Skip to content

GitLab

Commit 4a62a3eb authored by zhuwenwen's avatar zhuwenwen
Browse files

remove topk_ids

parent 58bbb720
Show whitespace changes
Inline Side-by-side
Showing with 0 additions and 5 deletions
vllm/model_executor/layers/fused_moe/fused_moe.py
View file @ 4a62a3eb
......@@ -652,7 +652,6 @@ def invoke_fused_moe_kernel(A: torch.Tensor,
B_scale: Optional[torch.Tensor],
B_zp: Optional[torch.Tensor],
topk_weights: Optional[torch.Tensor],
topk_ids: torch.Tensor,
sorted_token_ids: torch.Tensor,
expert_ids: torch.Tensor,
num_tokens_post_padded: torch.Tensor,
......@@ -1690,7 +1689,6 @@ def fused_experts_impl(
w1_scale,
w1_zp,
curr_topk_weights,
curr_topk_ids,
sorted_token_ids,
expert_ids,
num_tokens_post_padded,
......@@ -1730,7 +1728,6 @@ def fused_experts_impl(
w2_scale,
w2_zp,
curr_topk_weights,
curr_topk_ids,
sorted_token_ids,
expert_ids,
num_tokens_post_padded,
......@@ -2017,7 +2014,6 @@ class TritonExperts(mk.FusedMoEPermuteExpertsUnpermute):
w1_scale,
w1_zp,
None,
topk_ids,
sorted_token_ids,
expert_ids,
num_tokens_post_padded,
......@@ -2049,7 +2045,6 @@ class TritonExperts(mk.FusedMoEPermuteExpertsUnpermute):
w2_scale,
w2_zp,
None,
topk_ids,
sorted_token_ids,
expert_ids,
num_tokens_post_padded,
......
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