microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into loadams/adam-params

loadams committed 263 days ago

Verified d891a8bd

Update version after 0.17.1 release (#7345)

loadams committed 263 days ago

Verified 4686d5ef

Move pytest pinning from individual tests to requirements-dev.txt until fixed. (#7327)

loadams committed 263 days ago

Verified 2ce55057

Fix docs that are rendering Incorrectly (#7344)

felixgondwe committed 263 days ago

Verified 4d0c1596

Improve overflow handling in ZeRO (#6976)

tjruwase committed 264 days ago

Verified e440506b

Update folder name (#7343)

sfc-gh-truwase committed 264 days ago

Verified bb293aea

Fix LoRA arxiv reference (#7340)

emmanuel-ferdman committed 266 days ago

Verified 05818e90

fixed: Modified the topkgating function and modified the test_moe file for testing (#7163)

xiongjyu committed 266 days ago

Verified 770967f5

DeepNVMe update (#7215)

tjruwase committed 266 days ago

Verified 24a1d8f9

fp16 optimizer timers fix - TypeError: 'NoneType' object is not callable (#7330)

rraminen committed 267 days ago

Verified cb3ad0c1

Fix issue with symint input (#7243)

tohtana committed 267 days ago

Verified 7d0c3f78

Fix pytest version to 8.3.5 in hpu-gaudi actions (#7337)

raza-sikander committed 267 days ago

Verified 2ad2011c

Update config_utils.py (#7333)

qgallouedec committed 268 days ago

Verified d0f7091a

Improve Ulysses Plus Docs (#7335)

cynricfu committed 268 days ago

Verified b8d4b842

UlyssesPlus Docs take 2 (#7332)

stas00 committed 270 days ago

Verified 097f0637

Ulysses Plus Docs (#7331)

stas00 committed 270 days ago

Verified 81a47408

Update version.txt after v0.17.0 release (#7326)

loadams committed 270 days ago

Verified 8f3c3e78

Bump to v0.17.0 (#7324)

sfc-gh-mwyatt committed 270 days ago

Verified 720787e7

Fix ci hang in torch2.7& improve ut (#7321)

inkcherry committed 270 days ago

Verified 8b03a356

Ulysses SP for HF Integration (#7268)

stas00 committed 273 days ago

Verified 4d00b38a

fix asymmetric in dequantize (#7283)

pencil-hub committed 274 days ago

Verified 0baf79ea

anchor transformers version (#7316)

stas00 committed 275 days ago

Verified b66c8107

Update gaudi2 nightly,ci to latest 1.21.0 build (#7313)

raza-sikander committed 275 days ago

Verified ec6b254d

`tests/conftest.py`: automatically add local deepspeed repo when running tests (#7317)

stas00 committed 275 days ago

Verified e5afb887

CI: prefer bf16 over fp16 (#7304)

stas00 committed 276 days ago

Verified b4cc079e

Fix: Update grad norm calculation for CPU offload (#7302)

therealnaveenkamal committed 277 days ago

Verified b9af5d8d

Fix the GPU memory usage of ZeRO-Offload (only update stage_1_and_2.py) (#7309)

arminzhu committed 277 days ago

Verified 17c8be07

Fix AutoTP gathering replaced layer params when bias is not None (#7257)

HollowMan6 committed 279 days ago

Verified b666844f

Update COMMITTERS.md (#7305)

PKUWZP committed 281 days ago

Verified d4032ec7

Update next version in version.txt after 0.16.9 release. (#7306)

loadams committed 281 days ago

Verified e3bd16fd

Older