gair-prox/DCLM-pro
π DCLM-pro ArXiv | Models | Code DCLM-pro is refined from DCLM using the ProX refining framework. It contains about >500B high quality tokens, ready for general language model pre-training. License DCLM-pro is based on DCLM, which is made available under an cc-by-4.0 license. Citation @article{zhou2024programming, title={Programming Every Example: Lifting Pre-training Data Quality like Experts at Scale}, author={Zhou, Fan and Wangβ¦ See the full description on the dataset page: https://huggingface.co/datasets/gair-prox/DCLM-pro.
| Author / organization | gair-prox |
|---|---|
| License (as declared) | odc-by |
| Downloads | 2.2K (as reported by Hugging Face) |
| Likes | 13 |
| Created | 14 February 2025 |
| Last updated at source | 15 February 2025 |
Popularity, last 3 days
2,232 now (-9 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: π DCLM-pro ArXiv | Models | Code DCLM-pro is refined from DCLM using the ProX refining framework. It contains about >500B high quality tokens, ready for general language model pre-training. License DCLM-pro is based on DCLM, which is made β π DCLM-pro ArXiv | Models | Code DCLM-pro is refined from DCLM using the ProX refining framework. It contains about >500B high quality tokens, ready for general language model pre-training. License DCLM-pro is based on DCLM, which is made
Changes detected by Silytics between refreshes of the source.
More from gair-prox
- gair-prox/open-web-math-proDataset
- gair-prox/FineWeb-proDataset
- gair-prox/c4-proDataset
- gair-prox/RedPajama-proDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.