Skip to content

[DeepSeekNextN] fix: residual of head norm can be None #7398

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
Jun 20, 2025
Merged

[DeepSeekNextN] fix: residual of head norm can be None #7398

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
6 changes: 5 additions & 1 deletion python/sglang/srt/models/deepseek_nextn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,7 +107,11 @@ def forward(
)

if not forward_batch.forward_mode.is_idle():
hidden_states, _ = self.shared_head.norm(hidden_states, residual)
if residual is not None:
hidden_states, _ = self.shared_head.norm(hidden_states, residual)
else:
hidden_states = self.shared_head.norm(hidden_states)
Comment on lines +110 to +113
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Jun 20, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The conditional logic can be simplified to be more concise. Instead of calling self.shared_head.norm in two different branches, call it once and handle the output based on its type. The RMSNorm.forward method can handle residual being None.

output = self.shared_head.norm(hidden_states, residual)
hidden_states = output[0] if isinstance(output, tuple) else output


return hidden_states


Expand Down
Loading