Vector-Geometry Features for Screening

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

• Geometry of featuresconcept0.806
  Research thread within About Blank concerning the structure and relational properties of neural network feature representations; covariance pooling tangentially supports this thread.
• Geometry of features matters for representation quality.claim0.786
  General principle supported tangentially by covariance pooling work; relates to feature co-occurrence structure.
• Feature Visualizationmethod0.769
  Method of optimizing input to cause a neuron to fire maximally, used to characterize what a neuron detects; establishes causal link
• F-statistics and Linear Probes for Feature Selectionmethod0.767
  Method to select d_steer top-activated SAE features for constructing control vectors
• Geometryconcept0.765
  The actual shapes and spatial relationships of buildings, essential to living structure.
• geometry-based steeringconcept0.753
  Paradigm of finding the right geometry (manifold) for principled control.
• Vector-based navigation using grid-like representations in artificial agents (Banino et al., 2018)concept0.749
  Demonstrated grid cell emergence in RNNs trained on spatial navigation; related work category 4.
• Feature Manifoldsframework0.748
  Hypothesized extension of superposition where features may be higher-dimensional manifolds rather than 1D directions