Direct model editing via parameter subcomponent modification—emoticon eye recognition altered to predict shocked faces with no retraining

{
  "doc": "paper.md",
  "context": "Demonstrated that VPD-discovered subcomponents encode true computational machinery by enabling targeted, predictable behavior changes without gradient-based training.",
  "norm_label": "Direct model editing via parameter subcomponent modification—emoticon eye recognition altered to predict shocked faces with no retraining",
  "graphify_id": "emoticon_editing",
  "source_file": "paper.md",
  "imported_from": "/Users/antonborzov/Documents/Research.nosync/papers/extract_typed_out/paper/graph.json",
  "extracted_type": "finding",
  "source_location": "§2.2",
  "graphify_file_type": "finding"
}