mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-04 16:22:57 +08:00
fix arguement error (#3030)
This commit is contained in:
@@ -85,8 +85,8 @@ class MoEMethodBase(QuantMethodBase):
|
||||
layer.top_k,
|
||||
layer.hidden_size,
|
||||
layer.num_experts,
|
||||
layer.moe_config.num_max_dispatch_tokens_per_rank,
|
||||
layer.fd_config.parallel_config.splitwise_role,
|
||||
layer.fd_config.model_config.num_max_dispatch_tokens_per_rank,
|
||||
layer.ep_size,
|
||||
layer.ep_rank,
|
||||
layer.fd_config.model_config.redundant_experts_num,
|
||||
|
Reference in New Issue
Block a user