mattbonnell's picture
Upload tokenizer
f9c363c verified
raw
history blame
507 Bytes
{
"[PAD]": 43,
"[UNK]": 42,
"a": 29,
"b": 19,
"d": 33,
"e": 35,
"f": 39,
"h": 14,
"i": 15,
"j": 1,
"k": 34,
"l": 9,
"m": 13,
"n": 21,
"o": 40,
"p": 24,
"s": 0,
"t": 26,
"u": 36,
"v": 12,
"w": 41,
"z": 5,
"|": 23,
"æ": 6,
"ð": 32,
"ŋ": 17,
"ɐ": 4,
"ɑ": 30,
"ɔ": 22,
"ə": 18,
"ɚ": 7,
"ɛ": 2,
"ɜ": 31,
"ɡ": 38,
"ɪ": 11,
"ɹ": 10,
"ɾ": 37,
"ʃ": 28,
"ʊ": 8,
"ʌ": 3,
"ʒ": 20,
"ː": 25,
"θ": 16,
"ᵻ": 27
}