llama.cpp
[SYCL] Fix the sub group size of Intel
#8106
Merged
Overview
Commits
11
Changes
View On GitHub
Commits
  • use warp_size macro for all sycl kernels
    luoyu-intel committed 1 year ago
  • fix mask of permute_sub_group_by_xor
    luoyu-intel committed 1 year ago
  • fix rms_norm with correct warp number
    luoyu-intel committed 1 year ago
  • fix rms_norm_f32
    luoyu-intel committed 1 year ago
  • fix group_norm_f32
    luoyu-intel committed 1 year ago
  • move norm to norm.cpp file
    luoyu-intel committed 1 year ago
  • sync master
    luoyu-intel committed 1 year ago
  • fix quantize bug
    luoyu-intel committed 1 year ago
  • fix mmvq's batch size
    luoyu-intel committed 1 year ago
  • add warp_size macro
    luoyu-intel committed 1 year ago
  • fix compile issue
    luoyu-intel committed 1 year ago
FAQ Terms Privacy Refunds Impressum
Loading