Morph MORPH
SIGN IN SIGN UP
dhruvhead / Megatron-Bridge UNCLAIMED
0 0 1 Python

ACTIVITY

Commits Reviews
week month year
COMMITS
9
in the last week
CONTRIBUTORS
7
active
STARS
0
total
FORKS
0
total
TOP CONTRIBUTORS
Y
Yu Yao
3 commits
G
GavinZhu-GMI
1 commit
O
Oded Ovadia
1 commit
D
dafu
1 commit
H
Hersh Godse
1 commit
S
svcnemo-autobot
1 commit
J
Jingyue Wu
1 commit
RECENT COMMITS
G
[peft] fix: LoRA adapter init no longer depends on tensor parallel size (#5843)
GavinZhu-GMI
947e12e
O
fix(training): avoid double-counting MTP MoE metric layers (#5795)
Oded Ovadia
8eea7a9
D
fix(model): reject CP>1 in Gemma4DenseProvider with a clear error (#5895)
dafu
d69c7c4
H
fix: dequantize GLM 5 FP8 weights on import (#5851)
Hersh Godse
44c871b
Y
feat(model): add Muse Glimmer support (#5479)
Yu Yao
08fc1c6
S
chore(deps): AUT-2186 bump Megatron-Core nightly pins (#5891)
svcnemo-autobot
886afe9
Y
[docs] fix: correct Nemotron 3.5 Lightning page (#5936)
Yu Yao
d0c6228
Y
[build] chore: bump MCore main to 3134ff9 (#5932)
Yu Yao
c1e12f9
J
feat(training): support MFSDP V2 expert parallelism (#5521)
Jingyue Wu
70d8f3f