Skip to content

GitLab

Unverified Commit 9bd4872a authored by Trevor Morris's avatar Trevor Morris Committed by GitHub
Browse files

[bugfix] Fix typo in modelopt quant: 'FusedMoE' object has no attribute 'local_num_experts' (#8768)

parent 2fa0462c
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
python/sglang/srt/layers/quantization/modelopt_quant.py
View file @ 9bd4872a
...@@ -776,7 +776,7 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase): ...@@ -776,7 +776,7 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
# GEMM 1 # GEMM 1
w13_weight = ModelWeightParameter( w13_weight = ModelWeightParameter(
data=torch.empty( data=torch.empty(
layer.local_num_experts, layer.num_local_experts,
2 * intermediate_size_per_partition, 2 * intermediate_size_per_partition,
# 2 fp4 items are packed in the input dimension # 2 fp4 items are packed in the input dimension
hidden_size // 2, hidden_size // 2,
......
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