hrx: the attention kernel derives the transposed pitch from its key count (fixes a per-key-count recompile that merged with #289) - #291

hrx: the attention kernel derives the transposed pitch from its key count (fixes a per-key-count recompile that merged with #289)#291

bong-water-water-bong merged 1 commit into

1bit/hrx-vulkan-patched1bit-MONSTER/llama.cpp:1bit/hrx-vulkan-patchedfrom

1bit/1989-fa-pitch-runtime1bit-MONSTER/llama.cpp:1bit/1989-fa-pitch-runtimeCopy head branch name to clipboard