test-llama-tiny / tokenizer_config.json
kmfoda's picture
Upload tokenizer_config.json with huggingface_hub
abf7537 verified
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": "<|bos|>",
"eos_token": "<|eos|>",
"errors": "replace",
"extra_special_tokens": [
"<|user_start|>",
"<|user_end|>",
"<|assistant_start|>",
"<|assistant_end|>",
"<|system_start|>",
"<|system_end|>"
],
"is_local": false,
"model_max_length": 1024,
"pad_token": null,
"tokenizer_class": "GPT2Tokenizer",
"unk_token": "<|endoftext|>",
"chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}<|system_start|>{{ message['content'] }}<|system_end|>{% elif message['role'] == 'user' %}<|user_start|>{{ message['content'] }}<|user_end|>{% elif message['role'] == 'assistant' %}<|assistant_start|>{{ message['content'] }}<|assistant_end|>{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant_start|>{% endif %}"
}