PINDARO-AI-CODE / tokenizer_config.json
RthItalia's picture
Release AICE-v1: tokenizer_config.json
4edefce 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
},
"32000": {
"content": "<|noesis|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32001": {
"content": "<|end|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"additional_special_tokens": [
"<|noesis|>",
"<|end|>"
],
"bos_token": "<s>",
"chat_template": "{{ bos_token }}{% for message in messages %}<|noesis|>\n{% if message['role'] == 'system' %}### System\n{{ message['content'] }}\n{% elif message['role'] == 'user' %}### Question\n{{ message['content'] }}\n{% elif message['role'] == 'assistant' %}### Answer\n{{ message['content'] }}\n{% endif %}<|end|>\n{% endfor %}{% if add_generation_prompt %}<|noesis|>\n### Answer\n```\n{% endif %}",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"legacy": false,
"max_length": 512,
"model_max_length": 2048,
"pad_token": "</s>",
"padding_side": "right",
"sp_model_kwargs": {},
"stride": 0,
"tokenizer_class": "LlamaTokenizer",
"truncation_side": "right",
"truncation_strategy": "longest_first",
"unk_token": "<unk>",
"use_default_system_prompt": false
}