32 lines
658 B
JSON
32 lines
658 B
JSON
{
|
|
"backend": "tokenizers",
|
|
"bos_token": "<s>",
|
|
"clean_up_tokenization_spaces": false,
|
|
"cls_token": "<CLS>",
|
|
"eos_token": "</s>",
|
|
"extra_special_tokens": [
|
|
"[bib_ref]",
|
|
"[/bib_ref]",
|
|
"[fig_ref]",
|
|
"[/fig_ref]",
|
|
"[bib]",
|
|
"[/bib]",
|
|
"[fig]",
|
|
"[/fig]",
|
|
"[table]",
|
|
"[/table]",
|
|
"[formula]",
|
|
"[/formula]"
|
|
],
|
|
"is_local": false,
|
|
"legacy": true,
|
|
"mask_token": "<MASK>",
|
|
"model_max_length": 1000000000000000019884624838656,
|
|
"pad_token": "</s>",
|
|
"sep_token": "<SEP>",
|
|
"sp_model_kwargs": {},
|
|
"tokenizer_class": "TokenizersBackend",
|
|
"unk_token": "<unk>",
|
|
"use_default_system_prompt": false
|
|
}
|