llama_2_gsm8k_per_class_reflect / tokenizer_config.json
CharlesLi's picture
Training in progress, step 20
96a8b1b verified
{
"add_bos_token": true,
"add_eos_token": false,
"add_prefix_space": null,
"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
}
},
"bos_token": "<s>",
"chat_template": "{% set system_message = '<<SYS>>' + messages[0]['content'] | trim + '<</SYS>>' if messages[0]['role'] == 'system' else '' %}{% set messages = messages[1:] if messages[0]['role'] == 'system' else messages %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{{ bos_token + '[INST] ' + (system_message + message['content'] if loop.index0 == 0 else message['content']) | trim + ' [/INST]' if message['role'] == 'user' else ' ' + message['content'] | trim + ' ' + eos_token }}{% endfor %}",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"legacy": false,
"model_max_length": 2048,
"pad_token": "</s>",
"padding_side": "right",
"sp_model_kwargs": {},
"tokenizer_class": "LlamaTokenizer",
"unk_token": "<unk>",
"use_default_system_prompt": false
}