparable to baselinefinding0.743
  Shows that inducing deception via steering vectors preserves semantic coherence and does not cause random errors