Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
open-web-math-pro / README.md
koalazf99's picture
Update README.md
b246286 verified
metadata
license: odc-by
size_categories:
  - 1B<n<10B
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/*
task_categories:
  - text-generation
language:
  - en
pretty_name: open-web-math-pro
tags:
  - math
  - reasoning

πŸ“š Open-Web-Math-Pro

ArXiv | Models | Code

Open-Web-Math-Pro is refined from open-web-math using the ProX refining framework. It contains about 5B high quality math related tokens, ready for pre-training.

License

Open-Web-Math-Pro is based on open-web-math, which is made available under an ODC-By 1.0 license; users should also abide by the CommonCrawl ToU: https://commoncrawl.org/terms-of-use/. We do not alter the license of any of the underlying data.

Citation

@article{zhou2024programming,
  title={Programming Every Example: Lifting Pre-training Data Quality like Experts at Scale},
  author={Zhou, Fan and Wang, Zengzhi and Liu, Qian and Li, Junlong and Liu, Pengfei},
  journal={arXiv preprint arXiv:2409.17115},
  year={2024}
}