Skip to content

GitLab

Unverified Commit 707b240d authored by Xin Yang's avatar Xin Yang Committed by GitHub
Browse files

[Bugfix] Fix FusedMoE LoRA w2_output_size (#31949) 


Signed-off-by: default avatarXin Yang <xyangx@amazon.com>
parent 29ce4822
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
vllm/lora/layers/fused_moe.py
View file @ 707b240d
...@@ -720,7 +720,7 @@ class FusedMoE3DWithLoRA(FusedMoEWithLoRA): ...@@ -720,7 +720,7 @@ class FusedMoE3DWithLoRA(FusedMoEWithLoRA):
""" """
Full size Full size
""" """
return self.w2_lora_a_stacked[0].shape[-2] return self.base_layer.hidden_size
@classmethod @classmethod
def can_replace_layer( def can_replace_layer(
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment