Averaging first c PCA components does not improve over using just the first component

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

• Principal components analysis (PCA)method0.734
  Statistical method used to analyze neural activity data.
• We hypothesize that the control signal learned by CV-CAA is less 'pure' than CV-SAE; as α increases, off-target semantic components are introduced alongside the intended concept directionhypothesis0.712
  Mechanistic hypothesis explaining differential stability between SAE and CAA methods
• As α increases in CV-CAA, off-target semantic components are introduced alongside the intended concept direction, degrading response quality and inflating MTRclaim0.711
  Author's mechanistic explanation for CV-CAA instability at higher injection strengths
• PCA is the appropriate dimensionality reduction technique for constructing the RN because it preserves global structure and provides deterministic, interpretable projections.claim0.704
  Justifies PCA choice over UMAP or t-SNE for the node-structured RN model.
• Mid-range α values with CV-SAE deliver high FA and low errors, while CV-CAA peaks at smaller α and shows faster MTR growthfinding0.701
  Characterizes the differential sensitivity to injection strength between SAE and CAA methods
• The first principal component of persona space (the 'Assistant Axis') captures the assistant vs alternative persona distinction consistently across three models with role loading correlations exceeding 0.92finding0.689
  Finding establishing cross-model consistency of the assistant axis as the dominant structure in persona space
• Under a living process, as unfolding proceeds, the relative areas of these four components need to remain almost equal.claim0.685
  Invariant that living process maintains percentage balance.
• The contrast vector method is recommended over PC1 for reproducing the Assistant Axis in different models because it is not guaranteed that PC1 in every model will correspond to an Assistant Axisclaim0.681
  Practical methodological recommendation based on Llama 3.1 70B failure case