Model techniques map
Techniques

specific method · not yet filed

Global attention with a dense feed-forward network in the first Transformer block

Also called global attention with a dense FFN.

source
1
model
1
lab adopt it
1
strongest
core

How sources treat it

One count per evidence span, weakest treatment to strongest.

core 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

The first Transformer block uses global attention with a dense FFN.

coremodel architecturein MiMo-V2.6-Pro-RLXiaomi