mpt-7b-instruct2 / tokenizer_config.json
Matthew Stallone mstallone@ibm.com
Initial release
04ad1ab
raw
history blame
430 Bytes
{
"add_prefix_space": false,
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 2048,
"padding_side": "right",
"special_tokens_map_file": "/workspace/.cache/huggingface/hub/models--mosaicml--mpt-7b/snapshots/4ff95c4aec5c04ba509ddf517c56720541a7a487/special_tokens_map.json",
"tokenizer_class": "GPTNeoXTokenizer",
"unk_token": "<|endoftext|>"
}