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

Calculate appropriate max sequence length based on the dataset #175

Open

valedan opened this issue Apr 13, 2023 · 1 comment

Labels

Contributor

valedan commented Apr 13, 2023

Right now we always use 512, which is overkill for most of the models we are training. Only using what we need will speed up training.

Member

traeuker commented Jun 28, 2023

Seems less relevant; feel free to re-prio it if I am buffooning me

mivanit added the code-quality label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment