IBM / ModuleFormer

ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward experts. We released a collection of ModuleFormer-based Language Models (MoLM) ranging in scale from 4 billion to 8 billion parameters.

Date Created 2023-08-24 (about a year ago)

Commits 19 (last one 11 months ago)

Stargazers 216 (0 this week)

Watchers 11 (0 this week)

Forks 12

License apache-2.0

Ranking

RepositoryStats indexes 595,856 repositories, of these IBM/ModuleFormer is ranked #171,281 (71st percentile) for total stargazers, and #192,427 for total watchers. Github reports the primary language for this repository as Python, for repositories using this language it is ranked #30,437/119,431.

Other Information

IBM/ModuleFormer has 1 open pull request on Github, 4 pull requests have been merged over the lifetime of the repository.

Github issues are enabled, there are 2 open issues and 4 closed issues.

All Topics

lm moe

Star History

Github stargazers over time

Watcher History

Github watchers over time, collection started in '23

Recent Commit History

19 commits on the default branch (main) since jan '22

Yearly Commits

Commits to the default branch (main) per year

Issue History

Languages

The only known language in this repository is Python

updated: 2024-11-06 @ 11:02am, id: 682660919 / R_kgDOKLCUNw