Finished but no result

#1010
by ymcki - opened

Previously, I reported three jobs that ends with empty benchmark scores:
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard/discussions/1009

I found that two of them now have results of proper scores. That's good.

However, one of them while finished doesn't show up in the results.
https://huggingface.co/datasets/open-llm-leaderboard/requests/blob/main/ymcki/gemma-2-2b-ORPO-jpn-it-abliterated-18-merge_eval_request_False_bfloat16_Original.json
No result here:
https://huggingface.co/datasets/open-llm-leaderboard/results/tree/main/ymcki/gemma-2-2b-ORPO-jpn-it-abliterated-18-merge

Since this job is submitted already, I can't submit it again. Can someone re-submit for me or somehow dig out the disappeared result?

Thanks a lot in advance.

I can see it in the board.

Open LLM Leaderboard org

Hi @ymcki ,

No worries, everything is correct now! We recalculated all the models that had incomplete result files and indeed you can find this your model ymcki/gemma-2-2b-ORPO-jpn-it-abliterated-18-merge on the Leaderboard. Usually, it takes a couple of hours for the Leaderboard to restart and pull all new models from the Contents dataset

I'm closing this issue, feel free to ping me in case of any other questions about your missing models, or please open a new disucssion

alozowski changed discussion status to closed

Sign up or log in to comment