Interference Weights

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

• Feature Interferenceconcept0.781
  When non-orthogonal features cause logistic regression to identify a suboptimal probe direction
• Equal Weightingframework0.750
  Baseline MTL approach minimizing sum of task losses with equal weights; suffers from task balancing
• Task weightconcept0.731
  Coefficient weighting each task loss in the MTL objective.
• Weight Editingmethod0.722
  Editing network weights to test predictions about circuit function; proposed as falsifiability test for circuit claims
• cross-base interferenceconcept0.716
  Asymmetric transfer after fine-tuning: high-density bases (B10) are more robust.
• Untied Decoder Weightsconcept0.715
  Autoencoder design choice to learn separate encoder and decoder weights, increasing representational capacity by allowing encoder vectors to distinguish similar features
• Virtual Weightsconcept0.704
  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
• Weight spaceconcept0.702
  The space of the model's parameter matrices, where VPD operations take place.