{ "version": "1.0", "truncation": null, "padding": null, "added_tokens": [ { "id": 0, "content": "[PAD]", "single_word": false, "lstrip": false, "rstrip": false, "normalized": false, "special": true }, { "id": 1, "content": "[BOS]", "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": { "type": "TemplateProcessing", "single": [ { "Sequence": { "id": "A", "type_id": 0 } } ], "pair": [ { "Sequence": { "id": "A", "type_id": 0 } }, { "Sequence": { "id": "B", "type_id": 1 } } ], "special_tokens": {} }, "decoder": { "type": "BPEDecoder", "suffix": "" }, "model": { "type": "BPE", "dropout": null, "unk_token": "bloop", "continuing_subword_prefix": null, "end_of_word_suffix": null, "fuse_unk": false, "byte_fallback": false, "ignore_merges": false, "vocab": { "[PAD]": 0, "[BOS]": 1, "[EOS]": 2, "b": 3, "g": 4, "i": 5, "l": 6, "o": 7, "p": 8, "u": 9, "bl": 10, "bu": 11, "ub": 12, "blo": 13, "gl": 14, "lb": 15, "blub": 16, "bi": 17, "op": 18, "ulb": 19, "bll": 20, "blu": 21, "bub": 22, "bubl": 23, "bulb": 24, "blop": 25, "bloop": 26, "glu": 27, "glub": 28, "blubb": 29, "biulb": 30 }, "merges": [ [ "b", "l" ], [ "b", "u" ], [ "u", "b" ], [ "bl", "o" ], [ "g", "l" ], [ "l", "b" ], [ "bl", "ub" ], [ "b", "i" ], [ "o", "p" ], [ "u", "lb" ], [ "bl", "l" ], [ "bl", "u" ], [ "bu", "b" ], [ "bu", "bl" ], [ "bu", "lb" ], [ "blo", "p" ], [ "blo", "op" ], [ "gl", "u" ], [ "gl", "ub" ], [ "blub", "b" ], [ "bi", "ulb" ] ] } }