{ | |
"added_tokens_decoder": { | |
"0": { | |
"content": "[UNK]", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"1": { | |
"content": "[CLS]", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"2": { | |
"content": "[SEP]", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"3": { | |
"content": "[PAD]", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"4": { | |
"content": "[MASK]", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"5": { | |
"content": "<as>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"6": { | |
"content": "<bd>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"7": { | |
"content": "<bn>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"8": { | |
"content": "<dg>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"9": { | |
"content": "<en>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"10": { | |
"content": "<gom>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"11": { | |
"content": "<gu>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"12": { | |
"content": "<hi>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"13": { | |
"content": "<kha>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"14": { | |
"content": "<kn>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"15": { | |
"content": "<ks>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"16": { | |
"content": "<mai>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"17": { | |
"content": "<ml>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"18": { | |
"content": "<mni>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"19": { | |
"content": "<mr>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"20": { | |
"content": "<ne>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"21": { | |
"content": "<or>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"22": { | |
"content": "<pa>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"23": { | |
"content": "<sa>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"24": { | |
"content": "<sd>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"25": { | |
"content": "<sat>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"26": { | |
"content": "<ta>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"27": { | |
"content": "<te>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"28": { | |
"content": "<ur>", | |
"lstrip": false, | |
"normalized": false, | |
"rstrip": false, | |
"single_word": false, | |
"special": true | |
}, | |
"250000": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250001": { | |
"content": "π¦", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250002": { | |
"content": "πΎ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250003": { | |
"content": "π₯²", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250004": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250005": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250006": { | |
"content": "π¦Ύ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250007": { | |
"content": "π€«", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250008": { | |
"content": "π±", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250009": { | |
"content": "πΈ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250010": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250011": { | |
"content": "π₯", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250012": { | |
"content": "π«‘", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250013": { | |
"content": "π°", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250014": { | |
"content": "π₯±", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250015": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250016": { | |
"content": "π₯΅", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250017": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250018": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250019": { | |
"content": "πͺ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250020": { | |
"content": "βͺ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250021": { | |
"content": "π₯Ά", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250022": { | |
"content": "π₯", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250023": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250024": { | |
"content": "β", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250025": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250026": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250027": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250028": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250029": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250030": { | |
"content": "π«", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250031": { | |
"content": "π¦¬", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250032": { | |
"content": "πΊ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250033": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250034": { | |
"content": "π¨", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250035": { | |
"content": "π«’", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250036": { | |
"content": "π€ ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250037": { | |
"content": "β", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250038": { | |
"content": "πΊ", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250039": { | |
"content": "π¦°", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250040": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250041": { | |
"content": "π«Ά", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250042": { | |
"content": "π", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250043": { | |
"content": "π°", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250044": { | |
"content": "π΅", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250045": { | |
"content": "π₯", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
}, | |
"250046": { | |
"content": "π€", | |
"lstrip": false, | |
"normalized": true, | |
"rstrip": false, | |
"single_word": false, | |
"special": false | |
} | |
}, | |
"clean_up_tokenization_spaces": false, | |
"cls_token": "[CLS]", | |
"mask_token": "[MASK]", | |
"model_max_length": 1000000000000000019884624838656, | |
"pad_token": "[PAD]", | |
"sep_token": "[SEP]", | |
"tokenizer_class": "PreTrainedTokenizerFast", | |
"unk_token": "[UNK]" | |
} | |