low-resource-language-model-adaptation
/
adapted_tokenizers
/added-opt-yor
/opt_100-add_full-yor-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 | |
| }, | |
| "43998": { | |
| "content": "ì", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50096": { | |
| "content": "ò", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50101": { | |
| "content": "ù", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50265": { | |
| "content": "ọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50266": { | |
| "content": "ẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50267": { | |
| "content": "ọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50268": { | |
| "content": "wọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50269": { | |
| "content": "ṣ", | |
| "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": " awọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50272": { | |
| "content": " ọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50273": { | |
| "content": " ṣ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50274": { | |
| "content": " gb", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50275": { | |
| "content": " ati", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50276": { | |
| "content": "́", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50277": { | |
| "content": "lẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50278": { | |
| "content": " ṣe", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50279": { | |
| "content": " ẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50280": { | |
| "content": "̀", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50281": { | |
| "content": " wọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50282": { | |
| "content": " lati", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50283": { | |
| "content": "gba", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50284": { | |
| "content": "mọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50285": { | |
| "content": "ṣẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50286": { | |
| "content": "kọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50287": { | |
| "content": " ì", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50288": { | |
| "content": "jọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50289": { | |
| "content": " ní", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50290": { | |
| "content": " jẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50291": { | |
| "content": "Ọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50292": { | |
| "content": " rẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50293": { | |
| "content": "lọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50294": { | |
| "content": "gbe", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50295": { | |
| "content": "gbo", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50296": { | |
| "content": " lọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50297": { | |
| "content": " naa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50298": { | |
| "content": "pọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50299": { | |
| "content": "wọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50300": { | |
| "content": "rọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50301": { | |
| "content": " Ọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50302": { | |
| "content": "ún", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50303": { | |
| "content": "àn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50304": { | |
| "content": " sọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50305": { | |
| "content": " tó", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50306": { | |
| "content": "ṣe", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50307": { | |
| "content": " tí", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50308": { | |
| "content": " pẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50309": { | |
| "content": " àwọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50310": { | |
| "content": "rí", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50311": { | |
| "content": " ohun", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50312": { | |
| "content": " ń", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50313": { | |
| "content": " ò", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50314": { | |
| "content": "rẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50315": { | |
| "content": " pẹlu", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50316": { | |
| "content": " lẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50317": { | |
| "content": " pé", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50318": { | |
| "content": " gba", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50319": { | |
| "content": " ile", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50320": { | |
| "content": " ninu", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50321": { | |
| "content": " lori", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50322": { | |
| "content": " gbo", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50323": { | |
| "content": "sẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50324": { | |
| "content": "tọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50325": { | |
| "content": " è", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50326": { | |
| "content": " Awọn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50327": { | |
| "content": " yoo", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50328": { | |
| "content": "yin", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50329": { | |
| "content": "áa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50330": { | |
| "content": " ọmọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50331": { | |
| "content": " gbogbo", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50332": { | |
| "content": " lá", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50333": { | |
| "content": " nipa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50334": { | |
| "content": " mọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50335": { | |
| "content": " ló", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50336": { | |
| "content": "gbẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50337": { | |
| "content": " sí", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50338": { | |
| "content": "dun", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50339": { | |
| "content": " ara", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50340": { | |
| "content": "bẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50341": { | |
| "content": " yii", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50342": { | |
| "content": "”", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50343": { | |
| "content": " mẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50344": { | |
| "content": " máa", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50345": { | |
| "content": " iṣẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50346": { | |
| "content": "Ẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50347": { | |
| "content": "áà", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50348": { | |
| "content": " ó", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50349": { | |
| "content": "yẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50350": { | |
| "content": " ala", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50351": { | |
| "content": " “", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50352": { | |
| "content": "ìn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50353": { | |
| "content": "iṣẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50354": { | |
| "content": "Ṣ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50355": { | |
| "content": " kọ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50356": { | |
| "content": "gbà", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50357": { | |
| "content": "ùn", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50358": { | |
| "content": "tẹ", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50359": { | |
| "content": "ọna", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50360": { | |
| "content": " tabi", | |
| "lstrip": false, | |
| "normalized": true, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": false | |
| }, | |
| "50361": { | |
| "content": " Ì", | |
| "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>" | |
| } | |