-
Notifications
You must be signed in to change notification settings - Fork 3.1k
Pull requests: PaddlePaddle/PaddleNLP
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Normalize gates on expert dim before calculating seq_aux_loss
stale
#11160
opened Nov 3, 2025 by
lshpku
Loading…
[Auto-Paralllel] fix intermediate api use
stale
#11124
opened Sep 28, 2025 by
Xing-lil
Contributor
Loading…
2 tasks
【FlexCheckpoint】fix_the_optimizer_init
contributor
stale
#11123
opened Sep 27, 2025 by
zty-king
Contributor
Loading…
2 tasks
Feat: support chatglm v2 faster infer in p800
stale
#11118
opened Sep 25, 2025 by
mingMelody
Contributor
Loading…
hack offload optimizer减少一次master weight的offload&reload
stale
#11111
opened Sep 23, 2025 by
Wennie396
Contributor
Loading…
update using_post_norm_recompute
stale
#11093
opened Sep 16, 2025 by
chen2016013
Contributor
Loading…
2 tasks
add script for training gpt3 on XPU machine using flagcx as comm backend
contributor
stale
#11014
opened Aug 26, 2025 by
mikethegoblin
Loading…
2 tasks
Optimie moe and dense overlap
stale
#11013
opened Aug 26, 2025 by
phlrain
Collaborator
Loading…
2 tasks
[NOT MERGE]Pr adapt flex checkpoint
contributor
stale
#10996
opened Aug 25, 2025 by
zty-king
Contributor
Loading…
2 tasks
[BUG]: fix the bug in PretrainedModel.recompute_disable()
contributor
stale
#10988
opened Aug 21, 2025 by
hongjx175
Loading…
2 tasks
[do not merge] support llama2 flex ckpt save&load
stale
#10952
opened Aug 14, 2025 by
AndSonder
Contributor
Loading…
2 tasks
Previous Next
ProTip!
Updated in the last three days: updated:>2026-04-15.