Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Documentation

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
docs.yml
297 workflow run results
297 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Create a small tutorial on how to accelerate HF Llama models with Transformer-Engine Documentation #2320: Pull request #615 opened by sudhakarsingh27
January 19, 2024 00:48 53s sudhakarsingh27:llama_accelerate_tutorial
sudhakarsingh27:llama_accelerate_tutorial
January 19, 2024 00:48 53s
Activation offloading to CPU's for the Linear, Layernorm Linear and the Layernorm MLP modules Documentation #2319: Pull request #571 synchronize by sanandaraj5597
January 18, 2024 23:03 56s main
main
January 18, 2024 23:03 56s
[C++/PyTorch] Add alibi_slopes support Documentation #2318: Pull request #608 synchronize by cyanguwa
January 18, 2024 23:00 54s cyanguwa:benchmark/fa_fu_alibi
cyanguwa:benchmark/fa_fu_alibi
January 18, 2024 23:00 54s
[C++/PyTorch] Add alibi_slopes support Documentation #2316: Pull request #608 synchronize by cyanguwa
January 18, 2024 22:46 52s cyanguwa:benchmark/fa_fu_alibi
cyanguwa:benchmark/fa_fu_alibi
January 18, 2024 22:46 52s
make TransformerLayer accept a bshd or sbhd tensor format Documentation #2315: Pull request #557 synchronize by sudhakarsingh27
January 18, 2024 19:00 1m 21s sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
January 18, 2024 19:00 1m 21s
[JAX] Support SP + RoPE + GeLU Documentation #2310: Pull request #602 synchronize by mingxu1067
January 18, 2024 09:34 1m 7s mingxu1067:mingh/sp_rope_gelu
mingxu1067:mingh/sp_rope_gelu
January 18, 2024 09:34 1m 7s
make TransformerLayer accept a bshd or sbhd tensor format Documentation #2309: Pull request #557 synchronize by sudhakarsingh27
January 18, 2024 00:50 1m 4s sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
January 18, 2024 00:50 1m 4s
make TransformerLayer accept a bshd or sbhd tensor format Documentation #2308: Pull request #557 synchronize by sudhakarsingh27
January 18, 2024 00:43 1m 3s sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
sudhakarsingh27:plumb_tensor_format_thru_transformer_layer
January 18, 2024 00:43 1m 3s
Avoid using torch.compile for roll and fill_ Documentation #2307: Pull request #609 synchronize by ksivaman
January 17, 2024 18:20 1m 23s ksivaman:avoid_torch_compile_for_roll_kernel
ksivaman:avoid_torch_compile_for_roll_kernel
January 17, 2024 18:20 1m 23s
Avoid using torch.compile for roll and fill_ Documentation #2306: Pull request #609 opened by ksivaman
January 17, 2024 18:20 55s ksivaman:avoid_torch_compile_for_roll_kernel
ksivaman:avoid_torch_compile_for_roll_kernel
January 17, 2024 18:20 55s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2305: Pull request #532 synchronize by denera
January 17, 2024 17:41 1m 3s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 17, 2024 17:41 1m 3s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2304: Pull request #532 synchronize by denera
January 17, 2024 17:33 1m 24s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 17, 2024 17:33 1m 24s
[PyTorch] Deferred Initialization via device='meta' option Documentation #2303: Pull request #596 synchronize by denera
January 17, 2024 17:03 1m 5s denera:databricks/fp8-fsdp-oom-fix
denera:databricks/fp8-fsdp-oom-fix
January 17, 2024 17:03 1m 5s
Switch to torch.compile for dropout for torch v>2.2 Documentation #2302: Pull request #607 synchronize by ksivaman
January 17, 2024 06:29 49s ksivaman:fix_torch_compile_dropout_perf
ksivaman:fix_torch_compile_dropout_perf
January 17, 2024 06:29 49s
[JAX] Support SP + RoPE + GeLU Documentation #2301: Pull request #602 synchronize by mingxu1067
January 17, 2024 03:22 50s mingxu1067:mingh/sp_rope_gelu
mingxu1067:mingh/sp_rope_gelu
January 17, 2024 03:22 50s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2299: Pull request #532 synchronize by denera
January 17, 2024 00:45 1m 20s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 17, 2024 00:45 1m 20s
[C++/PyTorch] Add alibi_slopes support Documentation #2298: Pull request #608 synchronize by cyanguwa
January 16, 2024 23:52 46s cyanguwa:benchmark/fa_fu_alibi
cyanguwa:benchmark/fa_fu_alibi
January 16, 2024 23:52 46s
[C++/PyTorch] Add alibi_slopes support Documentation #2297: Pull request #608 opened by cyanguwa
January 16, 2024 23:29 1m 4s cyanguwa:benchmark/fa_fu_alibi
cyanguwa:benchmark/fa_fu_alibi
January 16, 2024 23:29 1m 4s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2296: Pull request #532 synchronize by denera
January 16, 2024 21:51 45s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 16, 2024 21:51 45s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2295: Pull request #532 synchronize by denera
January 16, 2024 21:46 1m 12s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 16, 2024 21:46 1m 12s
[JAX] Custom Op Workspace Tensors from XLA Buffers Documentation #2294: Pull request #532 synchronize by denera
January 16, 2024 21:23 48s denera:jax-gemm-workspace-xla-buffer
denera:jax-gemm-workspace-xla-buffer
January 16, 2024 21:23 48s
[PyTorch] Deferred Initialization via device='meta' option Documentation #2293: Pull request #596 synchronize by denera
January 16, 2024 17:39 46s denera:databricks/fp8-fsdp-oom-fix
denera:databricks/fp8-fsdp-oom-fix
January 16, 2024 17:39 46s
Switch to torch.compile for dropout for torch v>2.2 Documentation #2292: Pull request #607 opened by ksivaman
January 16, 2024 15:35 57s ksivaman:fix_torch_compile_dropout_perf
ksivaman:fix_torch_compile_dropout_perf
January 16, 2024 15:35 57s
Activation offloading to CPU's for the Linear, Layernorm Linear and the Layernorm MLP modules Documentation #2291: Pull request #571 synchronize by ksivaman
January 16, 2024 07:37 1m 6s main
main
January 16, 2024 07:37 1m 6s
[JAX] Support SP + RoPE + GeLU Documentation #2290: Pull request #602 synchronize by mingxu1067
January 16, 2024 05:40 45s mingxu1067:mingh/sp_rope_gelu
mingxu1067:mingh/sp_rope_gelu
January 16, 2024 05:40 45s