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

Merging new tokens into parts #778

Open
RitwikGupta opened this issue Jan 9, 2025 · 0 comments
Open

Merging new tokens into parts #778

RitwikGupta opened this issue Jan 9, 2025 · 0 comments
Labels
type/question An issue that's a question

Comments

@RitwikGupta
Copy link

❓ The question

I am attempting to pre-train OLMo on new data. I have tokenized these files using the OLMo tokenizer, resulting in millions of small npy files.

You merge your tokens into large parts. How? When I do this by naively concatenating npy files together, it results in nasty CUDA device-side assertions being thrown.

Is there a proper way to introduce new data for pre-training?
@RitwikGupta RitwikGupta added the type/question An issue that's a question label Jan 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
type/question An issue that's a question
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@RitwikGupta