method
active
method:hellinger-distanceHellinger distance
Metric used to define geometric space of output token probability distributions in behavior manifold analysis.
Neighborhood — ranked by edge-count
Concepts (2)
concept
- behavior manifoldimplementsOne-dimensional curved surface in output probability space; the paper shows this mirrors representation manifold structure.
- Behavior SpaceusesA geometric space of all output token probability distributions, equipped with Hellinger distance, used to visualize model behavior.
Related by similarity (4)
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.
- Primary quantitative measure of distributional divergence between natural and intervened representations
- Measure of how far cells travel during development; increased by stress sharing.
- Alternative alignment metric compared in appendix; computes edit distance between nearest neighbor lists