BWhat changed
Released a documented multi-source pretraining corpus and reproducible curation toolkit, with analyses of filtering, deduplication and source mixing for language-model research.
Scientific scope
Released a documented multi-source pretraining corpus and reproducible curation toolkit, with analyses of filtering, deduplication and source mixing for language-model research. Initial corpus release was August 2023; the January 2024 paper documents later corpus/toolkit versions and analyses. Distinct from OLMo's model framework; not a claim of bias-free, contamination-free or universally licensed data.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Initial corpus release was August 2023; the January 2024 paper documents later corpus/toolkit versions and analyses. Distinct from OLMo's model framework; not a claim of bias-free, contamination-free or universally licensed data.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
- Soldaini et al.: Dolma open pretraining corpus and curation toolkit — institutional/author source
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction