github/ROCm
1
1
Fork 0
mirror of https://github.com/ROCm/ROCm.git synced 2026-01-07 22:03:58 -05:00
Code Issues Packages Projects Releases Wiki Activity

clean up formatting in FA2 page (#5795)

Browse Source
This commit is contained in:
peterjunpark
2025-12-19 09:21:41 -05:00
committed by GitHub
parent 52c0a47e84
commit 7455fe57b8
1 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
docs/how-to/rocm-for-ai/inference-optimization/model-acceleration-libraries.rst
View File

@@ -41,16 +41,16 @@ Installing Flash Attention 2
`Flash Attention <https://github.com/Dao-AILab/flash-attention>`_ supports two backend implementations on AMD GPUs. `Flash Attention <https://github.com/Dao-AILab/flash-attention>`_ supports two backend implementations on AMD GPUs.
* `Composable Kernel (CK) <https://github.com/ROCm/composable_kernel>`_ - the default backend * `Composable Kernel (CK) <https://github.com/ROCm/composable_kernel>`__ - the default backend
* `OpenAI Triton <https://github.com/triton-lang/triton>`_ - an alternative backend * `OpenAI Triton <https://github.com/triton-lang/triton>`__ - an alternative backend
You can switch between these backends using the environment variable FLASH_ATTENTION_TRITON_AMD_ENABLE: You can switch between these backends using the environment variable ``FLASH_ATTENTION_TRITON_AMD_ENABLE``:
FLASH_ATTENTION_TRITON_AMD_ENABLE="FALSE" ``FLASH_ATTENTION_TRITON_AMD_ENABLE="FALSE"``
→ Use Composable Kernel (CK) backend (FlashAttention 2) → Use Composable Kernel (CK) backend (Flash Attention 2)
FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE" ``FLASH_ATTENTION_TRITON_AMD_ENABLE="TRUE"``
→ Use OpenAI Triton backend (FlashAttention 2) → Use OpenAI Triton backend (Flash Attention 2)
To install Flash Attention 2, use the following commands: To install Flash Attention 2, use the following commands:
@@ -74,7 +74,7 @@ For detailed installation instructions, see `Flash Attention <https://github.com
Benchmarking Flash Attention 2 Benchmarking Flash Attention 2
------------------------------ ------------------------------
Benchmark scripts to evaluate the performance of Flash Attention 2 are stored in the `flash-attention/benchmarks/` directory. Benchmark scripts to evaluate the performance of Flash Attention 2 are stored in the ``flash-attention/benchmarks/`` directory.
To benchmark the CK backend To benchmark the CK backend