.. |
__init__.py
|
541e423ae6
Enable tensor fragments for zero 2 & 3 (#2727)
|
1 year ago |
comms_logging.py
|
da84e60d98
add missing license info to top of all source code (#2889)
|
1 year ago |
debug.py
|
da84e60d98
add missing license info to top of all source code (#2889)
|
1 year ago |
exceptions.py
|
c0af6d90f7
Refactor MoE and Groups API to simplify model creation and mangement (#1798)
|
2 years ago |
groups.py
|
5fe9d61065
Tensor parallelism for Mixture of Experts (#2074)
|
2 years ago |
init_on_device.py
|
aa88137b8d
Add Inference support for running the BigScience-BLOOM Architecture (#2083)
|
2 years ago |
logging.py
|
da84e60d98
add missing license info to top of all source code (#2889)
|
1 year ago |
mixed_precision_linkage.py
|
541e423ae6
Enable tensor fragments for zero 2 & 3 (#2727)
|
1 year ago |
nvtx.py
|
da84e60d98
add missing license info to top of all source code (#2889)
|
1 year ago |
tensor_fragment.py
|
541e423ae6
Enable tensor fragments for zero 2 & 3 (#2727)
|
1 year ago |
timer.py
|
98cc35b6a8
Abstract accelerator (step 3) (#2677)
|
1 year ago |
types.py
|
da84e60d98
add missing license info to top of all source code (#2889)
|
1 year ago |
zero_to_fp32.py
|
541e423ae6
Enable tensor fragments for zero 2 & 3 (#2727)
|
1 year ago |