Forkoz 58998acc9f Fix cuda kernel for Pascal & Cuda 6/6.1 ...

When I left the other functions to use normal atomic add it seemed like a small speedup. 4.79 it/s vs 5.23 it/s

2023-03-23 07:33:57 -05:00

..

autograd_4bit.py

add more scripts and adjust code for transformer branch

2023-03-22 04:09:04 +00:00

gradient_checkpointing.py

Add gradient checkpointing

2023-03-23 08:25:29 +00:00

quant_cuda_kernel.cu

Fix cuda kernel for Pascal & Cuda 6/6.1

2023-03-23 07:33:57 -05:00

quant_cuda.cpp

add fast_4bit_matmul and auto switch 2 methods according to bottleneck

2023-03-21 08:43:07 +00:00