ted_tokenizer / tokenizer_config.json
Lakoc's picture
Upload tokenizer
e7001cc
raw
history blame contribute delete
414 Bytes
{
"bos_token": "<s>",
"cls_token": "<s>",
"do_lower_case": false,
"eos_token": "</s>",
"mask_token": "[MASK]",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<pad>",
"sep_token": "</s>",
"sp_model_kwargs": {
"alpha": 0.1,
"enable_sampling": true,
"nbest_size": -1
},
"split_by_punct": false,
"tokenizer_class": "DebertaV2Tokenizer",
"unk_token": "<unk>"
}