grc-alignment / tokenizer_config.json
TariqYousef
commit from TY
f58077f
raw
history blame
401 Bytes
{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "xlm-roberta-base", "tokenizer_class": "XLMRobertaTokenizer", "full_tokenizer_file": "/content/gdrive/MyDrive/ugaBert/xlm-roberta-base-finetuned-full_12M_10epp/tokenizer.json"}