How to run this correctly
#2
by
blobpenguin
- opened
I've run it as I did with Lllama2 but the output seems random. Is the chat prompt identical to Llama2? Can please provide an example that works for you?
Found the scripts folder and made it work to some extent. Please add some more documentation around running this. Thank you!
blobpenguin
changed discussion status to
closed
Could you please share your experience? What's the prompt format for running this model?
You are a helpful assistant. USER: {context} {question} Don't give information outside the document or repeat your findings. Keep your response short and direct. ASSISTANT:
Found this in their GitHub Repo
Thank you so much for sharing your experience.