Model techniques map
Techniquesmodel architecturetoken mixersoftmax attentiongrouped-query attention

ambiguous · filed under model architecture

Softplus-based per-head gating

A per-head attention gating mechanism using softplus; the evidence does not establish a more specific name or mechanism.

Also called Softplus-based per-head attention gating.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

and softplus-based per-head gating

usedmodel architecturein Laguna XS.2Poolside

Filed alongside

Other methods under model architecture :: token mixer :: softmax attention :: grouped-query attention.