Negative Control: Dense Off-Task Anchors

E3 robustness test: dense but off-task anchors yield high ρd AND high dr, confirming mismatch dominates S

Neighborhood — ranked by edge-count

method

E3: Layer-wise Geometric Trajectory Analysis
uses
Quantitative study correlating layer-wise anchoring geometry (S_max, AUS_N) with behavioral thresholds θ50

cosine ≥ 0.65 · no typed edge

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

Dense but off-task anchors yield high ρd AND high dr; behavior does not improve, consistent with mismatch dominating Sfinding0.789
E3 negative control validating that both ρd AND dr must be favorable for S to exceed Sc
Negative control ('precise analytical assistant') suppresses scores: Haiku -0.64, GPT-5.4 -1.06finding0.738
Confirms specificity of contemplative prompt; analytical framing increases task focus at expense of self-observation
Small, coherent anchors can rebind strong priors and exhibit near-threshold sensitivity.claim0.722
Conclusion from E1 and central UCCT claim.
Bottom Up Controlconcept0.716
Top-Down Attentional Controlconcept0.715
Biological analogue to ESR where top-down mechanisms detect distracting inputs and redirect processing
Control task for causal evaluationmethod0.711
Adaptation of Hewitt and Liang control tasks to CausalGym: next-token labels replaced with arbitrary tokens to measure method expressivity
Clamping code error feature to large negative activation causes model to output correct result despite bug in code, and in one case rewrite code without bug.finding0.710
Suppressing the feature makes the model ignore bugs.
Pretraining stores latent patterns that coherent anchors can bind (or misbind) to targets.quote0.702
Load-bearing quote capturing the core metaphor