low-resource-language-model-adaptation
/
adapted_tokenizers
/added-opt-hau
/opt_100-add_full-hau-opt
/tokenizer_config.json
| { | |
| "add_bos_token": true, | |
| "add_prefix_space": false, | |
| "added_tokens_decoder": { | |
| "1": { | |
| "content": "<pad>", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": true | |
| }, | |
| "2": { | |
| "content": "</s>", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": true | |
| }, | |
| "50265": { | |
| "content": " ku", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50266": { | |
| "content": " za", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50267": { | |
| "content": "yar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50268": { | |
| "content": " wan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50269": { | |
| "content": "sar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50270": { | |
| "content": "’", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50271": { | |
| "content": " yi", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50272": { | |
| "content": " kuma", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50273": { | |
| "content": "ikin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50274": { | |
| "content": " yan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50275": { | |
| "content": " shi", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50276": { | |
| "content": "uwa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50277": { | |
| "content": "yin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50278": { | |
| "content": " cikin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50279": { | |
| "content": " kar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50280": { | |
| "content": "ƙ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50281": { | |
| "content": " cewa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50282": { | |
| "content": " daga", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50283": { | |
| "content": " suka", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50284": { | |
| "content": "iyar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50285": { | |
| "content": "wam", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50286": { | |
| "content": " mai", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50287": { | |
| "content": " ji", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50288": { | |
| "content": " wannan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50289": { | |
| "content": " han", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50290": { | |
| "content": "dda", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50291": { | |
| "content": " wani", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50292": { | |
| "content": " sha", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50293": { | |
| "content": "ɗ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50294": { | |
| "content": "shin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50295": { | |
| "content": "ugaban", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50296": { | |
| "content": " kasa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50297": { | |
| "content": " kasar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50298": { | |
| "content": " ƙ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50299": { | |
| "content": " ci", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50300": { | |
| "content": " bayan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50301": { | |
| "content": "wamn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50302": { | |
| "content": " wanda", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50303": { | |
| "content": " tsa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50304": { | |
| "content": " masu", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50305": { | |
| "content": "anar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50306": { | |
| "content": "tsa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50307": { | |
| "content": " iya", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50308": { | |
| "content": " mata", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50309": { | |
| "content": "jeriya", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50310": { | |
| "content": " yana", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50311": { | |
| "content": " gaba", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50312": { | |
| "content": " dai", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50313": { | |
| "content": " zuwa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50314": { | |
| "content": " duk", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50315": { | |
| "content": " zai", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50316": { | |
| "content": " kai", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50317": { | |
| "content": " sai", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50318": { | |
| "content": " Najeriya", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50319": { | |
| "content": " jihar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50320": { | |
| "content": " gi", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50321": { | |
| "content": " tare", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50322": { | |
| "content": " shugaban", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50323": { | |
| "content": " wasu", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50324": { | |
| "content": " loka", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50325": { | |
| "content": " ‘", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50326": { | |
| "content": "wamnat", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50327": { | |
| "content": "anin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50328": { | |
| "content": " yadda", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50329": { | |
| "content": " suna", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50330": { | |
| "content": "yyan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50331": { | |
| "content": " kara", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50332": { | |
| "content": " mutan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50333": { | |
| "content": " matsa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50334": { | |
| "content": " “", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50335": { | |
| "content": "kwa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50336": { | |
| "content": " haka", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50337": { | |
| "content": " abin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50338": { | |
| "content": " tsar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50339": { | |
| "content": " wata", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50340": { | |
| "content": "yyana", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50341": { | |
| "content": " ake", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50342": { | |
| "content": " huk", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50343": { | |
| "content": " bayyana", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50344": { | |
| "content": " ranar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50345": { | |
| "content": " samu", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50346": { | |
| "content": "”", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50347": { | |
| "content": " kwa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50348": { | |
| "content": " inda", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50349": { | |
| "content": " biy", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50350": { | |
| "content": " wadan", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50351": { | |
| "content": "shen", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50352": { | |
| "content": " shekar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50353": { | |
| "content": " gwamnat", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50354": { | |
| "content": "bba", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50355": { | |
| "content": "kon", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50356": { | |
| "content": " lokacin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50357": { | |
| "content": " wajen", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50358": { | |
| "content": "iyyar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50359": { | |
| "content": " amf", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50360": { | |
| "content": " mutane", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50361": { | |
| "content": "kara", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50362": { | |
| "content": "ungiyar", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50363": { | |
| "content": " yin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50364": { | |
| "content": " kya", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| } | |
| }, | |
| "bos_token": "</s>", | |
| "clean_up_tokenization_spaces": true, | |
| "eos_token": "</s>", | |
| "errors": "replace", | |
| "model_max_length": 1000000000000000019884624838656, | |
| "pad_token": "<pad>", | |
| "tokenizer_class": "GPT2Tokenizer", | |
| "unk_token": "</s>" | |
| } | |