cl-a23 / tokenizer_config.json
MLmtl91's picture
model
4e4a7d8 verified
Raw History Blame Contribute Delete
827 Bytes
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"is_local": true,
"local_files_only": false,
"max_length": 200,
"model_max_length": 2048,
"pad_to_multiple_of": null,
"pad_token": "<|endoftext|>",
"pad_token_type_id": 0,
"padding_side": "left",
"return_token_type_ids": false,
"stride": 0,
"tokenizer_class": "CodeGenTokenizer",
"truncation_side": "right",
"truncation_strategy": "longest_first",
"unk_token": "<|endoftext|>",
"chat_template": "{{ bos_token }}{% for m in messages %}{% if m['role'] == 'system' %}[SYS] {{ m['content'] }} [/SYS]\n{% elif m['role'] == 'user' %}[INST] {{ m['content'] }} [/INST]{% elif m['role'] == 'assistant' %} {{ m['content'] }}{{ eos_token }}{% endif %}{% endfor %}"
}