Text Generation
Transformers
PyTorch
Safetensors
gpt2
conversational
text-generation-inference
Inference Endpoints
gpt-sw3-356m-instruct / tokenizer_config.json
Ekgren's picture
Add default chat template to tokenizer_config.json (#5)
6ac6a39 verified
raw
history blame
349 Bytes
{
"model_max_length": 2048,
"padding_side": "left",
"truncation_side": "left",
"chat_template": "{{ eos_token }}{{ bos_token }}{% for message in messages %}{% if message['role'] == 'user' %}{{ 'User: ' + message['content']}}{% else %}{{ 'Bot: ' + message['content']}}{% endif %}{{ message['text'] }}{{ bos_token }}{% endfor %}Bot:"
}