]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: fc18425)
metal : separate scale and mask from QKT in FA kernel (#9189)
authorGeorgi Gerganov <redacted>
Mon, 26 Aug 2024 15:31:02 +0000 (18:31 +0300)
committerGitHub <redacted>
Mon, 26 Aug 2024 15:31:02 +0000 (18:31 +0300)
commit06658ad7c37f440502de2b9486ce43c47b4ec710
tree26bfb9dbe7a81d4d7087fcd831a60ad4aaa06dddtree
parentfc18425b6a8ad03847383ce2b69d52edfd49b0ffcommit | diff
metal : separate scale and mask from QKT in FA kernel (#9189)

* metal : separate scale and mask from QKT in FA kernel

* metal : ne01 check no longer necessary

* metal : keep data in local memory
ggml/src/ggml-metal.metal diff | blob | history
Packaging of ggml-org/llama.cpp