]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
sync : llama (mul_mat_id + get_rows kernels, typos) (#649)
authorGeorgi Gerganov <redacted>
Wed, 13 Dec 2023 19:53:20 +0000 (21:53 +0200)
committerGitHub <redacted>
Wed, 13 Dec 2023 19:53:20 +0000 (21:53 +0200)
commit5c7bd24f848aae5b8d348c4a7186c9c533f29c39
tree12657467110e485aa09de1fe5f1b8ce290529fec
parent5bf85a52211a2e4a17a257eed8a24ac6ea75627a
sync : llama (mul_mat_id + get_rows kernels, typos) (#649)

* sync : llama (mul_mat_id + get_rows kernels, typos)

ggml-ci

* cuda : restore im2col

ggml-ci

* metal : fix accuracy of dequantization kernels

* cuda : restore correct im2col kernel

ggml-ci

* metal : fix moe test by reducing the expert size

ggml-ci

* cuda : fix bin bcast when src1 and dst have different types

---------

Co-authored-by: slaren <redacted>
include/ggml/ggml-alloc.h
include/ggml/ggml.h
src/ggml-cuda.cu
src/ggml-metal.m
src/ggml-metal.metal
src/ggml-quants.c
src/ggml.c
tests/test-backend-ops.cpp
tests/test-grad0.cpp
tests/test-quantize-perf.cpp