forked from NVIDIA/Megatron-LM
-
Notifications
You must be signed in to change notification settings - Fork 1
Pull requests: scitix/Megatron-LM
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[GRL] Land PR400 sparse/delta megatron-lm dependency
#16
opened Jun 25, 2026 by
flowerfox-scitix
Loading…
feat(moe): shared MoE expert-routing stats (capture/reduce/scalars + router hook + CLI)
#15
opened Jun 24, 2026 by
ilsa001
Collaborator
Loading…
feat(moe): support sonic-moe for small model training
#12
opened Jun 15, 2026 by
hscos
Loading…
6 tasks
feat(qat): add MXFP4 fake-QAT STE on TEGroupedLinear expert weights
#4
opened May 13, 2026 by
tonic-scitix
Loading…
6 tasks
ProTip!
Follow long discussions with comments:>50.