find_tr2 / vocab.json
pnparam's picture
Upload tokenizer
bc4e17c
raw
history blame
265 Bytes
{
"[PAD]": 23,
"[UNK]": 22,
"a": 7,
"b": 15,
"c": 16,
"d": 10,
"e": 5,
"f": 13,
"g": 4,
"h": 9,
"i": 6,
"k": 11,
"l": 12,
"m": 17,
"n": 14,
"o": 21,
"p": 8,
"r": 1,
"s": 3,
"t": 19,
"u": 0,
"w": 18,
"y": 2,
"|": 20
}