Loreon
Labs
Platform
Docs
Home
Ecosystems
Python
Megatron-LM
Ongoing research training transformer models at scale
Python
Emerging
GitHub
Website
Stars
—
Forks
—
Contributors
8
Last push
9d ago
Recent commits
Latest commits.
Fix: Perform sigmoid calculation in fp32 for aux loss stability (#2765)
5f668c1
Srijan Upadhyay
4mo ago
fix: async_utils: explicit GC in persistent checkpoint worker loop (#3591)
e1a9ac9
Seonmyeong Bak
4mo ago
Add debug info to an assert. (#3588)
18c94d2
Vitaly Kurin
4mo ago
ci: Increase changelog generation max PRs fetched (#3620)
027e0f3
Charlie Truong
4mo ago
docs: Fix version picker urls (#3621)
1b1f5c4
Charlie Truong
4mo ago
ci: Re-add release tag prefix (#3619)
36d8a9d
oliver könig
4mo ago
ci: No comment for release workflow (#3615)
6161f7a
oliver könig
4mo ago
ci: Skip cleanup-taint-node jobs during deployments (#3612)
9100119
oliver könig
4mo ago
Top contributors
Builders behind this project.
ko3n1g
1.7K commits
jaredcasper
1K commits
lmcafee-nvidia
399 commits
shanmugamr1992
391 commits
shoeybi
339 commits
deepakn94
317 commits
ericharper
282 commits
mikolajblaz
242 commits