Skip to content

Bumping flash attention version to 2.6.3 and adding option for softcap in attention and lm_head logits. #7261

Bumping flash attention version to 2.6.3 and adding option for softcap in attention and lm_head logits.

Bumping flash attention version to 2.6.3 and adding option for softcap in attention and lm_head logits. #7261

Triggered via pull request July 23, 2024 22:33
Status Cancelled
Total duration 9m 19s
Artifacts

pr-cpu.yaml

on: pull_request
Matrix: pytest-cpu
Coverage Results  /  coverage
Coverage Results / coverage
Fit to window
Zoom out
Zoom in

Annotations

2 errors
cpu-2.3.1 / pytest-cpu
Canceling since a higher priority waiting request for 'PR CPU tests-1374' exists
cpu-2.3.1 / pytest-cpu
The operation was canceled.