finding
active
finding:helpful-and-harmful-weights-are-scattered-across-the-entire-range-of-virtual-weight-magnitudesHelpful and harmful weights are scattered across the entire range of virtual weight magnitudes
IN Tokens→Logits shows weak VW-helpfulness relationship.
Source paper
extracted_fromNicholas L. Turner · Jeffrey Wu · Joshua Batson
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Large virtual weights are not inherently misleading but not guaranteed to be helpful or effectivesupportsCaution for practitioners.
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.
- Three regimes by effectiveness.
- Motivating example.
- Implicit weights directly connecting any pair of layers computed by multiplying output weights of one layer with input weights of another through the residual stream
- Used to project features through QK weights and find aligned features.
- Interpretation of motivating example.
- Reads loss change directly.
- Estimates density floor.
- Discussion.