chrisvinsen's picture
add tokenizer
161855f
raw
history blame
282 Bytes
{"b": 0, "o": 1, "e": 2, "l": 3, ",": 4, "s": 5, "a": 6, "!": 7, "w": 8, "u": 9, "f": 10, "d": 11, "g": 12, "i": 13, "p": 14, "é": 15, "z": 16, "n": 17, "v": 19, "k": 20, "j": 21, "x": 22, "y": 23, "c": 24, "m": 25, "h": 26, "t": 27, "r": 28, "|": 18, "[UNK]": 29, "[PAD]": 30}