-
Notifications
You must be signed in to change notification settings - Fork 558
Refactormc2 #2164
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.
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
Refactormc2 #2164
Conversation
|
👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:
If CI fails, you can run linting and testing checks locally according Contributing and Testing. |
vllm_ascend/ops/fused_moe.py
Outdated
| hidden_states = torch.cat(hidden_states, dim=0) | ||
| return hidden_states | ||
|
|
||
| w1 = w1.transpose(1, 2) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Duplicate code?
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
|
|
||
| return final_hidden_states | ||
|
|
||
| class QuantizedTokenDispatcherWithAllGather(MoETokenDispatcher): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
fused_experts_with_allgather is not transferred here. when expert_map is not None, this is really slow
What this PR does / why we need it?
Does this PR introduce any user-facing change?
How was this patch tested?