We follow the [open-instruct](https://github.com/allenai/open-instruct?tab=readme-ov-file#dataset-preparation) repo to prepare hour instruction tuning datasets. In our project, we utilize a combination of four training datasets: Flan v2, COT, Dolly, and Open Assistant. For the purposes of evaluation, we employ three additional datasets: MMLU, Tydiqa, and BBH. A processed version of these files will be made available [here] [TODO].
We follow the [open-instruct](https://github.com/allenai/open-instruct?tab=readme-ov-file#dataset-preparation) repo to prepare hour instruction tuning datasets. In our project, we utilize a combination of four training datasets: Flan v2, COT, Dolly, and Open Assistant. For the purposes of evaluation, we employ three additional datasets: MMLU, Tydiqa, and BBH. A processed version of these files are available [here](https://huggingface.co/datasets/princeton-nlp/less_data).
