{ "additional_special_tokens": [ { "content": "ar_AR", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "cs_CZ", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "de_DE", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "en_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "es_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "et_EE", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "fi_FI", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "fr_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "gu_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "hi_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "it_IT", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ja_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "kk_KZ", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ko_KR", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "lt_LT", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "lv_LV", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "my_MM", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ne_NP", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "nl_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ro_RO", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ru_RU", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "si_LK", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "tr_TR", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "vi_VN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "zh_CN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "af_ZA", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "az_AZ", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "bn_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "fa_IR", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "he_IL", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "hr_HR", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "id_ID", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ka_GE", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "km_KH", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "mk_MK", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ml_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "mn_MN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "mr_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "pl_PL", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ps_AF", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "pt_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "sv_SE", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "sw_KE", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ta_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "te_IN", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "th_TH", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "tl_XX", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "uk_UA", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "ur_PK", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "xh_ZA", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "gl_ES", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "sl_SI", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "bn_NG", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, { "content": "lg_YY", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false } ], "bos_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "cls_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "eos_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "mask_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "pad_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "sep_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false }, "unk_token": { "content": "", "lstrip": false, "normalized": false, "rstrip": false, "single_word": false } }