viswavi's picture
Add tokenizer to model (#4)
bacfce1
{"<</m>>": 30523, "<<m>>": 30522}