[MoE Refactor] Move the shared/fused expert output sum into MoERunnerBase (#35949)
Signed-off-by: Bill Nell <bnell@redhat.com> Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
This commit is contained in:
@@ -349,7 +349,6 @@ class Gemma4MoE(nn.Module):
|
||||
"moe_intermediate_size",
|
||||
getattr(config, "expert_intermediate_size", None),
|
||||
),
|
||||
reduce_results=True,
|
||||
renormalize=True,
|
||||
quant_config=quant_config,
|
||||
prefix=f"{prefix}.experts",
|
||||
|
||||
Reference in New Issue
Block a user