same-story / tokenizer_config.json
emilys's picture
initial commit
46f5f7c verified
raw
history blame
463 Bytes
{
"bos_token": "<s>",
"cls_token": "<s>",
"do_lower_case": true,
"eos_token": "</s>",
"mask_token": "<mask>",
"model_max_length": 512,
"name_or_path": "/home/silcock/.cache/torch/sentence_transformers/sentence-transformers_all-mpnet-base-v2/",
"pad_token": "<pad>",
"sep_token": "</s>",
"special_tokens_map_file": null,
"strip_accents": null,
"tokenize_chinese_chars": true,
"tokenizer_class": "MPNetTokenizer",
"unk_token": "[UNK]"
}