Unverified Commit 63dc3426 authored by Jee Jee Li's avatar Jee Jee Li Committed by GitHub
Browse files

[Model] Add packed_modules_mapping for Qwen3-MOE (#18118)


Signed-off-by: default avatarJee Jee Li <pandaleefree@gmail.com>
parent 8f5dc414
...@@ -475,6 +475,17 @@ class Qwen3MoeModel(nn.Module): ...@@ -475,6 +475,17 @@ class Qwen3MoeModel(nn.Module):
class Qwen3MoeForCausalLM(nn.Module, SupportsPP): class Qwen3MoeForCausalLM(nn.Module, SupportsPP):
packed_modules_mapping = {
"qkv_proj": [
"q_proj",
"k_proj",
"v_proj",
],
"gate_up_proj": [
"gate_proj",
"up_proj",
],
}
fall_back_to_pt_during_load = False fall_back_to_pt_during_load = False
......
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