Actions: microsoft/DeepSpeed
Actions
2,958 workflow runs
2,958 workflow runs
Add FALCON-40B Inference-Kernel Support
python
#3112:
Pull request #3656
synchronize
by
RezaYazdaniAminabadi
fix the bug of save bf16 optimizer state in the bf16+zero1+pp mode
python
#3111:
Pull request #3759
opened
by
L-hongbin
[squash] styoun/triton fp16 transformer (#530)
python
#3110:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3109:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3108:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3107:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3106:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3105:
Pull request #3748
synchronize
by
stephen-youn
[squash] styoun/triton fp16 transformer (#530)
python
#3104:
Pull request #3748
synchronize
by
stephen-youn
readme for blog
python
#3103:
Commit f41e279
pushed
by
stephen-youn
[profiling]add show_straggler argument to log_summary()
python
#3100:
Pull request #3579
synchronize
by
loadams
fix error :Dictionary expression not allowed in type annotation Pylance
python
#3099:
Pull request #3708
synchronize
by
loadams
fix ccl_backend and residual_add problems
python
#3098:
Pull request #3642
synchronize
by
loadams
fix the bug of save model in the bf16+zero1+pp mode
python
#3097:
Pull request #3756
opened
by
L-hongbin
fix error :Dictionary expression not allowed in type annotation Pylance
python
#3096:
Pull request #3708
synchronize
by
digger-yu
Account for expert parameters when calculating the total number of pa…
python
#3095:
Pull request #3720
synchronize
by
alito
fix ccl_backend and residual_add problems
python
#3093:
Pull request #3642
synchronize
by
loadams
add Chinese Zhihu social account
python
#3092:
Pull request #3755
opened
by
conglongli
[squash] styoun/triton fp16 transformer (#530)
python
#3091:
Pull request #3748
synchronize
by
stephen-youn
Triton kernels and BERT inference using triton in float16 (#459)
python
#3090:
Commit b978eae
pushed
by
stephen-youn
fix ccl_backend and residual_add problems
python
#3089:
Pull request #3642
synchronize
by
loadams
Add H100 workflow
python
#3088:
Pull request #3754
synchronize
by
loadams