diff options
| author | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-20 15:20:50 +0300 |
|---|---|---|
| committer | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-22 12:02:52 +0300 |
| commit | e05cca9ef652eee7b42927485a3821b14e3c565f (patch) | |
| tree | d75bda9704a09ad75357d12996ad3d9ea28b3d78 /ggml-cuda/scale.cu | |
| parent | 36374ab37dac8fadb634f802aaa3ee7b816fe727 (diff) | |
bitnet(scale in a separate tensor): CPU improvements
Arrange Q8 quants in blocks of 128 and adapt iqk_mul_mat
to deal with that. This improves PP speef by a few percent.
Diffstat (limited to 'ggml-cuda/scale.cu')
0 files changed, 0 insertions, 0 deletions