Skip to content

Adds RULER benchmark #722

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

Open
wants to merge 11 commits into
base: main
Choose a base branch
from
Open

Adds RULER benchmark #722

Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
65275d5
adds RULE
NathanHB May 15, 2025
0ef15e3
adds RULE
NathanHB May 15, 2025
9cafd75
adds RULE
NathanHB May 15, 2025
ed3d907
use llama 3.2 no chat template
NathanHB May 19, 2025
a4394ad
Merge branch 'main' into nathan-adds-helet
NathanHB May 21, 2025
248bb67
Update src/lighteval/tasks/default_prompts.py
NathanHB May 21, 2025
a1aee68
Update src/lighteval/models/vllm/vllm_model.py
NathanHB May 21, 2025
775705c
Merge branch 'main' into nathan-adds-helet
NathanHB May 28, 2025
461b8cb
fix typo
eliebak Jun 2, 2025
57f2921
put tuler in extedned tasks
NathanHB Jun 18, 2025
79e6a6e
added params for Nouamane
clefourrier Jun 23, 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
4 changes: 4 additions & 0 deletions src/lighteval/models/vllm/vllm_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,6 +105,8 @@ class VLLMModelConfig(ModelConfig):
max_num_batched_tokens: PositiveInt = 2048 # maximum number of tokens per batch
subfolder: str | None = None
is_async: bool = False # Whether to use the async version or sync version of the model
use_dual_chunk_attention: bool = False
Copy link
Member Author

Choose a reason for hiding this comment

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

what version of vllm are you using for this ? I get TypeError: EngineArgs.__init__() got an unexpected keyword argument 'use_dual_chunk_attention' with vllm == 0.8.5.post1

Copy link
Member

Choose a reason for hiding this comment

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

I was on 0.9.1 I think

Copy link
Member

Choose a reason for hiding this comment

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

(changed my env to same as you now)

enforce_eager: bool = False


class VLLMModel(LightevalModel):
Expand Down Expand Up @@ -187,6 +189,8 @@ def _create_auto_model(self, config: VLLMModelConfig) -> Optional[LLM]:
"seed": int(config.seed),
"max_num_seqs": int(config.max_num_seqs),
"max_num_batched_tokens": int(config.max_num_batched_tokens),
"enforce_eager": config.enforce_eager,
"use_dual_chunk_attention": config.use_dual_chunk_attention,
}

if config.quantization is not None:
Expand Down
Loading