]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 4feb4b3)
ggml : importance matrix support for legacy quants (#4969)
authorKawrakow <redacted>
Tue, 16 Jan 2024 17:51:26 +0000 (19:51 +0200)
committerGitHub <redacted>
Tue, 16 Jan 2024 17:51:26 +0000 (19:51 +0200)
commit334a835a1ccc8106a5fa355683a965efb1bfa24b
tree83172d25be464c4f041dd43eb56592c372b2c784tree
parent4feb4b33eeb1756e46084a4db9230b279af1a480commit | diff
ggml : importance matrix support for legacy quants (#4969)

* imatrix: adding support for legacy quants

* imatrix: guard Q4_0/Q5_0 against ffn_down craziness

---------

Co-authored-by: Iwan Kawrakow <redacted>
ggml-quants.c diff | blob | history
ggml-quants.h diff | blob | history
ggml.c diff | blob | history
llama.cpp diff | blob | history
Packaging of ggml-org/llama.cpp