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

sgd-with-adam & weight decay scheduler & Cautious & C4 dataset #29

Open
wants to merge 16 commits into
base: soap
Choose a base branch
from
Open

sgd-with-adam & weight decay scheduler & Cautious & C4 dataset #29

wants to merge 16 commits into from

Conversation

Andron00e
Copy link
Collaborator

No description provided.

@Andron00e
Copy link
Collaborator Author

Andron00e commented Jan 7, 2025 •
edited
Loading

this PR introduces a family of Cautious optimizers and weight decay scheduling (cosine, WSD, stable-decay; with different cooldown shapes).
in addition, this PR include mix of SGD and Adam: Adam trains lm_head, layernorms, embeds by default, all other parameters are optimized by SGD. however this all can be customized further.

TODO:

  • customize div_factor of schedulers (default value is 1e2) so we can experiment with down to 10%, down to zero, etc. [done]
  • add C4 dataset [ongoing]
  • handle the large filename oseerror [ongoing]

@Andron00e Andron00e changed the title sgd-with-adam updates & weight decay scheduler code sgd-with-adam & weight decay scheduler code & Cautious & C4 dataset Jan 7, 2025
@Andron00e Andron00e changed the title sgd-with-adam & weight decay scheduler code & Cautious & C4 dataset sgd-with-adam & weight decay scheduler & Cautious & C4 dataset Jan 7, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@Andron00e