Skip to content

Actions: erhoo82/TransformerEngine

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
33 workflow runs
33 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[JAX] Use default factory for not sharing mutable default values (#1364) Deploy nightly docs #45: Commit e4c99b0 pushed by erhoo82
December 12, 2024 05:32 1m 20s main
main
December 12, 2024 05:32 1m 20s
Convert non-kernel cuda files to cpp (#1322) Deploy nightly docs #44: Commit 68adf45 pushed by erhoo82
November 12, 2024 06:07 1m 5s main
main
November 12, 2024 06:07 1m 5s
[PyTorch] Userbuffers support in operation-based API (#1142) Deploy nightly docs #43: Commit 095b27d pushed by erhoo82
November 6, 2024 07:22 1m 4s main
main
November 6, 2024 07:22 1m 4s
Support using fp16 master weights and fp16/fp8 optimizer states in Fu… Deploy nightly docs #42: Commit 05c0fb0 pushed by erhoo82
November 1, 2024 18:11 1m 10s main
main
November 1, 2024 18:11 1m 10s
[PyTorch] Move block_table argument to FA varlen function (#1222) Deploy nightly docs #41: Commit 10cceae pushed by erhoo82
October 3, 2024 20:35 2m 7s main
main
October 3, 2024 20:35 2m 7s
Update list of CI users (#1198) Deploy nightly docs #40: Commit a68acd7 pushed by erhoo82
September 24, 2024 04:55 1m 1s main
main
September 24, 2024 04:55 1m 1s
Restore compatibility with Python 3.8 (#1189) Deploy nightly docs #39: Commit 0c74535 pushed by erhoo82
September 21, 2024 00:52 1m 37s main
main
September 21, 2024 00:52 1m 37s
[PyTorch] Improve logging/messaging in attention (#1074) Deploy nightly docs #38: Commit 121ff62 pushed by erhoo82
August 6, 2024 17:14 1m 21s main
main
August 6, 2024 17:14 1m 21s
Initialize output tensors to 0 for THD (temporary) (#1009) Deploy nightly docs #37: Commit 238df4c pushed by erhoo82
July 19, 2024 23:36 1m 21s main
main
July 19, 2024 23:36 1m 21s
Add cuDNN sliding window and set_deterministic_algorithm (#992) Deploy nightly docs #36: Commit 8e039fd pushed by erhoo82
July 11, 2024 22:23 1m 14s main
main
July 11, 2024 22:23 1m 14s
Fix local cpp tests after inplace build (#911) Deploy nightly docs #35: Commit 78efc93 pushed by erhoo82
June 12, 2024 18:56 1m 11s main
main
June 12, 2024 18:56 1m 11s
Fix minor security vulnerability when triggering CI (#898) Deploy nightly docs #34: Commit c6ce2b8 pushed by erhoo82
June 8, 2024 04:55 1m 8s main
main
June 8, 2024 04:55 1m 8s
[JAX] Fixes for the issue with ActLuPrimitive in PAXML (#837) Deploy nightly docs #33: Commit 87e4d6c pushed by erhoo82
May 10, 2024 23:43 1m 21s main
main
May 10, 2024 23:43 1m 21s
Add SM margin to LayerNorm in inference (#772) Deploy nightly docs #32: Commit 5d34b2a pushed by erhoo82
April 15, 2024 19:17 1m 20s main
main
April 15, 2024 19:17 1m 20s
Fix undefined symbol issue for transformer_engine::getenv (#763) Deploy nightly docs #31: Commit 1b20f2d pushed by erhoo82
April 11, 2024 01:18 1m 57s main
main
April 11, 2024 01:18 1m 57s
[JAX] Adapt latest JAX/PAX image (#744) Deploy nightly docs #30: Commit bfe21c3 pushed by erhoo82
April 9, 2024 17:26 1m 37s main
main
April 9, 2024 17:26 1m 37s
userbuffer: support fp8 buffer for individual overlap instance (#750) Deploy nightly docs #29: Commit 7d8ef9b pushed by erhoo82
April 5, 2024 22:09 1m 32s main
main
April 5, 2024 22:09 1m 32s
Fixing potential integer overflow on sequence counter (#729) Deploy nightly docs #28: Commit e1e2b76 pushed by erhoo82
April 4, 2024 03:31 1m 32s main
main
April 4, 2024 03:31 1m 32s
[PyTorch] Fix backward compatibility with checkpoint API (#740) Deploy nightly docs #27: Commit 12cbd86 pushed by erhoo82
March 31, 2024 06:09 1m 37s main
main
March 31, 2024 06:09 1m 37s
Enable TP-AG overlap with return_layernorm_output (#727) Deploy nightly docs #26: Commit c1a68f6 pushed by erhoo82
March 23, 2024 19:38 1m 11s main
main
March 23, 2024 19:38 1m 11s
TP-RS overlap with send/recv ring-exchange (#724) Deploy nightly docs #25: Commit b855656 pushed by erhoo82
March 21, 2024 23:04 1m 57s main
main
March 21, 2024 23:04 1m 57s
Llama accelerate tutorial (#720) Deploy nightly docs #24: Commit c38779b pushed by erhoo82
March 20, 2024 21:18 1m 30s main
main
March 20, 2024 21:18 1m 30s
Ln force no weight sharding (#715) Deploy nightly docs #23: Commit ffa2447 pushed by erhoo82
March 14, 2024 21:48 2m 1s main
main
March 14, 2024 21:48 2m 1s
[Common] Fix build errors with recent cuDNN frontend versions (#696) Deploy nightly docs #22: Commit a38b291 pushed by erhoo82
March 12, 2024 05:34 1m 24s main
main
March 12, 2024 05:34 1m 24s
[PyTorch] Update doc for checkpoint API (#695) Deploy nightly docs #21: Commit 24f78ac pushed by erhoo82
March 5, 2024 00:12 2m 3s main
main
March 5, 2024 00:12 2m 3s