Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Bugfix] Fix layer skip logic with bitsandbytes #9887

Merged
merged 2 commits into from
Nov 1, 2024
Merged

[Bugfix] Fix layer skip logic with bitsandbytes #9887

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7747f1d
Fix layer skip logic with bitsandbytes
mgoin Oct 31, 2024
2275ced
format
mgoin Oct 31, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 6 additions & 1 deletion vllm/model_executor/layers/quantization/bitsandbytes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -119,7 +119,12 @@ def get_scaled_act_names(self) -> List[str]:


def is_layer_skipped_bnb(prefix: str, llm_int8_skip_modules: List[str]):
return any(module_name in prefix for module_name in llm_int8_skip_modules)
# Split the prefix into its dot-separated components
components = prefix.split('.')

# Check if any of the skip modules exactly matches any component
return any(module_name in components
for module_name in llm_int8_skip_modules)


class BitsAndBytesLinearMethod(LinearMethodBase):
Expand Down