Adds RULER benchmark #722
Open
Adds RULER benchmark #722
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
clefourrier
approved these changes
May 19, 2025
src/lighteval/tasks/default_tasks.py
Outdated
| @@ -24,6 +24,1238 @@ | |||
| from lighteval.tasks.lighteval_task import LightevalTaskConfig | |||
|
|
|||
|
|
|||
| ruler_niah_single_1_131072 = LightevalTaskConfig( | |||
There was a problem hiding this comment.
Imo we should keep the task names in alphabetical order to make the file easier to browse
NathanHB
commented
May 21, 2025
There was a problem hiding this comment.
Pull Request Overview
This PR integrates the RULER benchmark into the lighteval framework, expanding its evaluation capabilities.
- Implements a new prompt function for RULER in default_prompts.py
- Wraps task iteration with tqdm progress bars in lighteval_task.py
- Adds new RULER metrics in metrics.py and introduces a debugging breakpoint in vllm_model.py
Reviewed Changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| src/lighteval/tasks/lighteval_task.py | Adds tqdm import and progress bars to task iteration loops |
| src/lighteval/tasks/default_prompts.py | Adds a new RULER prompt function; contains a typo in the arc_with_options function |
| src/lighteval/models/vllm/vllm_model.py | Modifies logging format and introduces a breakpoint for debugging |
| src/lighteval/metrics/metrics.py | Adds new RULER metrics for evaluation |
| @@ -254,7 +265,7 @@ def arc_with_options(line, task_name: str = None): | |||
| query += "".join([f"\n{key}. {choice}" for key, choice in zip(LETTER_INDICES, line["choices"]["text"])]) | |||
| query += "\nAnswer:" | |||
| return Doc( | |||
| task_name=task_name, | |||
| mm task_name=task_name, | |||
There was a problem hiding this comment.
It appears that an extraneous 'mm' has been introduced before the task_name parameter. Please remove it to restore valid syntax.
Suggested change
| mm task_name=task_name, | |
| task_name=task_name, |
Copilot uses AI. Check for mistakes.
NathanHB
commented
May 21, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
https://github.com/NVIDIA/RULER
Available context size: 4096, 8192, 16384, 32768, 65536, 131072