can we use these parameter subcomponents to perform clean, targeted changes?

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Parameter subcomponents cleanly isolate true mechanisms of the modelclaim0.787
  Interpretive claim that the subcomponents correspond to real functional units.
• Parameter subcomponentconcept0.772
  One of the simple rank-one matrices resulting from VPD that sums with others to reconstruct the original model weights and has a specific functional role.
• A good parameter subcomponent is causally important only for specific roles and can be removed from the model without hurting performance on irrelevant promptsclaim0.760
  Definitional principle guiding VPD: subcomponents should encode narrow, targeted computational roles rather than distributed, multi-purpose functionality.
• are the resulting parameter subcomponents actually interpretable objects?question0.756
  First question posed after applying VPD, investigating whether the subcomponents make sense.
• Model editing via direct subcomponent overwritemethod0.750
  Technique to alter model behavior by directly editing a parameter subcomponent without training, demonstrated by changing an emoticon eye subcomponent.
• Direct model editing via parameter subcomponent modification—emoticon eye recognition altered to predict shocked faces with no retrainingfinding0.741
  Demonstrated that VPD-discovered subcomponents encode true computational machinery by enabling targeted, predictable behavior changes without gradient-based training.
• Attention computations distribute across heads via parameter subcomponents with interpretable rolesfinding0.736
  Mechanistic discovery about how attention mechanisms decompose into interpretable parameter components.
• Parameters did not change; the binding did.quote0.732
  Emphasizes that anchoring is a binding process without weight updates.