NX2411's picture
add tokenizer
2b3e824
raw
history blame
268 Bytes
{"i": 1, "d": 2, "t": 3, "j": 4, "n": 5, "s": 6, "g": 7, "a": 8, "x": 9, "m": 10, "y": 11, "z": 12, "l": 13, "w": 14, "k": 15, "u": 16, "f": 17, "o": 18, "v": 19, "'": 20, "h": 21, "e": 22, "p": 23, "b": 24, "r": 25, "c": 26, "q": 27, "|": 0, "[UNK]": 28, "[PAD]": 29}