internlm2-base-20b / tokenizer_config.json
RangiLyu's picture
fast tokenizer and stream_chat fix (#3)
ca3f2c7 verified
raw
history blame
393 Bytes
{
"auto_map": {
"AutoTokenizer": [
"tokenization_internlm2.InternLM2Tokenizer",
"tokenization_internlm2_fast.InternLM2TokenizerFast"
]
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "</s>",
"tokenizer_class": "InternLM2Tokenizer",
"unk_token": "<unk>"
}