concept
active
concept:object-pronoun-upweightingObject pronoun upweighting
The other pathway in the 'her' subnetwork, where the verb 'lost' upweights object pronouns (including 'her').
Neighborhood — ranked by edge-count
Findings (1)
finding
- Attribution graph reveals a pathway that detects the verb 'lost' and upweights object pronounsassociated_withSecond component of the subnetwork for 'her', complementing the femaleness signal.
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.
- Loss balancing using homoscedastic uncertainty.
- Platform used for earlier human feedback data collection.
- Editing network weights to test predictions about circuit function; proposed as falsifiability test for circuit claims
- Baseline MTL approach minimizing sum of task losses with equal weights; suffers from task balancing
- The language into which Elephant programs are compiled, containing conventional data structures.
- Tuples are referenced by matching typed fields, akin to a relational database.
- Dynamic condition: increasing scale to assert importance.
- The phenomenon that separately trained models of the same architecture converge to the same basin and can be merged