jihyun's picture
Upload tokenizer
4e972ff
raw
history blame
397 Bytes
{
"'": 31,
"1": 7,
"2": 26,
"3": 22,
"4": 5,
"5": 2,
"[": 1,
"[PAD]": 35,
"[UNK]": 34,
"]": 23,
"a": 6,
"b": 29,
"c": 18,
"d": 8,
"e": 16,
"f": 28,
"g": 25,
"h": 32,
"i": 4,
"j": 14,
"k": 13,
"l": 10,
"m": 30,
"n": 33,
"o": 12,
"p": 27,
"q": 3,
"r": 20,
"s": 11,
"t": 9,
"u": 0,
"w": 15,
"x": 24,
"y": 19,
"z": 17,
"|": 21
}