Skip to content

Actions: vasunvidia/TransformerEngine

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
32 workflow runs
32 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[PyTorch] Remove Megatron-LM convergence test (#1521)
Deploy nightly docs #51: Commit f090551 pushed by vasunvidia
March 10, 2025 18:54 1m 20s main
March 10, 2025 18:54 1m 20s
Add NVTX ranges to FP8 amax AR and grad output preprocessing (#1530)
Deploy nightly docs #50: Commit de06a34 pushed by vasunvidia
March 7, 2025 01:49 1m 12s main
March 7, 2025 01:49 1m 12s
Delete extra tensor objects after restoring float8 tensors (#1500)
Deploy nightly docs #49: Commit d3efaeb pushed by vasunvidia
February 28, 2025 22:05 1m 22s main
February 28, 2025 22:05 1m 22s
[PyTorch] Skip context parallelism tests if not enough GPUs (#1508)
Deploy nightly docs #48: Commit 2834e4a pushed by vasunvidia
February 27, 2025 08:05 1m 28s main
February 27, 2025 08:05 1m 28s
Minor fixes for attention (#1504)
Deploy nightly docs #47: Commit 8744188 pushed by vasunvidia
February 25, 2025 21:32 1m 28s main
February 25, 2025 21:32 1m 28s
[PyTorch] Add contiguous check for te_grouped_gemm (#1146)
Deploy nightly docs #46: Commit ddc5774 pushed by vasunvidia
September 3, 2024 23:16 1m 31s main
September 3, 2024 23:16 1m 31s
[JAX] Propagate sm_margin to the underly layernorm kernels (#1089)
Deploy nightly docs #45: Commit ba0fe9a pushed by vasunvidia
August 14, 2024 19:23 1m 15s main
August 14, 2024 19:23 1m 15s
Bug fix for num_warmup_iters=0 case (#1095)
Deploy nightly docs #44: Commit 44c8924 pushed by vasunvidia
August 12, 2024 21:29 1m 36s main
August 12, 2024 21:29 1m 36s
Add user to TE CI (#1081)
Deploy nightly docs #43: Commit 6717554 pushed by vasunvidia
August 7, 2024 15:28 1m 24s main
August 7, 2024 15:28 1m 24s
Fix an argument issue when flash_attn>=2.5.7 (#1068)
Deploy nightly docs #42: Commit 27c6342 pushed by vasunvidia
August 6, 2024 02:20 1m 9s main
August 6, 2024 02:20 1m 9s
[Paddle] Update Paddle image (#1053)
Deploy nightly docs #41: Commit 81dd6ad pushed by vasunvidia
July 30, 2024 02:52 1m 24s main
July 30, 2024 02:52 1m 24s
Update minimum CMake version (#1037)
Deploy nightly docs #40: Commit 9edcaf0 pushed by vasunvidia
July 24, 2024 18:28 2m 4s main
July 24, 2024 18:28 2m 4s
Script to run pre-commit hooks locally (#969)
Deploy nightly docs #39: Commit 7326af9 pushed by vasunvidia
July 1, 2024 12:00 1m 23s main
July 1, 2024 12:00 1m 23s
A hot fix to disable CE deadlock check (#926)
Deploy nightly docs #38: Commit d71fc94 pushed by vasunvidia
June 14, 2024 22:55 1m 34s main
June 14, 2024 22:55 1m 34s
Change norm_factor into softmax_scale and add kwarg into `DotProd…
Deploy nightly docs #37: Commit 7d576ed pushed by vasunvidia
June 13, 2024 19:07 1m 13s main
June 13, 2024 19:07 1m 13s
[C] Allow bias support for sm80/86/89 for cuDNN 9+ (#863)
Deploy nightly docs #36: Commit 223050a pushed by vasunvidia
May 27, 2024 20:02 1m 28s main
May 27, 2024 20:02 1m 28s
[JAX] Fixes for the issue with ActLuPrimitive in PAXML (#837)
Deploy nightly docs #35: Commit 87e4d6c pushed by vasunvidia
May 10, 2024 17:20 1m 22s main
May 10, 2024 17:20 1m 22s
[JAX] Generalizing Activation Primitives (#810)
Deploy nightly docs #34: Commit aad4e17 pushed by vasunvidia
May 6, 2024 18:54 1m 49s main
May 6, 2024 18:54 1m 49s
Handle the scaling factor when amax is too tiny that leads to an infi…
Deploy nightly docs #33: Commit 7acb5e2 pushed by vasunvidia
May 1, 2024 17:37 1m 40s main
May 1, 2024 17:37 1m 40s
[JAX] SwiGLU Implementation (#773)
Deploy nightly docs #32: Commit f85553e pushed by vasunvidia
April 25, 2024 06:30 1m 31s main
April 25, 2024 06:30 1m 31s
[JAX] Allow multi-dims for dgamma and dbeta in LN descriptor. (#780)
Deploy nightly docs #31: Commit aaf9354 pushed by vasunvidia
April 19, 2024 23:32 1m 22s main
April 19, 2024 23:32 1m 22s
[PyTorch] Use __torch_function__ as a class method (#783)
Deploy nightly docs #30: Commit d3552dd pushed by vasunvidia
April 16, 2024 17:24 1m 28s main
April 16, 2024 17:24 1m 28s
[JAX] Adapt latest JAX/PAX image (#744)
Deploy nightly docs #29: Commit bfe21c3 pushed by vasunvidia
April 8, 2024 19:08 1m 13s main
April 8, 2024 19:08 1m 13s
Revert "Update FA version to 2.5.6 (#714)"
Deploy nightly docs #28: Commit 47276e1 pushed by ksivaman
April 3, 2024 02:38 1m 51s main
April 3, 2024 02:38 1m 51s
Enable TP-AG overlap with return_layernorm_output (#727)
Deploy nightly docs #27: Commit c1a68f6 pushed by vasunvidia
March 26, 2024 17:28 1m 22s main
March 26, 2024 17:28 1m 22s