Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

server: recieving <|im_end|> in all responses of llama 3 #6873

Closed
infozzdatalabs opened this issue Apr 24, 2024 · 2 comments
Closed

server: recieving <|im_end|> in all responses of llama 3 #6873

infozzdatalabs opened this issue Apr 24, 2024 · 2 comments
Labels
bug-unconfirmed

Comments

@infozzdatalabs
Copy link

I have been experiencing this problem since I have tried different models of llama 3 for example:
https://huggingface.co/QuantFactory/Meta-Llama-3-8B-Instruct-GGUF
https://huggingface.co/QuantFactory/dolphin-2.9-llama3-8b-GGUF

In all responses calling "/chat/completions" returns at the end the '<|im_end|>'.

I'm using the latest docker version for cuda: 'ghcr.io/ggerganov/llama.cpp:server-cuda'

Thanks in advance.
@slavonnet slavonnet mentioned this issue Apr 24, 2024
Closed
@infozzdatalabs
Copy link
Author

Sorry, I didn't see that an issue was already open.

@ggerganov
Copy link
Owner

Fixed in #6860

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
bug-unconfirmed
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ggerganov @infozzdatalabs