bpe_tokenizer.json

{
  "version": "1.0",
  "truncation": null,
  "padding": null,
  "added_tokens": [
    {
      "id": 0,
      "content": "[UNK]",
      "single_word": false,
      "lstrip": false,
      "rstrip": false,
      "normalized": false,
      "special": true
    },
    {
      "id": 1,
      "content": "[PAD]",
      "single_word": false,
      "lstrip": false,
      "rstrip": false,
      "normalized": false,
      "special": true
    },
    {
      "id": 2,
      "content": "[EOS]",
      "single_word": false,
      "lstrip": false,
      "rstrip": false,
      "normalized": false,
      "special": true
    }
  ],
  "normalizer": null,
  "pre_tokenizer": {
    "type": "Whitespace"
  },
  "post_processor": null,
  "decoder": null,
  "model": {
    "type": "BPE",
    "dropout": null,
    "unk_token": "[UNK]",
    "continuing_subword_prefix": null,
    "end_of_word_suffix": null,
    "fuse_unk": false,
    "byte_fallback": false,
    "ignore_merges": false,
    "vocab": {
      "[UNK]": 0,
      "[PAD]": 1,
      "[EOS]": 2,
      "a": 3,
      "d": 4,
      "e": 5,
      "f": 6,
      "g": 7,
      "h": 8,
      "i": 9,
      "l": 10,
      "n": 11,
      "o": 12,
      "r": 13,
      "s": 14,
      "t": 15,
      "u": 16,
      "v": 17,
      "w": 18,
      "x": 19,
      "y": 20,
      "nd": 21,
      "and": 22,
      "ty": 23,
      "th": 24,
      "ou": 25,
      "re": 26,
      "sand": 27,
      "thou": 28,
      "thousand": 29,
      "hu": 30,
      "ndre": 31,
      "hundre": 32,
      "hundred": 33,
      "en": 34,
      "ne": 35,
      "ev": 36,
      "even": 37,
      "ei": 38,
      "gh": 39,
      "eigh": 40,
      "tw": 41,
      "seven": 42,
      "fi": 43,
      "ix": 44,
      "six": 45,
      "ine": 46,
      "nine": 47,
      "rty": 48,
      "eight": 49,
      "ve": 50,
      "fou": 51,
      "four": 52,
      "two": 53,
      "one": 54,
      "five": 55,
      "thre": 56,
      "three": 57,
      "thi": 58,
      "fif": 59,
      "sixty": 60,
      "fo": 61,
      "forty": 62,
      "enty": 63,
      "twenty": 64,
      "seventy": 65,
      "thirty": 66,
      "fifty": 67,
      "eighty": 68,
      "ninety": 69,
      "een": 70,
      "teen": 71,
      "el": 72,
      "twel": 73,
      "twelve": 74,
      "rteen": 75,
      "thirteen": 76,
      "fifteen": 77,
      "nineteen": 78,
      "seventeen": 79,
      "eleven": 80,
      "fourteen": 81,
      "eighteen": 82,
      "ten": 83,
      "sixteen": 84
    },
    "merges": [
      [
        "n",
        "d"
      ],
      [
        "a",
        "nd"
      ],
      [
        "t",
        "y"
      ],
      [
        "t",
        "h"
      ],
      [
        "o",
        "u"
      ],
      [
        "r",
        "e"
      ],
      [
        "s",
        "and"
      ],
      [
        "th",
        "ou"
      ],
      [
        "thou",
        "sand"
      ],
      [
        "h",
        "u"
      ],
      [
        "nd",
        "re"
      ],
      [
        "hu",
        "ndre"
      ],
      [
        "hundre",
        "d"
      ],
      [
        "e",
        "n"
      ],
      [
        "n",
        "e"
      ],
      [
        "e",
        "v"
      ],
      [
        "ev",
        "en"
      ],
      [
        "e",
        "i"
      ],
      [
        "g",
        "h"
      ],
      [
        "ei",
        "gh"
      ],
      [
        "t",
        "w"
      ],
      [
        "s",
        "even"
      ],
      [
        "f",
        "i"
      ],
      [
        "i",
        "x"
      ],
      [
        "s",
        "ix"
      ],
      [
        "i",
        "ne"
      ],
      [
        "n",
        "ine"
      ],
      [
        "r",
        "ty"
      ],
      [
        "eigh",
        "t"
      ],
      [
        "v",
        "e"
      ],
      [
        "f",
        "ou"
      ],
      [
        "fou",
        "r"
      ],
      [
        "tw",
        "o"
      ],
      [
        "o",
        "ne"
      ],
      [
        "fi",
        "ve"
      ],
      [
        "th",
        "re"
      ],
      [
        "thre",
        "e"
      ],
      [
        "th",
        "i"
      ],
      [
        "fi",
        "f"
      ],
      [
        "six",
        "ty"
      ],
      [
        "f",
        "o"
      ],
      [
        "fo",
        "rty"
      ],
      [
        "en",
        "ty"
      ],
      [
        "tw",
        "enty"
      ],
      [
        "seven",
        "ty"
      ],
      [
        "thi",
        "rty"
      ],
      [
        "fif",
        "ty"
      ],
      [
        "eigh",
        "ty"
      ],
      [
        "nine",
        "ty"
      ],
      [
        "e",
        "en"
      ],
      [
        "t",
        "een"
      ],
      [
        "e",
        "l"
      ],
      [
        "tw",
        "el"
      ],
      [
        "twel",
        "ve"
      ],
      [
        "r",
        "teen"
      ],
      [
        "thi",
        "rteen"
      ],
      [
        "fif",
        "teen"
      ],
      [
        "nine",
        "teen"
      ],
      [
        "seven",
        "teen"
      ],
      [
        "el",
        "even"
      ],
      [
        "four",
        "teen"
      ],
      [
        "eight",
        "een"
      ],
      [
        "t",
        "en"
      ],
      [
        "six",
        "teen"
      ]
    ]
  }
}