[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:
@@ -90,7 +90,6 @@ class JambaMoE(nn.Module):
|
||||
self.intermediate_size,
|
||||
tp_size=tp_size,
|
||||
params_dtype=params_dtype,
|
||||
reduce_results=True,
|
||||
renormalize=False,
|
||||
use_grouped_topk=False,
|
||||
quant_config=quant_config,
|
||||
|
||||
Reference in New Issue
Block a user