-
Notifications
You must be signed in to change notification settings - Fork 250
Issues: stanford-crfm/helm
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
Installation Guide: Required Dependencies for Supported Operating Systems and Distributions
user question
#3160
opened Nov 14, 2024 by
shakatoday
Issue with running HEIM
documentation
Improvements or additions to documentation
HEIM (Text2Image)
user question
#3080
opened Oct 22, 2024 by
sudhir-mcw
How to use this package when I have the pompts and images across models
HEIM (Text2Image)
user question
#3062
opened Oct 14, 2024 by
snehith3195
Not able to install install-heim-extras.sh for heim leaderboard
HEIM (Text2Image)
user question
#3060
opened Oct 11, 2024 by
snehith3195
HatefulMemesScenario get_instances returning error
bug
Something isn't working
HEIM (Text2Image)
methodology
Evaluation methodology
user question
#3056
opened Oct 10, 2024 by
dxwu2
Add MMLU-Pro
additions
New models or scenarios
good first issue
Good for newcomers
scenarios
#3018
opened Sep 24, 2024 by
yifanmai
Add amazon bedrock 3p models to the model configuration.
user question
#2967
opened Sep 3, 2024 by
subhaviv
Incorrect scoring due to answer format mismatch in MMLU evaluation
user question
#2939
opened Aug 16, 2024 by
DerryChan
Add sorting/filtering functionality to predictions pages
frontend
Frontend issues
#2887
opened Aug 1, 2024 by
farzaank
HEIMHumanEvalScenario requires permissions to download data from codalab
HEIM (Text2Image)
user question
#2865
opened Jul 30, 2024 by
slymane
Sort by correctness functionality on Predictions page in front end
#2783
opened Jun 29, 2024 by
farzaank
Don't set trust_remote_code=True by default in HuggingFaceClient
#2645
opened May 13, 2024 by
yifanmai
helm-summarize gives extra weight to runs in multiple groups
framework
methodology
Evaluation methodology
p2
Priority 2 (Good to have for release)
#2541
opened Apr 2, 2024 by
yifanmai
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.