internlm-chat-7b / tokenizer_config.json
Matt
Move to in-library checkpoint
d3397bb
raw
history blame
No virus
960 Bytes
{
"add_bos_token": true,
"add_eos_token": false,
"bos_token": "<s>",
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.last and message['role'] != 'user' %}{{ raise_exception('Most recent message must come from user!') }}{% endif %}{% if message['role'] == 'user' %}{{ '<|User|>:' + message['content'] + '<eoh>\n'}}{% elif message['role'] == 'assistant' %}{{ '<|Bot|>:' + message['content'] + '<eoa>\n'}}{% else %}{{ raise_exception('Only user and assistant roles are supported in this model!') }}{% endif %}{% endfor %}{{ '<|Bot|>:' }}",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"legacy": true,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "</s>",
"tokenizer_class": "LlamaTokenizer",
"unk_token": "<unk>"
}