microsoft/DeepSpeed

Pull Requests Commits

molly-smith committed 2 years ago

e1a0c539

Merge branch 'master' into mosm/inf-refactor

molly-smith committed 2 years ago

Verified ea49e41b

Some gpt2 autotp path fixes

molly-smith committed 2 years ago

334a2f92

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

Merge branch 'master' into mosm/inf-refactor

molly-smith committed 2 years ago

Verified ae85c1b9

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

molly-smith committed 2 years ago

e68e8e21

Merge branch 'master' into mosm/inf-refactor

molly-smith committed 2 years ago

6578d4e1

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

Engine side fix for loading llama checkpoint fine-tuned with zero3 (#3981)

minjiaz committed 2 years ago

Verified 15f94ae7

Merge branch 'master' into mosm/inf-refactor

molly-smith committed 2 years ago

Verified 0274068e

polishing timers and log_dist (#3996)

Alexander Jipa committed 2 years ago

Verified b354c28b

Merge branch 'master' into mosm/inf-refactor

molly-smith committed 2 years ago

Verified ebdf58d3

Remove self from loading class

molly-smith committed 2 years ago

3573994c

Move redundant functions to autotp

molly-smith committed 2 years ago

7e3e7cd1

molly-smith committed 2 years ago

9c48fb47

Refactor autoTP inference for HE

molly-smith committed 2 years ago

f33910c7

fix(pipe): `strict` parameter of method `PipelineModule.load_state_dir` is not work (#4020)

hughpu committed 2 years ago

Verified 09601bb8

Monitored Loss Calculations (#4030)

jomayeri committed 2 years ago

Verified eeab613a

Remove print of weight parameter in RMS norm (#4031)

Puneesh Khanna committed 2 years ago

Verified ad661b8e

Pass missing positional arguments in `DeepSpeedHybridEngine.generate()` (#4026)

XuehaiPan committed 2 years ago

Verified 43188ff0

remove duplicate check for pp and zero stage (#4033)

inkcherry committed 2 years ago

Verified e8600b7f

Older