Paulo Villegas
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- StarCoder: may the source be with you!Saved credit: Contributor · 2023Explore people connected to this work →
- SantaCoder: don't reach for the stars!Saved credit: Contributor · 2023Explore people connected to this work →
- BERTIN: Efficient Pre-Training of a Spanish Language Model using Perplexity SamplingSaved credit: Contributor · 2022Explore people connected to this work →
- The ROOTS Search Tool: Data Transparency for LLMsSaved credit: Contributor · 2023Explore people connected to this work →
- The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual DatasetSaved credit: Contributor · 2022Explore people connected to this work →