bertweetbr / tokenizer_config.json
Fernando Carneiro
Set max_size=128
c941901
raw
history blame
230 Bytes
{"normalization": false, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "max_len": 128, "tokenizer_class": "BertweetTokenizer"}