hypothesis
active
hypothesis:if-all-of-2-retained-weights-were-interpretable-the-result-would-be-closer-to-an-upstairs-liftIf all of ~2% retained weights were interpretable, the result would be closer to an 'upstairs' lift
Speculative possibility.
Source paper
extracted_fromNicholas L. Turner · Jeffrey Wu · Joshua Batson
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (1)
finding
- Sampled helpfulness mass.
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.
- Emphasizes the importance of full-scale physical judgment over scaled drawings.
- Estimates density floor.
- Battery does not detect epistemic humility alone; contemplative prompt does something distinct
- Empirical validation of the theory of centers in architecture.
- Load-bearing claim about the tractability of circuit analysis; central thesis of Claim 2
- Statement about the client's satisfaction after anxious delays.
- Do we assume that the model knows something we don't, or are these interference weights contributing noise?question0.716Motivating question.
- Large virtual weights are not inherently misleading but not guaranteed to be helpful or effectiveclaim0.716Caution for practitioners.