ambiguous · filed under model architecture
Routed experts
Refers to experts selected for a token, but the evidence does not establish a distinct routing algorithm.
Also called 192 routed experts (top-8).
- 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.
192 routed experts (top-8) + 1 shared expert
coremodel architecturein Hy3Tencent
Filed alongside
Other methods under model architecture :: channel mixer :: mixture of experts :: expert routing.
Top-8 expert routingToken-level expert routingAnticipatory RoutingTop-4 expert routingDP-aware routingFixed Top-k routing with frozen biasHash routingKeep RoutingLatent-space routingLoss-spike-triggered Anticipatory RoutingToken-choice routingToken-choice routing with softplus gatingTop-6 expert routingTop-k expert routing with softmax over selected experts