Let say I have large corpuses of texts, which I then tokenize and convert to List-of-Lists ? How would you store those LoL\'s in files and use them later for other tasks ?