concept
active
concept:ginibre-matrix-distributionGinibre Matrix Distribution
A class of random matrices with Gaussian entries used to characterize the baseline eigenvalue distribution of OV circuits at initialization, against which learned positive eigenvalue clustering is compared
Neighborhood — ranked by edge-count
Methods (1)
method
- Eigenvalue-Based Copying Detectionassociated_withA summary statistic using positive eigenvalues of the OV circuit matrix to detect copying behavior in attention heads
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.
- Prescribes transitions among hidden state factors under action; encodes policy-dependent dynamics
- The distribution of latent representations produced by the model under unperturbed inputs
- Constraint in VPD where each parameter subcomponent is constrained to be a rank-one matrix for simplicity.
- The simple matrix form into which VPD constrains subcomponents to enforce mechanistic simplicity.
- Conjugate prior for categorical variables; used for beliefs about likelihood matrix A.
- Lashley's finding that memory is not localized but encoded globally; cited as an example of wholeness in neurophysiology.
- An OV circuit matrix that maps tokens to increasing the logit of those same tokens; detectable via positive eigenvalues
- The core idea of decomposing weight matrices into components for interpretability.