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.

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

Quantizing last layer for output prediction/logits #27

Open
DeepCowProductions opened this issue Jan 25, 2024 · 0 comments
Open

Quantizing last layer for output prediction/logits #27

DeepCowProductions opened this issue Jan 25, 2024 · 0 comments
Labels
bug Something isn't working

Comments

@DeepCowProductions
Copy link
Contributor

Currently, we can not quantize this layer, otherwise our model does not converge properly.
When training from scratch, it does not converge at all when all other layers are also quantized.

@DeepCowProductions DeepCowProductions added the bug Something isn't working label Jan 25, 2024
@DeepCowProductions DeepCowProductions changed the title Last layer for output prediction/logits Quantizing last layer for output prediction/logits Jan 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

1 participant