microsoft/DeepSpeed

Pull Requests Commits

tjruwase committed 2 years ago

2ef73009

Respect memory pinning config

tjruwase committed 2 years ago

3d5ea143

Update nightly workflows to open an issue if CI fails (#3952)

loadams committed 2 years ago

Verified 0c75f4a3

Add ops (#4119)

loadams committed 2 years ago

Verified d300517f

Fix Issue 4083 (#4084)

jomayeri committed 2 years ago

Verified 8a8683d3

enable pipeline checkpoint loading mode (#3629)

leiwen83 committed 2 years ago

Verified 1e0c39c6

add deepspeed chat arxiv report (#4110)

conglongli committed 2 years ago

Verified 78d985ab

Pass correct node size for ZeRO++ (#4085)

cmikeh2 committed 2 years ago

Verified f0463b4d

Disable z3 tracing profiler (#4106)

tjruwase committed 2 years ago

Verified 977254c1

use correct ckpt path when base_dir not available (#4101)

polisettyvarma committed 2 years ago

Verified abe293b4

set temperature to avoid config validation error (#4107)

mrwyattii committed 2 years ago

Verified 975bcbc0

add type checker ignore to resolve that pylance can't resolved noqa annotation (#4102)

EeyoreLee committed 2 years ago

Verified 57a27b08

zero_to_fp32 script adds support for tag argument (#4089)

EeyoreLee committed 2 years ago

Verified 241ae39a

update ut/doc for glm/codegen (#4057)

inkcherry committed 2 years ago

Verified 85dc854b

fix typo: change polciies to policies (#4090)

digger-yu committed 2 years ago

Verified 4cde5da8

Spread layers more uniformly when using partition_uniform (#4053)

marcobellagente93 committed 2 years ago

Verified e8318634

Fix Stable Diffusion Injection (#4078)

lekurile committed 2 years ago

Verified 1ba40989

unpin datasets in UT (#4079)

mrwyattii committed 2 years ago

Verified a7fe3bcc

Refactor autoTP inference for HE (#4040)

molly-smith committed 2 years ago

Verified 94c7233a

fix: remove unnessary `#` punct in the second `sed` command (#4061)

hughpu committed 2 years ago

Verified e31b4041

add reproducible compilation environment (#3943)

fecet committed 2 years ago

Verified f763b93d

save_non_zero_checkpoint on first partition group (#3787)

zarzen committed 2 years ago

Verified 8a63754b

Fix deadlock when SHM based allreduce spin too fast (#4048)

delock committed 2 years ago

Verified 82c498d9

Multiple zero stage 3 related fixes (#3886)

tjruwase committed 2 years ago

Verified 7f90ef4b

faster allreduce with omp parallel for reduce kernel (#4049)

delock committed 2 years ago

Verified 7f26bb6a

autoTP for fused qkv weight (#3844)

inkcherry committed 2 years ago

Verified 6b877d2d

enable autoTP for MPT (#3861)

sywangyi committed 2 years ago

Verified 0bafeac4

fix opt-350m shard loading issue in AutoTP (#3600)

sywangyi committed 2 years ago

Verified 76953a37

fix comm logging for inference (#4043)

delock committed 2 years ago

Verified 0b507253

fix: Remove duplicate word the (#4051)

digger-yu committed 2 years ago

Verified 389bf693

Older