Spaces:
Running
on
CPU Upgrade
Open LLM Leaderboard Results Dataset
Hi,
I am looking at the OpenLLM Results dataset but the rows do not seem to be loading. https://huggingface.co/datasets/open-llm-leaderboard/results. I also could not find detailed documentation of what these results are. I am looking for the more detailed results, like for benchmark subtasks, of all of the models on the leaderboard.
Thank you!
Hi
@natalialanzoni
!
Sorry, I missed your issue! Have you taken a look at our doc page?
You should be able to download the results dataset using the datasets library, to browse it locally, or to explore the individual files here online. Models are stored in org/model/run_name.json, and if you open one file (like here for example), you'll see that all the subtasks results are reported.
Feel free to reopen or comment again if you need further indications!