A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 17:00 UTC
[Bugfix] Fix weight loading issue by rename variable. (#8293)
Wenxiang committed
Sep 13, 2024 at 07:25 UTC
a480939e8e3b8e5b5571531c30212a1a947ee32e
1 file changed
+1
-1
vllm/model_executor/models/phimoe.py
+1
-1
@@ -600,7 +600,7 @@ class PhiMoEForCausalLM(nn.Module, SupportsLoRA):
600
weight_loader(
601
param,
602
loaded_weight,
603
- weight_name,
603
+ name,
604
shard_id=shard_id,
605
expert_id=expert_id,
606
)