Spaces:
Running
on
CPU Upgrade
Model failed to evaluate
Hi,
Thanks for this great leaderboard 🤗!
I have 2 models that failed to evaluate (with the status "FAILED") and got the "This model has been already submitted." error when I tried to re-submit them.
https://huggingface.co/datasets/open-llm-leaderboard/requests/blob/main/bofenghuang/vigogne-2-13b-instruct_eval_request_False_float16_Original.json
https://huggingface.co/datasets/open-llm-leaderboard/requests/blob/main/bofenghuang/vigogne-33b-instruct_eval_request_False_float16_Original.json
How could I re-submit them? And any ideas that they failed, please? They are of the same architecture as other models that I've succeeded to submit such as bofenghuang/vigogne-2-7b-instruct.
Hi!
Thanks for your interest!
Your vigogne-2-13b
model failed because it could not be read (OSError: Unable to load weights from pytorch checkpoint file for '.../models--bofenghuang--vigogne-2-13b-instruct/snapshots/5cdd81383316cc77a9368cab7096241caf7f4556/pytorch_model-00014-of-00014.bin'
) - could you try converting your model to safetensors and resubmitting?
The vigogne-33b
got an OOM error, this is likely to be a problem on our side.
I removed your models from requests so you can submit them again.