Skip to content

GitLab

Unverified Commit 3d962d72 authored by Kyungmin Lee's avatar Kyungmin Lee Committed by GitHub
Browse files

[BugFix] fix FusedMoE.make_expert_params_mapping in EXAONE-MoE (#32196) 


Signed-off-by: default avatarlkm2835 <lkm2835@gmail.com>
parent 20228cb8
Show whitespace changes
Inline Side-by-side
Showing with 1 addition and 0 deletions
vllm/model_executor/models/exaone_moe.py
View file @ 3d962d72
......@@ -338,6 +338,7 @@ class ExaoneMoeModel(nn.Module):
# Params for weights, fp8 weight scales, fp8 activation scales
# (param_name, weight_name, expert_id, shard_id)
return FusedMoE.make_expert_params_mapping(
self,
ckpt_gate_proj_name="gate_proj",
ckpt_down_proj_name="down_proj",
ckpt_up_proj_name="up_proj",
......
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