Open Models Almanac
Methodsmodel architecturetoken mixersparse attentionfixed-pattern sparse attention

single mention · model architecture

Sparse Global Attention Anchors

Sparse global attention anchors are used in a hybrid Mamba-2 stack; the evidence does not specify their selection mechanism.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See fixed-pattern sparse attention for the methods filed beside it.

Filed under model architecture › token mixer › sparse attention › fixed-pattern sparse attention