Description

Large text dataset, classically used for next word prediction.