Skip to content

GitLab

"examples/vscode:/vscode.git/clone" did not exist on "3b747de8452beed08df0397522920220f62fd7ff"
Unverified Commit 9dcdf5da authored by fzyzcjy's avatar fzyzcjy Committed by GitHub
Browse files

Tiny fix wrong comments (#9589)

parent f8b757bc
Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 2 deletions
python/sglang/srt/layers/quantization/modelopt_quant.py
View file @ 9dcdf5da
...@@ -876,7 +876,6 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase): ...@@ -876,7 +876,6 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
data=torch.empty( data=torch.empty(
layer.num_local_experts, layer.num_local_experts,
2 * intermediate_size_per_partition, 2 * intermediate_size_per_partition,
# 2 fp4 items are packed in the input dimension
hidden_size // self.quant_config.group_size, hidden_size // self.quant_config.group_size,
dtype=weight_scale_dtype, dtype=weight_scale_dtype,
), ),
...@@ -895,7 +894,6 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase): ...@@ -895,7 +894,6 @@ class ModelOptNvFp4FusedMoEMethod(FusedMoEMethodBase):
data=torch.empty( data=torch.empty(
layer.num_local_experts, layer.num_local_experts,
hidden_size, hidden_size,
# 2 fp4 items are packed in the input dimension
intermediate_size_per_partition // self.quant_config.group_size, intermediate_size_per_partition // self.quant_config.group_size,
dtype=weight_scale_dtype, dtype=weight_scale_dtype,
), ),
......
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