view article Article Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models Mar 20 • 69
OpenCulture Collection A multilingual dataset of public domain books and newspapers. • 27 items • Updated Nov 6 • 121