language: - en - de
A sampled version of the CCMatrix dataset for the German-English pair, containing 1M train entries.
