tokenizer / tokenizer_config.json
adriancowham's picture
Upload tokenizer_config.json
d998b94 verified
{
"add_bos_token": true,
"add_eos_token": false,
"added_tokens_decoder": {
"0": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "</s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"additional_special_tokens": [],
"bos_token": "<s>",
"chat_template": "{% if messages[0]['role'] == 'system' %}{% if messages[1]['role'] == 'user' %}{% set sequence_start = 'user' %}{% elif messages[1]['role'] == 'assistant' %}{% set sequence_start = 'assistant' %}{% else %}{{ raise_exception('Second message must be either user or assistant after system') }}{% endif %}{% else %}{{ raise_exception('First message must be from system') }}{% endif %}{{ bos_token }}{% for message in messages %}{% if sequence_start == 'user' %}{% if (message['role'] == 'user') != ((loop.index0) % 2 == 1) %}{{ raise_exception('Conversation roles must alternate starting with user after system') }}{% endif %}{% elif sequence_start == 'assistant' %}{% if (message['role'] == 'assistant') != ((loop.index0) % 2 == 1) %}{{ raise_exception('Conversation roles must alternate starting with assistant after system') }}{% endif %}{% endif %}{% if message['role'] == 'system' %}{{ '[INST] <<SYS>>' + message['content'].strip() + '<</SYS>>' }}{% elif message['role'] == 'user' %}{{ '[INST] ' + message['content'].strip() + ' [/INST] ' }}{% elif message['role'] == 'assistant' %}{{ message['content'].strip() + eos_token}}{% else %}{{ raise_exception('Only system, user, and assistant roles are supported!') }}{% endif %}{% endfor %}",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"legacy": true,
"model_max_length": 1000000000000000019884624838656,
"pad_token": null,
"sp_model_kwargs": {},
"spaces_between_special_tokens": false,
"tokenizer_class": "LlamaTokenizer",
"unk_token": "<unk>",
"use_default_system_prompt": false
}