Investigating eval request failure

#495
by SanjiWatsuki - opened

Hey all!

I was trying to investigate https://huggingface.co/datasets/open-llm-leaderboard/requests/blob/main/SanjiWatsuki/Loyal-Toppy-Bruins-Maid-7B-DARE_eval_request_False_bfloat16_Original.json

I'm able to load it up locally with transformers and I ran some of the lm_eval harness. It seems like it was working. Are there any logs I could inspect to figure out what's wrong?

Cheers!

Open LLM Leaderboard org

Hi,
Thanks for pointing to the request file of your model - we have a connectivity issue with our eval cluster atm so the leaderboard has been put on hold, and all models which failed in the last 2 days because of this will be relaunched as soon as it's fixed.

Thanks for the response! I'll hold on tight, good luck debugging the issues 🫂

SanjiWatsuki changed discussion status to closed

Sign up or log in to comment