keyi10's picture
add tokenizer
4dd1cda
{"s": 0, "z": 1, "v": 2, "d": 3, "f": 4, "i": 5, "j": 6, "x": 7, "e": 8, "r": 9, "k": 11, "u": 12, "w": 13, "y": 14, "t": 15, "g": 16, "c": 17, "o": 18, "p": 19, "q": 20, "a": 21, "h": 22, "l": 23, "n": 24, "b": 25, "m": 26, "|": 10, "[UNK]": 27, "[PAD]": 28}