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

Evaluate falcon3 family #251

Open
the-crypt-keeper opened this issue Dec 17, 2024 · 2 comments
Open

Evaluate falcon3 family #251

the-crypt-keeper opened this issue Dec 17, 2024 · 2 comments
Labels
model request Evaluate performance of a new model

Comments

@the-crypt-keeper
Copy link
Owner

https://huggingface.co/collections/tiiuae/falcon3-67605ae03578be86e4e87026

The falcon has landed
@the-crypt-keeper the-crypt-keeper added the model request Evaluate performance of a new model label Dec 17, 2024
the-crypt-keeper pushed a commit that referenced this issue Dec 17, 2024
@KryptK420
#251 evaluate falcon3 family
ad999f6
@the-crypt-keeper
Copy link
Owner Author

Complete. A bit of an odd result: Every GGUF except 7B is behind the FP16, while the 7B GGUF mysteriously outperforms even 10B FP16.

@the-crypt-keeper
Copy link
Owner Author

ggerganov/llama.cpp#10876

It's been reverted due to problems with the tokenizer, going to have to re-run GGUFs once it's fixed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
model request Evaluate performance of a new model
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@the-crypt-keeper