jerryyan/dclm-pool-400m-1x-student-cleaned-rescued

DCLM 400M-1x, cleaned by a small model Pre-training corpora built from the DCLM dclm-pool-400m-1x pool (a 10% shard sample). Boilerplate is stripped and low-value pages are dropped; what is published here is the result, already tokenized with EleutherAI/gpt-neox-20b and ready to feed straight into DCLM's training.train. Read this first: two different page sets The corpora here do not all cover the same set of pages, so their token counts are not interchangeable.… See the full description on the dataset page: https://huggingface.co/datasets/jerryyan/dclm-pool-400m-1x-student-cleaned-rescued.

<iframe src="https://silytics.si/embed/item/7af90076-470e-43a7-b035-288a7fc6a7d5" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

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 Information may be incomplete or out of date.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.