1 line
463 KiB
JSON
1 line
463 KiB
JSON
|
{"version":"1.0","truncation":{"max_length":512,"strategy":"LongestFirst","stride":0},"padding":{"strategy":"BatchLongest","direction":"Right","pad_to_multiple_of":null,"pad_id":0,"pad_type_id":0,"pad_token":"[PAD]"},"added_tokens":[{"id":0,"special":true,"content":"[PAD]","single_word":false,"lstrip":false,"rstrip":false,"normalized":false},{"id":101,"special":true,"content":"[UNK]","single_word":false,"lstrip":false,"rstrip":false,"normalized":false},{"id":102,"special":true,"content":"[CLS]","single_word":false,"lstrip":false,"rstrip":false,"normalized":false},{"id":103,"special":true,"content":"[SEP]","single_word":false,"lstrip":false,"rstrip":false,"normalized":false},{"id":104,"special":true,"content":"[MASK]","single_word":false,"lstrip":false,"rstrip":false,"normalized":false}],"normalizer":{"type":"BertNormalizer","clean_text":true,"handle_chinese_chars":true,"strip_accents":null,"lowercase":false},"pre_tokenizer":{"type":"BertPreTokenizer"},"post_processor":{"type":"TemplateProcessing","single":[{"SpecialToken":{"id":"[CLS]","type_id":0}},{"Sequence":{"id":"A","type_id":0}},{"SpecialToken":{"id":"[SEP]","type_id":0}}],"pair":[{"SpecialToken":{"id":"[CLS]","type_id":0}},{"Sequence":{"id":"A","type_id":0}},{"SpecialToken":{"id":"[SEP]","type_id":0}},{"Sequence":{"id":"B","type_id":1}},{"SpecialToken":{"id":"[SEP]","type_id":1}}],"special_tokens":{"[CLS]":{"id":"[CLS]","ids":[102],"tokens":["[CLS]"]},"[SEP]":{"id":"[SEP]","ids":[103],"tokens":["[SEP]"]}}},"decoder":{"type":"WordPiece","prefix":"##","cleanup":true},"model":{"type":"WordPiece","unk_token":"[UNK]","continuing_subword_prefix":"##","max_input_chars_per_word":100,"vocab":{"[PAD]":0,"unused0":1,"unused1":2,"unused2":3,"unused3":4,"unused4":5,"unused5":6,"unused6":7,"unused7":8,"unused8":9,"unused9":10,"unused10":11,"unused11":12,"unused12":13,"unused13":14,"unused14":15,"unused15":16,"unused16":17,"unused17":18,"unused18":19,"unused19":20,"unused20":21,"unused21":22,"unused22":23,"unused23":24,"unused24":25,"unused25":26,"unused26":27,"unused27":28,"unused28":29,"unused29":30,"unused30":31,"unused31":32,"unused32":33,"unused33":34,"unused34":35,"unused35":36,"unused36":37,"unused37":38,"unused38":39,"unused39":40,"unused40":41,"unused41":42,"unused42":43,"unused43":44,"unused44":45,"unused45":46,"unused46":47,"unused47":48,"unused48":49,"unused49":50,"unused50":51,"unused51":52,"unused52":53,"unused53":54,"unused54":55,"unused55":56,"unused56":57,"unused57":58,"unused58":59,"unused59":60,"unused60":61,"unused61":62,"unused62":63,"unused63":64,"unused64":65,"unused65":66,"unused66":67,"unused67":68,"unused68":69,"unused69":70,"unused70":71,"unused71":72,"unused72":73,"unused73":74,"unused74":75,"unused75":76,"unused76":77,"unused77":78,"unused78":79,"unused79":80,"unused80":81,"unused81":82,"unused82":83,"unused83":84,"unused84":85,"unused85":86,"unused86":87,"unused87":88,"unused88":89,"unused89":90,"unused90":91,"unused91":92,"unused92":93,"unused93":94,"unused94":95,"unused95":96,"unused96":97,"unused97":98,"unused98":99,"unused99":100,"[UNK]":101,"[CLS]":102,"[SEP]":103,"[MASK]":104,"d":105,"##on":106,"c":107,"##er":108,"s":109,"p":110,"a":111,"##to":112,"##re":113,"##en":114,"##ta":115,"##ti":116,"##in":117,"##la":118,"##el":119,"di":120,"##ri":121,"##ra":122,"##li":123,"##te":124,"##le":125,"e":126,"m":127,"##un":128,"##ia":129,"n":130,"##ro":131,"##an":132,"##ion":133,"i":134,"##ss":135,"del":136,"f":137,"##he":138,"in":139,"##or":140,"un":141,"##lo":142,"v":143,"##zion":144,"##om":145,"la":146,"##qu":147,"##os":148,"##ar":149,"##ic":150,"S":151,"C":152,"con":153,"t":154,"##es":155,"per":156,"al":157,"che":158,"g":159,"##io":160,"qu":161,"il":162,"##al":163,"##men":164,"##do":165,"##is":166,"##zione":167,"##et":168,"##no":169,"##ca":170,"su":171,"##co":172,"P":173,"##ma":174,"ri":175,"##ere":176,"o":177,"L":178,"-":179,"A":180,"l":181,"b":182,"##gli":183,"I":184,"##mo":185,"##za":186,"##va":187,"de":188,"M":189,"N":190,"##ci":191,"##ono":192,"##ol":193,"##ce":194,"##di":195,"com":196,"pro":197,"è":198,"le":199,"##tu":200,"##tto":201,"1":202,"da"
|