- Put your MMLU data folder in
/mmlu/data - (Optional) Create a virtual env (
python -m venv myenv,source myenv/bin/activate) - Run the
gen_mmlu.pyfirst, and then look what the JSON that's generated is called and use that withineval_mmlu.py
ahumay/calico
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|