method
active
method:within-vs-across-event-cosine-similarityWithin vs across event cosine similarity
Quantifies event alignment of codes.
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- Event boundariesaboutDiscourse structure used as evaluation target.
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.
- Fig. 7c over 50 GPT-5 stories.
- Used to measure alignment between DIM direction and cone basis vectors to assess overlap
- Geometric evaluation of truth direction alignment across layers and prompt templates.
- Third empirical pattern.
- Even basic methods become unreliable.
- Cosine similarity between feature activations restricted to tokens where one of the features fires; used to identify feature splitting relationships
- Used to quantify the semantic clustering of adjective-set embeddings across model families and conditions
- Detects tiling in real SAEs.