Skip to content

Allow for model kwargs when loading transformers from pretrained #754

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

Merged
merged 4 commits into from
May 21, 2025
Merged

Allow for model kwargs when loading transformers from pretrained #754

Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
31b59c0
add way to pass kwargs when loading transformers model
NathanHB May 20, 2025
a2c4dc7
add way to pass kwargs when loading transformers model
NathanHB May 20, 2025
5eeb8fa
Update src/lighteval/models/transformers/transformers_model.py
NathanHB May 20, 2025
6ff02c6
suggestion from copilot
NathanHB May 20, 2025
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: 3 additions & 4 deletions src/lighteval/models/transformers/transformers_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,8 +137,8 @@ class TransformersModelConfig(ModelConfig):
subfolder: str | None = None
revision: str = "main"
batch_size: PositiveInt | None = None
generation_size: PositiveInt = 256
max_length: PositiveInt | None = None
model_loading_kwargs: dict = {}
add_special_tokens: bool = True
model_parallel: bool | None = None
dtype: str | None = None
Expand Down Expand Up @@ -384,9 +384,8 @@ def _create_auto_model(self) -> transformers.PreTrainedModel:

pretrained_config = self.transformers_config

kwargs = {}
if "quantization_config" not in pretrained_config.to_dict():
kwargs["quantization_config"] = quantization_config
self.config.model_loading_kwargs["quantization_config"] = quantization_config
Copy link
Preview

Copilot AI May 20, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Mutating config.model_loading_kwargs in place can lead to unexpected state if _create_auto_model is called multiple times. Consider merging into a local dict and passing that to from_pretrained.

Copilot uses AI. Check for mistakes.


model = AutoModelForCausalLM.from_pretrained(
self.config.model_name,
Expand All @@ -395,7 +394,7 @@ def _create_auto_model(self) -> transformers.PreTrainedModel:
device_map=device_map,
torch_dtype=torch_dtype,
trust_remote_code=self.config.trust_remote_code,
**kwargs,
**self.config.model_loading_kwargs,
)
# model.to(self.device)
model.eval()
Expand Down
Loading