GitLab

* New TE wrapper for PyTorch FullyShardedDataParallel to make TE modules distribute their activations after the forward pass and gather them before the backward pass
Signed-off-by: Alp Dener <adener@nvidia.com>

* simplified TE module setup for FSDP comms
Signed-off-by: Alp Dener <adener@nvidia.com>

* FSDP scatter/gather for tensors saved into autograd ctx now working for base TE modules
Signed-off-by: Alp Dener <adener@nvidia.com>

* make sure activation recompute disables FSDP scatter/gather
Signed-off-by: Alp Dener <adener@nvidia.com>

* make sure Fp8 weight buffers are sharded at the end of the backward pass and gathered before forward
Signed-off-by: Alp Dener <adener@nvidia.com>

* Fixed typo in attribute name
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed bug in finding FSDP-wrapped TE modules
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed typo in fp8 weight tensor name
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed incorrect # of gradients
Signed-off-by: Alp Dener <adener@nvidia.com>

* Added fp8 amax gradient hook tensor to the parameter reset
Signed-off-by: Alp Dener <adener@nvidia.com>

* get rid of erroneous dummy tensor leftover from incorrect rebase
Signed-off-by: Alp Dener <adener@nvidia.com>

* Linting fixes
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixing git snafu and removing debug statements
Signed-off-by: Alp Dener <adener@nvidia.com>

---------
Signed-off-by: Alp Dener <adener@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Fix editable build for paddle
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Remove interval arg from recipe
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Remove usage of interval and use explicit kwarg for testing recipes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Cleanup
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Merged CT+dbias+dact into a single template
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Moved gated activations ifrom the cast_transpose_fused ito a sseparate cpp file
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Code clean up
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Code clean up
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Code clean up
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Code clean up
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

* Update transformer_engine/common/transpose/cast_transpose_fusion.cu
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com>

* Update transformer_engine/common/transpose/cast_transpose_fusion.cu
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com>

* Reverted the change with the file split
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>

---------
Signed-off-by: Oleg Goncharov <ogoncharov@nvidia.com>
Signed-off-by: Oleg Goncharov <64355998+Oleg-Goncharov@users.noreply.github.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Modify CUDA graph tests to use grad accumulation steps
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Initialize grad buffers before capturing CUDA graph in CUDA graph tests
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Only use BS=2 in CUDA graph tests
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Update tests/pytorch/test_cuda_graphs.py
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Llama 3 update
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Times update
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Times update
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* utils.py fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* utils.py fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* utils.py fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* update te llama tutorial to allow running with llama 3 weights
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* small fixes
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* small fix
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* small fix
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* add llama 3 vs llama 2 distinctions
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* paraphrasing and corrected facts
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* fix
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* fix
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

---------
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>
Co-authored-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Co-authored-by: Sudhakar Singh <sudhakars@nvidia.com>

Replace int8_t in PyTorch extensions with int64_t
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* add THD support
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add seq_offsets_o and use new offset calculation
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* addition to previous commit; fix unit test
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add None for offset_o gradient
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix lint
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* WIP: test padding between sequences
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* WIP: fix tests for padding between sequences
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix tests for sbhd/bshd layouts; clean up
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* update cudnn-frontend and add tests for max_seqlen_q=1 and d=256 for inference
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* test sbhd/bshd layouts for sq1, d256 inference case
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix lint
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* replace wording from accumulative to cumulative
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add offset tensors to custom fp8 mha tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add version control for cuDNN
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add sm>=90 constraint for thd support
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix cuDNN support for sq=1, d=256
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix lint and minor tweak for fp8 tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* modify cudnn version and restrict MQA/GQA support for THD
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add notes for seq offset tensors
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add dummy tensor to pass jax build
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add dummy tensor to pass paddle build
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix Jax CI
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

---------
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>
Signed-off-by: cyanguwa <8636796+cyanguwa@users.noreply.github.com>

* add multi-tensor kernels
Signed-off-by: Xin Yao <xiny@nvidia.com>

* add FusedAdam
Signed-off-by: Xin Yao <xiny@nvidia.com>

* add test to qa
Signed-off-by: Xin Yao <xiny@nvidia.com>

* add FusedSGD
Signed-off-by: Xin Yao <xiny@nvidia.com>

* fix lint
Signed-off-by: Xin Yao <xiny@nvidia.com>

---------
Signed-off-by: Xin Yao <xiny@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Initial refactor of FP8 workspaces in Linear module
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Remove extra kernel launch
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Minor perf optimizations

Tensor base class functions in Float8Tensor have significant overhead.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Debug FP8 recipe test
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Refactor FP8 workspaces in LayerNormLinear and LayerNormMLP
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Document FP8 workspace function
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Revert changes to FP8 recipe tests
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Add support for lazy FP8 transpose caching

Previous caching behavior (always fill cache) incorrectly filled cache during CUDA graph warmup steps.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Fix Pylint warnings
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Debug ONNX export

ONNX FP8 cast ops assumed that FP8 scales were created during model export (i.e. not initialized during training).
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Debug fused attention tests
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Make sure Float8Tensor.transpose_2d is backward compatible
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Revert changes to ONNX export operations

Work around ONNX test failures by filling FP8 scale tensors instead of copying into them.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Debug scale factor update in Float8Tensor transpose_2d
Signed-off-by: Tim Moon <tmoon@nvidia.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Change the documentation footer
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

* Update docs toolchain versions
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

---------
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

Make sure RoPE frequencies are in FP32
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Use correct FP8 group in multi-GPU docs

FP8 process group should be tensor-parallel group
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Synchronize FP8 scales over world group in multi-GPU docs
Signed-off-by: Tim Moon <tmoon@nvidia.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>

Signed-off-by: Tim Moon <tmoon@nvidia.com>

allow bias support for sm80/86/89 for cuDNN 9+
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* Fixed Llama tutorial. Changed batch size and added fused=True.
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Tutorial updated but not complete yet.
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Tutorial notebook reseted - removed fuse=true
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Removed fused=true
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Batch size back to 8
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Typo and commented out line
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* fixed whitespace
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* fixed whitespace
Signed-off-by: root <root@ipp2-0037.nvidia.com>

* Added comment to attention line. Fixed potential bug with loading weights - now loading works correctly, confirmed by the generation code.
Signed-off-by: root <root@ipp2-1661.nvidia.com>

* Comments
Signed-off-by: root <root@ipp2-1661.nvidia.com>

* Models cast added again
Signed-off-by: root <root@ipp2-1661.nvidia.com>

* Weight download info
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Moved parameter gate_proj_size to config
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* gate_proj_size removed and put immediate_size instead
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Llama 3 added to tutorial
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Typos fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Typos fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Fixed model loading
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Loading fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Different dim for attention
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Reversed other commit
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Changed name to kv_channels
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Fixed typo
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Back to kv_channels in transformer layer
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Back to kv_channels in transformer layer
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Small bug fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Small bug fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Test fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* changed file modes
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* lint fix and resolved conflict
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* lint fix and resolved conflict
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Lint fix, hopefully last
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

---------
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Signed-off-by: root <root@ipp2-0037.nvidia.com>
Signed-off-by: root <root@ipp2-1661.nvidia.com>
Co-authored-by: root <root@ipp2-2373.nvidia.com>
Co-authored-by: root <root@ipp2-1588.nvidia.com>
Co-authored-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Co-authored-by: root <root@ipp2-0037.nvidia.com>
Co-authored-by: root <root@ipp2-1661.nvidia.com>
Co-authored-by: root <root@ipp2-2371.nvidia.com>
Co-authored-by: root <root@ipp2-1589.nvidia.com>
Co-authored-by: Sudhakar Singh <sudhakars@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

TE checkpoint now preserves the torch autocast context from the forward pass during the recompute phase
Signed-off-by: Alp Dener <adener@nvidia.com>

* Fixed the shape mismatching issue in MLP.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Add a corresponding test
Signed-off-by: Ming Huang <mingh@nvidia.com>

---------
Signed-off-by: Ming Huang <mingh@nvidia.com>
Co-authored-by: Phuong Nguyen <36155692+phu0ngng@users.noreply.github.com>

* added alignment requirements for CuBLAS heuristics
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* minor rewords
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* added unit test for gemm with unaligned inputs
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* added pytest skip if fp8 is not available
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* changed offset so that it has alignment with 128
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

replaced deprecated pkg_resources with packaging
Signed-off-by: Alp Dener <adener@nvidia.com>

* Calibration fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

* Lint fix
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>

---------
Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
Co-authored-by: Pawel Gadzinski <pgadzinski@nvidia.com>

Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

* fix inconsistency for attn mask; now True means participating in attn
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix sliding window window_size for decoder+padding combination
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* revert paddle changes regarding mask
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* revert softmax to 1-mask;0-keep
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* enforce 1-mask out; 0-keep rule for jax masks
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix jax lint
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* revert pytorch mask changes; some kept in tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* revert to jax fused attn on main
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* inverse mask logic for get_cu_seqlens/_and_indices in PyTorch implementation and mask generation in unit tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* temporarily disable update_weight_scale_inv
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* enforce window_size for decoder
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add docstring for mask definition 1-mask out;0-keep
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add aux_ctx_tensors to save_for_backward
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* tweak make_decoder_mask and make_mask in jax tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* skip dBias for shapes other than 1HSS; otherwise dq/dk/dv NaNs
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* expand attn_biases from list to variables in save_for_backward
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix use of variable before assignment in jax dact_lu
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* remove window size definition for decoder
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add change notes in README for padding mask in PyTorch
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* tweak padding mask notes in README
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* expand list to tensors for save_for_backwards
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

---------
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>
Signed-off-by: cyanguwa <8636796+cyanguwa@users.noreply.github.com>

* support main_grad
Signed-off-by: Shijie Wang <jaywan@nvidia.com>

* update main_grad and fuse_wgrad_accumulation
Signed-off-by: Shijie Wang <jaywan@nvidia.com>

* fix ci errors
Signed-off-by: Shijie Wang <jaywan@nvidia.com>

* minor change
Signed-off-by: Shijie Wang <jaywan@nvidia.com>

---------
Signed-off-by: Shijie Wang <jaywan@nvidia.com>

* added squared relu in te-torch
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

Remove act_enum from the del list ActLuPrimitive.partition
Signed-off-by: Ming Huang <mingh@nvidia.com>
Co-authored-by: Phuong Nguyen <36155692+phu0ngng@users.noreply.github.com>

This reverts commit 07291027

.
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* use 3rdparty cudnn-frontend cmake to find cuDNN
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* add check for 3rdparty/cudnn-frontend module
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* switch order of CUDA and cuDNN
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

---------
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

fixed batcher in dbias_cast_transpose primitive
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

Signed-off-by: kunlunl <kunlunl@nvidia.com>
Co-authored-by: cyanguwa <8636796+cyanguwa@users.noreply.github.com>

* renamed gelu to act

* added relu, srelu, qgelu

* fixes initialization for layernorm_fp8_mlp tests

* moved activation_fp8 prim into testunit file

* Moved NVTE_Activation_Enum to common/.../activation.h

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* Import frameworks lazily
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Only load modules lazily after an import error

Pylint doesn't handle lazy loading gracefully.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* fixes for ActLuPrimitive in PAXML

* changed indices for arg_infos in sharding func in dbias_cast_transpose primitive

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

Bump FA version to 2.5.8
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Update FP8 recipe test to handle recipe changes
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* templated primitives and respective C++ functions
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* fixes for LayerNormMLP, tests in test_custom_compute all passed
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* added default arg for pybind get_workspace_size funcs
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* fixes for TestTransFormer with non-gated act tests
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* renamed gelu to act
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* improved enum implementation, avoid using magic numbers
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* Exposed C++ ActivationEnum to python side
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* Changed error messages
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* changed conditional check on input shape for dbias_cast_transpose
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* changed dtype (tol) for bias grad tests
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* fixes so that layer_norm_fp8_mlp can take bias = None
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* Set bias = None in flax modules
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

---------
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>

* Add layernorm_fp8_dot unit test
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update the softmax primitives support conditions
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add tests for the softmax primitives
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Round1 refactor of test_layer
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Split dropout arguments of ref code and add hidden/intermediate dropout elementwise comparison
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add dropout_braodcast_dim, self_attn_mask tests and clean a few code
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Abstract test layer and fix a rope reference code diff
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add bias tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add epsilon and float32 tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add relpos_bias and attention dropout tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Loose the atol
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Move common fixtures to conftest.py
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add doc string for test_layer
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add doc string for test_layer
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix conflicts of test_layer
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Avoid to left bias parameters in graph when use_bias=False
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>