]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 25f1045)
CANN: Resolve soft_max precision issue (#15730)
authorhipudding <redacted>
Tue, 2 Sep 2025 09:12:37 +0000 (17:12 +0800)
committerGitHub <redacted>
Tue, 2 Sep 2025 09:12:37 +0000 (17:12 +0800)
commit9961d244f2df6baf40af2f1ddc0927f8d91578c8
tree33d6218382d989c43e718c60c5477491c60b999etree
parent25f1045f07cf0daf667d63e35618842e3174a8c7commit | diff
CANN: Resolve soft_max precision issue (#15730)

Previously, the slope tensor was set to fp16 to improve efficiency.
While this worked correctly in FA, it caused precision issues in soft_max.
This change applies different data types for different operators
to balance both accuracy and performance.
ggml/src/ggml-cann/aclnn_ops.cpp diff | blob | history
Packaging of ggml-org/llama.cpp