Upload tokenizer
e4f298e
|
{ |
|
"kaz": { |
|
"\t": 0, |
|
"\n": 1, |
|
"\r": 2, |
|
"#": 4, |
|
"&": 5, |
|
"(": 6, |
|
")": 7, |
|
"+": 8, |
|
"/": 9, |
|
"0": 10, |
|
"1": 11, |
|
"2": 12, |
|
"3": 13, |
|
"4": 14, |
|
"5": 15, |
|
"6": 16, |
|
"7": 17, |
|
"8": 18, |
|
"9": 19, |
|
"[": 20, |
|
"[PAD]": 140, |
|
"[UNK]": 139, |
|
"\\": 21, |
|
"]": 22, |
|
"_": 23, |
|
"a": 24, |
|
"b": 25, |
|
"c": 26, |
|
"e": 27, |
|
"f": 28, |
|
"i": 29, |
|
"k": 30, |
|
"m": 31, |
|
"n": 32, |
|
"o": 33, |
|
"p": 34, |
|
"r": 35, |
|
"s": 36, |
|
"t": 37, |
|
"y": 38, |
|
"z": 39, |
|
"{": 40, |
|
"|": 3, |
|
"}": 41, |
|
"~": 42, |
|
" ": 43, |
|
"¥": 44, |
|
"©": 45, |
|
"«": 46, |
|
"¬": 47, |
|
"": 48, |
|
"·": 49, |
|
"»": 50, |
|
"ë": 51, |
|
"ŋ": 52, |
|
"ɕ": 53, |
|
"ə": 54, |
|
"ɵ": 55, |
|
"ʨ": 56, |
|
"̀": 57, |
|
"́": 58, |
|
"̃": 59, |
|
"̆": 60, |
|
"̇": 61, |
|
"̈": 62, |
|
"̜": 63, |
|
"̟": 64, |
|
"̥": 65, |
|
"̩": 66, |
|
"̰": 67, |
|
"̶": 68, |
|
"̹": 69, |
|
"υ": 70, |
|
"а": 71, |
|
"б": 72, |
|
"в": 73, |
|
"г": 74, |
|
"д": 75, |
|
"е": 76, |
|
"ж": 77, |
|
"з": 78, |
|
"и": 79, |
|
"й": 80, |
|
"к": 81, |
|
"л": 82, |
|
"м": 83, |
|
"н": 84, |
|
"о": 85, |
|
"п": 86, |
|
"р": 87, |
|
"с": 88, |
|
"т": 89, |
|
"у": 90, |
|
"ф": 91, |
|
"х": 92, |
|
"ц": 93, |
|
"ч": 94, |
|
"ш": 95, |
|
"щ": 96, |
|
"ъ": 97, |
|
"ы": 98, |
|
"ь": 99, |
|
"э": 100, |
|
"ю": 101, |
|
"я": 102, |
|
"ё": 103, |
|
"і": 104, |
|
"ї": 105, |
|
"ў": 106, |
|
"ѳ": 107, |
|
"ғ": 108, |
|
"қ": 109, |
|
"ң": 110, |
|
"ү": 111, |
|
"ұ": 112, |
|
"һ": 113, |
|
"ӊ": 114, |
|
"ӏ": 115, |
|
"ә": 116, |
|
"ө": 117, |
|
"": 118, |
|
"": 119, |
|
"‐": 120, |
|
"‒": 121, |
|
"–": 122, |
|
"—": 123, |
|
"―": 124, |
|
"’": 125, |
|
"‚": 126, |
|
"„": 127, |
|
"…": 128, |
|
"№": 129, |
|
"−": 130, |
|
"─": 131, |
|
"●": 132, |
|
"☺": 133, |
|
"♥": 134, |
|
"⠀": 135, |
|
"": 136, |
|
"": 137, |
|
"-": 138 |
|
} |
|
} |
|
|