Skip to content

Reorder loop in shared expert weight loading #5719

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 26, 2025
Merged

Reorder loop in shared expert weight loading #5719

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions python/sglang/srt/models/deepseek_nextn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,11 +215,11 @@ def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
"up_proj.weight_scale_inv",
]
names_to_remove = []
for num_repeat in range(self.n_share_experts_fusion):
for suffix in suffix_list:
shared_expert_weight_name = (
f"model.layers.0.mlp.shared_experts.{suffix}"
)
for suffix in suffix_list:
shared_expert_weight_name = (
f"model.layers.0.mlp.shared_experts.{suffix}"
)
for num_repeat in range(self.n_share_experts_fusion):
weights_list.append(
(
f"model.layers.0."
Expand All @@ -229,7 +229,7 @@ def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
weights_dict[shared_expert_weight_name],
)
)
names_to_remove += [shared_expert_weight_name]
names_to_remove += [shared_expert_weight_name]
weights = [w for w in weights_list if w[0] not in names_to_remove]

# Params for weights, fp8 weight scales, fp8 activation scales
Expand Down
12 changes: 6 additions & 6 deletions python/sglang/srt/models/deepseek_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1650,11 +1650,11 @@ def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
desc=f"Cloning {self.n_share_experts_fusion} "
"replicas of the shared expert into MoE",
):
for num_repeat in range(self.n_share_experts_fusion):
for suffix in suffix_list:
shared_expert_weight_name = (
f"model.layers.{moe_layer}.mlp.shared_experts.{suffix}"
)
for suffix in suffix_list:
shared_expert_weight_name = (
f"model.layers.{moe_layer}.mlp.shared_experts.{suffix}"
)
for num_repeat in range(self.n_share_experts_fusion):
weights_list.append(
(
f"model.layers.{moe_layer}."
Expand All @@ -1664,7 +1664,7 @@ def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
weights_dict[shared_expert_weight_name],
)
)
names_to_remove += [shared_expert_weight_name]
names_to_remove += [shared_expert_weight_name]
weights = [w for w in weights_list if w[0] not in names_to_remove]

# Params for weights, fp8 weight scales, fp8 activation scales
Expand Down
Loading