Tim Dettmers 7dc198feb7 Added 32-bit optimizer for bfloat16 gradients.		2023-04-17 18:01:49 -07:00
..
common.cpp	Fixed 2^31 max size issue for cpu blockwise quant.	2022-09-11 11:55:09 -07:00
common.h	Fixed 2^31 max size issue for cpu blockwise quant.	2022-09-11 11:55:09 -07:00
cpu_ops.cpp	Remove trailing whitespace & ensure newline at EOF	2022-10-27 13:11:29 +02:00
cpu_ops.h	Fixed 2^31 max size issue for cpu blockwise quant.	2022-09-11 11:55:09 -07:00
kernels.cu	Added 32-bit optimizer for bfloat16 gradients.	2023-04-17 18:01:49 -07:00
kernels.cuh	First draft of NF4.	2023-04-02 16:10:35 -07:00
ops.cu	Added 32-bit optimizer for bfloat16 gradients.	2023-04-17 18:01:49 -07:00
ops.cuh	First draft of NF4.	2023-04-02 16:10:35 -07:00
pythonInterface.c	Added 32-bit optimizer for bfloat16 gradients.	2023-04-17 18:01:49 -07:00