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

Jump to bottom

希望支持评估模型数据泄漏的可能性 #867

Open

WSC741606 opened this issue May 6, 2024 · 0 comments

Labels

WSC741606 commented May 6, 2024

Describe the feature
在Eval评测数据集的同时，希望同时输出对所测试的模型已经训练过同个数据集的可能性的评估

Paste any useful information
https://github.com/swj0419/detect-pretrain-code-contamination/

Additional context
有助于确保模型调优效果的增加来源不是数据泄漏
类似https://github.com/abacusai/smaug ，会同时给出训练前后的泄漏评估对比

tastelikefeet added the enhancement label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment