biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[LoRA] Cleanup LoRA unused code (#29611)

Browse Source 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
This commit is contained in:
Jee Jee Li
2025-11-29 14:52:58 +08:00
committed by GitHub
parent 4a80ad0a25
commit 39e63dec7c
46 changed files with 126 additions and 173 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/model_executor/models/transformers/base.py
View File

@@ -93,7 +93,6 @@ ALL_ATTENTION_FUNCTIONS["vllm"] = vllm_flash_attention_forward
class Base(nn.Module, VllmModel, SupportsQuant, SupportsLoRA, SupportsPP):
embedding_padding_modules = ["lm_head"]
embedding_modules = ["embed_tokens"] # TODO transformers will have a util to get it
hf_to_vllm_mapper = WeightsMapper(
orig_to_new_prefix={