Datasets:

Modalities:
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
RedPajama-pro / README.md
koalazf99's picture
Update README.md
6856fe4 verified
|
raw
history blame
933 Bytes
---
license: apache-2.0
task_categories:
- text-generation
language:
- en
tags:
- web
- common crawl
size_categories:
- 10B<n<100B
---
# 📚 RedPajama-pro
<p align="center">
<img src="prox-teaser.png">
</p>
[ArXiv](http://arxiv.org/abs/xxxx) | [Models](https://huggingface.co/collections/gair-prox/prox-general-models-65f1674f0607712c4d6eec76) | [Code](https://github.com/GAIR-NLP/ProX)
RedPajama-pro is refined from [RedPajama-Data-V2](https://huggingface.co/datasets/togethercomputer/RedPajama-Data-V2) using the **ProX** refining framework.
It contains about 30B high quality tokens, ready for general language model pre-training.
## License
RedPajama-pro is based on RedPajama-Data-V2, which is made available under an apache-2.0 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
```
@misc{TBD
}
```