Text Generation
Transformers
PyTorch
5 languages
RefinedWeb
falcon-40b
long-context
falcon
NTK-YaRN
conversational
custom_code
text-generation-inference

Prompt for chat conversation?

#3
by thomlevy - opened

Thanks for sharing your model.
I saw your example prompt, for a user query after a system message.
Can we extend it for a chat conversation with several user/assistant exchanges in the following way?
Prompt:

<start_system>You are Alfred, a helpful assistant trained by LightOn. Knowledge cutoff: November 2022. Current date: 16 November, 2023<end_message>
<start_user>{user query1}<end_message><start_assistant>{alfred resp1}<end_message>
<start_user>{user query2}<end_message><start_assistant>

Thanks

Hi, thansk for your question. That is correct, the model can be used for multi-turn conversations. However, don't forget to use the chat tokens:

<start_system>You are Alfred, a helpful assistant trained by LightOn. Knowledge cutoff: November 2022. Current date: 16 November, 2023<end_message><start_user>{user query1}<end_message><start_assistant>{alfred resp1}<end_message><start_user>{user query2}<end_message><start_assistant>{alfred resp2}<end_message><start_user>{user query3}<end_message><start_assistant>{alfred resp3}

and so on

thomlevy changed discussion status to closed

Sign up or log in to comment