New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

feat: support kFactor 8 used in mma tensor op tile iterator #1512

Open

gavinchen430 wants to merge 1 commit into NVIDIA:main from gavinchen430:feat_kfactor_8

gavinchen430 commented Apr 29, 2024

support Layout::kFactor with 8
loading data from shared memory:

0	16	32	48	64	80	96	112
T0	T1	T2	T3	T4	T5	T6	T7
T9	T8	T11	T10	T13	T12	T15	T14
T18	T19	T16	T17	T22	T23	T20	T21
T27	T26	T25	T24	T31	T30	T29	T28


          feat: support kFactor 8 used in mma tensor op tile iterator

5c2f535

Author

gavinchen430 commented May 7, 2024

@hwu36 Could you help review my pull request, please

gavinchen430 mentioned this pull request

[feat]: Support weight only gemm with 2bit NVIDIA/TensorRT-LLM#1568

Open

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment