| { | |
| "version": "1.0", | |
| "truncation": null, | |
| "padding": null, | |
| "added_tokens": [ | |
| { | |
| "id": 177, | |
| "content": "<unk>", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| } | |
| ], | |
| "normalizer": { | |
| "type": "Sequence", | |
| "normalizers": [ | |
| { | |
| "type": "Lowercase" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "Regex": "[^_;:,.!?\u00a1\u00bf\u2014\u2026\"\u00ab\u00bb\u201c\u201d ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\u0251\u0250\u0252\u00e6\u0253\u0299\u03b2\u0254\u0255\u00e7\u0257\u0256\u00f0\u02a4\u0259\u0258\u025a\u025b\u025c\u025d\u025e\u025f\u0284\u0261\u0260\u0262\u029b\u0266\u0267\u0127\u0265\u029c\u0268\u026a\u029d\u026d\u026c\u026b\u026e\u029f\u0271\u026f\u0270\u014b\u0273\u0272\u0274\u00f8\u0275\u0278\u03b8\u0153\u0276\u0298\u0279\u027a\u027e\u027b\u0280\u0281\u027d\u0282\u0283\u0288\u02a7\u0289\u028a\u028b\u2c71\u028c\u0263\u0264\u028d\u03c7\u028e\u028f\u0291\u0290\u0292\u0294\u02a1\u0295\u02a2\u01c0\u01c1\u01c2\u01c3\u02c8\u02cc\u02d0\u02d1\u02bc\u02b4\u02b0\u02b1\u02b2\u02b7\u02e0\u02e4\u02de\u2193\u2191\u2192\u2197\u2198\u0329']" | |
| }, | |
| "content": "" | |
| }, | |
| { | |
| "type": "Strip", | |
| "strip_left": true, | |
| "strip_right": true | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "Regex": "(?=.)|(?<!^)$" | |
| }, | |
| "content": "_" | |
| } | |
| ] | |
| }, | |
| "pre_tokenizer": { | |
| "type": "Split", | |
| "pattern": { | |
| "Regex": "" | |
| }, | |
| "behavior": "Isolated", | |
| "invert": false | |
| }, | |
| "post_processor": null, | |
| "decoder": null, | |
| "model": { | |
| "vocab": { | |
| "_": 0, | |
| ";": 1, | |
| ":": 2, | |
| ",": 3, | |
| ".": 4, | |
| "!": 5, | |
| "?": 6, | |
| "\u00a1": 7, | |
| "\u00bf": 8, | |
| "\u2014": 9, | |
| "\u2026": 10, | |
| "\"": 11, | |
| "\u00ab": 12, | |
| "\u00bb": 13, | |
| "\u201c": 14, | |
| "\u201d": 15, | |
| " ": 16, | |
| "A": 17, | |
| "B": 18, | |
| "C": 19, | |
| "D": 20, | |
| "E": 21, | |
| "F": 22, | |
| "G": 23, | |
| "H": 24, | |
| "I": 25, | |
| "J": 26, | |
| "K": 27, | |
| "L": 28, | |
| "M": 29, | |
| "N": 30, | |
| "O": 31, | |
| "P": 32, | |
| "Q": 33, | |
| "R": 34, | |
| "S": 35, | |
| "T": 36, | |
| "U": 37, | |
| "V": 38, | |
| "W": 39, | |
| "X": 40, | |
| "Y": 41, | |
| "Z": 42, | |
| "a": 43, | |
| "b": 44, | |
| "c": 45, | |
| "d": 46, | |
| "e": 47, | |
| "f": 48, | |
| "g": 49, | |
| "h": 50, | |
| "i": 51, | |
| "j": 52, | |
| "k": 53, | |
| "l": 54, | |
| "m": 55, | |
| "n": 56, | |
| "o": 57, | |
| "p": 58, | |
| "q": 59, | |
| "r": 60, | |
| "s": 61, | |
| "t": 62, | |
| "u": 63, | |
| "v": 64, | |
| "w": 65, | |
| "x": 66, | |
| "y": 67, | |
| "z": 68, | |
| "\u0251": 69, | |
| "\u0250": 70, | |
| "\u0252": 71, | |
| "\u00e6": 72, | |
| "\u0253": 73, | |
| "\u0299": 74, | |
| "\u03b2": 75, | |
| "\u0254": 76, | |
| "\u0255": 77, | |
| "\u00e7": 78, | |
| "\u0257": 79, | |
| "\u0256": 80, | |
| "\u00f0": 81, | |
| "\u02a4": 82, | |
| "\u0259": 83, | |
| "\u0258": 84, | |
| "\u025a": 85, | |
| "\u025b": 86, | |
| "\u025c": 87, | |
| "\u025d": 88, | |
| "\u025e": 89, | |
| "\u025f": 90, | |
| "\u0284": 91, | |
| "\u0261": 92, | |
| "\u0260": 93, | |
| "\u0262": 94, | |
| "\u029b": 95, | |
| "\u0266": 96, | |
| "\u0267": 97, | |
| "\u0127": 98, | |
| "\u0265": 99, | |
| "\u029c": 100, | |
| "\u0268": 101, | |
| "\u026a": 102, | |
| "\u029d": 103, | |
| "\u026d": 104, | |
| "\u026c": 105, | |
| "\u026b": 106, | |
| "\u026e": 107, | |
| "\u029f": 108, | |
| "\u0271": 109, | |
| "\u026f": 110, | |
| "\u0270": 111, | |
| "\u014b": 112, | |
| "\u0273": 113, | |
| "\u0272": 114, | |
| "\u0274": 115, | |
| "\u00f8": 116, | |
| "\u0275": 117, | |
| "\u0278": 118, | |
| "\u03b8": 119, | |
| "\u0153": 120, | |
| "\u0276": 121, | |
| "\u0298": 122, | |
| "\u0279": 123, | |
| "\u027a": 124, | |
| "\u027e": 125, | |
| "\u027b": 126, | |
| "\u0280": 127, | |
| "\u0281": 128, | |
| "\u027d": 129, | |
| "\u0282": 130, | |
| "\u0283": 131, | |
| "\u0288": 132, | |
| "\u02a7": 133, | |
| "\u0289": 134, | |
| "\u028a": 135, | |
| "\u028b": 136, | |
| "\u2c71": 137, | |
| "\u028c": 138, | |
| "\u0263": 139, | |
| "\u0264": 140, | |
| "\u028d": 141, | |
| "\u03c7": 142, | |
| "\u028e": 143, | |
| "\u028f": 144, | |
| "\u0291": 145, | |
| "\u0290": 146, | |
| "\u0292": 147, | |
| "\u0294": 148, | |
| "\u02a1": 149, | |
| "\u0295": 150, | |
| "\u02a2": 151, | |
| "\u01c0": 152, | |
| "\u01c1": 153, | |
| "\u01c2": 154, | |
| "\u01c3": 155, | |
| "\u02c8": 156, | |
| "\u02cc": 157, | |
| "\u02d0": 158, | |
| "\u02d1": 159, | |
| "\u02bc": 160, | |
| "\u02b4": 161, | |
| "\u02b0": 162, | |
| "\u02b1": 163, | |
| "\u02b2": 164, | |
| "\u02b7": 165, | |
| "\u02e0": 166, | |
| "\u02e4": 167, | |
| "\u02de": 168, | |
| "\u2193": 169, | |
| "\u2191": 170, | |
| "\u2192": 171, | |
| "\u2197": 172, | |
| "\u2198": 173, | |
| "null": 174, | |
| "\u0329": 175, | |
| "'": 176, | |
| "<unk>": 177 | |
| } | |
| } | |
| } |