Skip to content
Files Size

WIP: [GPU] Use FP32 accumulator for QK multiplication for 2nd+ token calculation in PagedAttention #158851

Sign in to view logs

WIP: [GPU] Use FP32 accumulator for QK multiplication for 2nd+ token calculation in PagedAttention

WIP: [GPU] Use FP32 accumulator for QK multiplication for 2nd+ token calculation in PagedAttention #158851

Triggered via pull request January 24, 2025 15:36
@sshlyapnsshlyapn
opened #28673
sshlyapn:paged_attention_2nd_token_fp32_acc
Status Success
Total duration 26s
Artifacts –

files_size.yml

on: pull_request
Check_Files_Size
14s
Check_Files_Size
Fit to window
Zoom out
Zoom in