fix check for fp8 capability (#3324)

* fix check for fp8 capability

* handle non-cuda compute

* reduce concurrency of tests
This commit is contained in:
Wing Lian
2025-12-22 13:58:25 -05:00
committed by GitHub
parent faaff6c792
commit efeb5a4e41
4 changed files with 22 additions and 3 deletions

View File

@@ -29,7 +29,6 @@ flex_attention: true
flex_attn_compile_kwargs:
dynamic: false
mode: max-autotune-no-cudagraphs
save_strategy: no
torch_compile: true
wandb_project: