cpm-bee-10b-shard-1GB / tokenizer_config.json
WHJ1998's picture
Upload tokenizer
04e7224
raw
history blame contribute delete
427 Bytes
{
"auto_map": {
"AutoTokenizer": [
"openbmb/cpm-bee-10b--tokenization_cpmbee.CpmBeeTokenizer",
null
]
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"eos_token": "</s>",
"line_token": "\n",
"mask_token": "<mask>",
"model_max_length": 4096,
"pad_token": "<pad>",
"padding_side": "left",
"space_token": " ",
"tokenizer_class": "CpmBeeTokenizer",
"unk_token": "<unk>"
}