Model techniques map
Techniquesmodel architecturetoken mixersparse attentionsparse attention indexer

implementation detail · filed under model architecture

Index Branch output

The output of the Index Branch, which is added to the layer output.

Also called Index Branch value head output.

source
1
model
1
labs adopt it
0
strongest
not used

How sources treat it

One count per evidence span, weakest treatment to strongest.

not used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

the Index Branch output is added to the layer output

not usedmodel architecturein MiniMax Sparse AttentionMiniMax

Filed alongside

Other methods under model architecture :: token mixer :: sparse attention :: sparse attention indexer.