john-hewitt/backpacks-flash-attn

The original Backpack Language Model implementation, a fork of FlashAttention

PythonBSD-3-Clause

Issues

EOFError when np.load
#4 opened a year ago by Tangyiming205069
0
CUDA version in nvcc -V and nvidia-smi different
#2 opened a year ago by Tangyiming205069
0
Hydra's cascading AttributeErrors
#3 opened a year ago by nuankw
3
Fix some figures in the paper from arXiv
#1 opened 2 years ago by JackeyWang777
0