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

Could I change the max_seq_length? #43

Open
asenasen123 opened this issue Oct 8, 2023 · 3 comments
Open

Could I change the max_seq_length? #43

asenasen123 opened this issue Oct 8, 2023 · 3 comments

Comments

@asenasen123
Copy link

I need to input a length greater than the maximum sequence length.Could I need change the max_seq_length?

@Muennighoff
Copy link
Owner

Hey yes, you can increase it up to 2048.
I'm not sure if the performance will be better than just cutting your inputs to be shorter.

@asenasen123
Copy link
Author

Hey yes, you can increase it up to 2048. I'm not sure if the performance will be better than just cutting your inputs to be shorter.

Thank for your replay. Could I increase it up to 4096? Does max_seq_length refer to the longest sequence length of tokens? If it is in Chinese, what is the approximate length limit?

@Muennighoff
Copy link
Owner

4096 will work for the sgpt-bloom models I think as they use Alibi embeddings so you could theoretically use any sequence length you want.

Yes it does

It is the same for all languages

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants