Llama 2 Prompt Template
gpus.llm-utils.org
gpus.llm-utils.org
I took Meta's generation.py and modified the code to output the raw prompt text before it’s fed to the tokenizer, to get an updated prompt template.
Result is:
[INST] <<SYS>>
{your_system_message}
<</SYS>>
{user_message_1} [/INST] {model_reply_1}
[INST] {user_message_2} [/INST]