Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Deploy nightly docs

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
deploy_nightly_docs.yml
155 workflow run results
155 workflow run results

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Support FP16 for user buffer (#690) Deploy nightly docs #372: Commit 8255f87 pushed by ksivaman
March 7, 2024 18:01 1m 36s main
main
March 7, 2024 18:01 1m 36s
[PyTorch] Adjusted the logic of MHA and DPA to enable speculative dec… Deploy nightly docs #371: Commit b459ccc pushed by ksivaman
March 6, 2024 20:37 1m 49s main
main
March 6, 2024 20:37 1m 49s
Fix types for forward attention for JAX. (#704) Deploy nightly docs #370: Commit 728e335 pushed by ksivaman
March 6, 2024 19:16 1m 35s main
main
March 6, 2024 19:16 1m 35s
Return layernorm output in the gathered form (#697) Deploy nightly docs #369: Commit d8f678d pushed by ksivaman
March 6, 2024 14:15 1m 23s main
main
March 6, 2024 14:15 1m 23s
Disable UB bulk wgrad when weights are frozen (#702) Deploy nightly docs #368: Commit b0f6535 pushed by ksivaman
March 5, 2024 21:06 1m 36s main
main
March 5, 2024 21:06 1m 36s
Update README.rst to show the table in FP8 Convergence. (#678) Deploy nightly docs #367: Commit 3f8baf9 pushed by ksivaman
March 5, 2024 01:39 1m 29s main
main
March 5, 2024 01:39 1m 29s
[PyTorch] Update doc for checkpoint API (#695) Deploy nightly docs #366: Commit 24f78ac pushed by ksivaman
March 4, 2024 23:41 1m 37s main
main
March 4, 2024 23:41 1m 37s
Enable incremental CMake build (#684) Deploy nightly docs #365: Commit 509ab0b pushed by ksivaman
March 4, 2024 23:40 1m 16s main
main
March 4, 2024 23:40 1m 16s
[PyTorch] Use dummy amax for Float8Tensor cast (#693) Deploy nightly docs #364: Commit 4e2ce51 pushed by ksivaman
March 1, 2024 14:33 1h 13m 19s main
main
March 1, 2024 14:33 1h 13m 19s
Create a small tutorial on how to accelerate HF Llama models with Tra… Deploy nightly docs #363: Commit 0bd84ed pushed by sudhakarsingh27
March 1, 2024 07:57 1m 17s main
main
March 1, 2024 07:57 1m 17s
Slightly more explicit error message for invalid FP8 GEMM dims (#692) Deploy nightly docs #362: Commit df4bf79 pushed by timmoon10
February 29, 2024 23:40 1m 28s main
main
February 29, 2024 23:40 1m 28s
[C/PyTorch/Jax] Add support for more bias shapes (#677) Deploy nightly docs #361: Commit b8eea8a pushed by cyanguwa
February 28, 2024 18:37 2m 6s main
main
February 28, 2024 18:37 2m 6s
[JAX] Bugfix for softmax primitives accepting invalid input sharding … Deploy nightly docs #360: Commit 0404095 pushed by denera
February 28, 2024 15:45 1m 38s main
main
February 28, 2024 15:45 1m 38s
[JAX] Support various implementations of RoPE. (#655) Deploy nightly docs #359: Commit 8bba5ee pushed by denera
February 27, 2024 15:14 1m 35s main
main
February 27, 2024 15:14 1m 35s
[PyTorch] Non-reentrant mode for activation recompute (#670) Deploy nightly docs #358: Commit 82bc797 pushed by ksivaman
February 24, 2024 00:19 1m 52s main
main
February 24, 2024 00:19 1m 52s
[JAX] Refine MHA API and add DPA API (#653) Deploy nightly docs #357: Commit 9b2fed5 pushed by denera
February 22, 2024 18:09 1m 37s main
main
February 22, 2024 18:09 1m 37s
Add __version__ attribute to Python module (#675) Deploy nightly docs #356: Commit fb2f952 pushed by timmoon10
February 21, 2024 18:23 2m 0s main
main
February 21, 2024 18:23 2m 0s
[Paddle] Add RMSNorm, RoPE and SwiGLU (#599) Deploy nightly docs #355: Commit 7172509 pushed by timmoon10
February 21, 2024 17:41 2m 25s main
main
February 21, 2024 17:41 2m 25s
Move distributed tests to L1 (#673) Deploy nightly docs #354: Commit 2187a8f pushed by ksivaman
February 20, 2024 18:21 1m 12s main
main
February 20, 2024 18:21 1m 12s
Changed VERSION to 1.5.0dev Deploy nightly docs #353: Commit a3840c1 pushed by ptrendx
February 20, 2024 16:22 1m 22s main
main
February 20, 2024 16:22 1m 22s
QuickGELU activation from HuggingFace/Transformers (#475) Deploy nightly docs #352: Commit 0e116d5 pushed by ptrendx
February 17, 2024 16:25 2m 31s main
main
February 17, 2024 16:25 2m 31s
Use unoptimized layernorm kernel if pointers are not aligned (#490) Deploy nightly docs #351: Commit d5c088d pushed by ptrendx
February 17, 2024 05:31 1m 15s main
main
February 17, 2024 05:31 1m 15s
Use fused implementation of RoPE in MultiHeadAttention (#658) Deploy nightly docs #350: Commit 8d62d5c pushed by ksivaman
February 15, 2024 19:06 1m 43s main
main
February 15, 2024 19:06 1m 43s
[PyTorch] Add Float8Tensor option to avoid updating transpose cache w… Deploy nightly docs #349: Commit 1e78094 pushed by timmoon10
February 15, 2024 18:53 1m 34s main
main
February 15, 2024 18:53 1m 34s
Use arguments instead of env vars for TP comm overlap (#649) Deploy nightly docs #348: Commit bdf1afe pushed by timmoon10
February 14, 2024 21:45 1m 22s main
main
February 14, 2024 21:45 1m 22s