method
active
method:ocean-trait-covariance-matrix-mOCEAN Trait Covariance Matrix M
5x5 Pearson correlation matrix of OCEAN traits computed from MDS injection sweeps to assess cross-trait leakage
Neighborhood — ranked by edge-count
Concepts (1)
concept
- Cross-Trait LeakageimplementsUnintended movement of non-target OCEAN traits when steering toward a target trait; quantified via lambda metric
Claims (1)
claim
- Interpretive conclusion from Big Two mismatch finding; tentative due to only 46.15% match rate
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Testing five phrasings of the self-referential prompt to confirm robustness to wording variation
- Novel aggregation technique replacing mean pooling; preserves joint activation structure (feature co-occurrence) in token embeddings.
- Five variants of the experimental prompt tested to confirm the effect is robust to changes in specific wording
- Supported by qualitative experiments showing fluent and coherent steering for three additional models
- Validates the statement synthesis pipeline as producing behavior-specific content comparable to established methods
- Cao & Yamins principle: solution set for an easy goal is large, for a challenging goal comparatively smaller; cited as theoretical basis for multitask scaling hypothesis
- Property where a rule learned on fixed-size grid generalizes to larger grids, observed in checkerboard and lizard experiments