Spaces:
Running
on
CPU Upgrade
FAQ - v2
Please feel free to ask all your questions here
I remember this was previously a v1 leaderboard, and now it has been replaced with a v2 version. I have updated my model and would like to submit it to the v2 leaderboard. However, it shows 'has already been evaluated (status FINISHED)'. Is there any way to update submissions for both v1 and v2 leaderboards? This is really important to me!
I remember this was previously a v1 leaderboard, and now it has been replaced with a v2 version. I have updated my model and would like to submit it to the v2 leaderboard. However, it shows 'has already been evaluated (status FINISHED)'. Is there any way to update submissions for both v1 and v2 leaderboards? This is really important to me!
I would greatly appreciate your prompt response to this message. This is an extremely urgent and important matter for me. Thank you in advance for your assistance!
I saw that the benchmarks used are here: https://github.com/huggingface/lighteval/blob/main/examples/tasks/OALL_v2_tasks.txt
And that the results are here: https://huggingface.co/datasets/OALL/v2_results/blob/main
But I can't find the script that runs the evals anywhere. Could you point me to the script that runs all of the evals and saves it to v2_results?