Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Benchmark links
#60
by
oiwio
- opened
May you share the link to the evaluation benchmark that is capable of running ko-ARC, ko-Hellaswag. and ko-CommonGen?
Hi. For Ko-H5, we didn't release the test data, so there is no link to evaluate it.
However, for English, you can evaluate it through lm-evaluation-harness and there is a Korean benchmark similar to Ko-h5 at https://huggingface.co/datasets/HAERAE-HUB/KMMLU