]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
llama : fix integer overflow during quantization (#4284)
authorGeorgi Gerganov <redacted>
Fri, 1 Dec 2023 16:42:11 +0000 (18:42 +0200)
committerGitHub <redacted>
Fri, 1 Dec 2023 16:42:11 +0000 (18:42 +0200)
commit880f57973b8e0091d0f9f50eb5ab4cd4e31582ca
treecc39f90d141084cbe36bff6357d61e9002a32d2c
parent8d6d9f033b8101f929e445cf45b39e1557ca7934
llama : fix integer overflow during quantization (#4284)

happens with multi-threaded quantization of Qwen-72B

ggml-ci
llama.cpp