luganda-ner-v4 / tokenizer_config.json
Conrad747's picture
Training in progress, epoch 1
9bc29c7
raw
history blame
435 Bytes
{
"add_prefix_space": true,
"bos_token": "[CLS]",
"cls_token": "[CLS]",
"do_lower_case": false,
"eos_token": "[SEP]",
"mask_token": "[MASK]",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "[PAD]",
"sep_token": "[SEP]",
"sp_model_kwargs": {},
"special_tokens_map_file": null,
"split_by_punct": false,
"tokenizer_class": "DebertaV2Tokenizer",
"unk_token": "[UNK]",
"vocab_type": "spm"
}