deepspeedai/DeepSpeed

Pull Requests Commits

Merge branch 'master' into loadams/megatron-new-pypi

loadams committed 2 years ago

Verified 70279766

Use git tag matching other version

loadams committed 2 years ago

1c92fdfd

Add HE support for the rest of model containers (#3191)

RezaYazdaniAminabadi committed 2 years ago

Verified 3e856464

save tensors in context of memory_efficient_linear (#3413)

tohtana committed 2 years ago

Verified 42858a98

Test megatron-lm replacement for 1.1.5

loadams committed 2 years ago

e43245cd

Revert "remove megatron-lm, no longer pip installable (#3389)"

loadams committed 2 years ago

ec6b94a9

doc updates (#3415)

goodship1 committed 2 years ago

Verified b4b63f52

Fix pipeline module evaluation when contiguous activation checkpointing is enabled (#3005)

BacharL committed 2 years ago

Verified 7ddc3b01

remove megatron-lm, no longer pip installable (#3389)

jeffra committed 2 years ago

Verified a094c976

[docs] add twitter badge to readme

jeffra committed 2 years ago

Verified f7d71ec1

Fix memory leak in zero2 contiguous gradients (#3306)

BacharL committed 2 years ago

Verified 01d17492

stage_1_and_2.py: do gradient scale only for fp16 (#3166)

guoyejun committed 2 years ago

Verified 0e357666

[DRAFT] Tentative implementation of MiCS (#2964)

zarzen committed 2 years ago

Verified 2e99f6ed

Auto TP Tutorial with T5 Example (#2962)

molly-smith committed 2 years ago

Verified d9253950

fixing default communication_data_type for bfloat16_enabled and docs (#3370)

Alexander Jipa committed 2 years ago

Verified d56268f3

Fix redundant shared_params in zero_to_fp32.py (#3149)

ShijieZZZZ committed 2 years ago

Verified 39825a90

add deepspeed chat blog links, add tags (#3369)

conglongli committed 2 years ago

Verified 085981bf

Update DS-Chat issue template (#3368)

mrwyattii committed 2 years ago

Verified 3031eec4

fix mpich launcher issue in multi-node (#3078)

sywangyi committed 2 years ago

Verified a748bfc6

hybrid_engine: check tuple size when fusing lora params (#3311)

adammoody committed 2 years ago

Verified c9196b61

README.md: Update MosaicML docs link (#3344)

kobindra committed 2 years ago

Verified 39b429d5

add bf16 cuda kernel support (#3092)

dc3671 committed 2 years ago

Verified 297cd9ed

print default values (#3347)

mrwyattii committed 2 years ago

Verified 2f8d384e

Diffusers 0.15.0 bug fix (#3345)

molly-smith committed 2 years ago

Verified 496a9a3a

Hipify cooperative_groups headers (#3323)

rraminen committed 2 years ago

Verified 6e1cbebe

Fix formatting (#3343)

mrwyattii committed 2 years ago

Verified 4a3ca4e2

jeffra committed 2 years ago

78afab04

Explicitly check for OPT activation function (#3278)

cmikeh2 committed 2 years ago

Verified 793c23e5

Fix missing scale attributes for GPTJ (#3256)

cmikeh2 committed 2 years ago

Verified 145c3a75

Fix for dist not being initialized when constructing main config (#3324)

mrwyattii committed 2 years ago

Verified ad168a69

Older