add tokenizer

vocab.json

@@ -1 +1 @@
-{"'": 0, "m": 1, "s": 2, "h": 3, "f": 4, "k": 5, "y": 6, "v": 7, "l": 9, "t": 10, "q": 11, "n": 12, "p": 13, "i": 14, "w": 15, "r": 16, "z": 17, "e": 18, "d": 19, "j": 20, "b": 21, "g": 22, "o": 23, "x": 24, "c": 25, "u": 26, "a": 27, "|": 8, "[UNK]": 28, "[PAD]": 29}
+{"g": 0, "k": 1, "b": 2, "m": 3, "z": 4, "w": 5, "y": 6, "i": 7, "h": 8, "n": 9, "f": 10, "'": 11, "d": 12, "x": 13, "o": 14, "t": 16, "e": 17, "s": 18, "c": 19, "l": 20, "p": 21, "q": 22, "u": 23, "a": 24, "j": 25, "v": 26, "r": 27, "|": 15, "[UNK]": 28, "[PAD]": 29}