wangyuxin87

Does 'dcn_v2_psroi_pooling_forward' means deformable position-sensitive roi pooling?

测试精度问题

9

作者你好 我想问下这个tensorflow版本能达到原文精度吗？ 略次还是略高？

how to use tutel on Megatron Deepspeed

4

can tutel be used with Megatron Deepspeed?

Thanks for your excellent work. However, GAU is slower than the original MHSA in my implementation, **3.5s vs 0.7s**. As I simply use "from flash_pytorch import GAU" with the default...

will the flash attention embed self-extend?

1

https://github.com/datamllab/LongLM

The publication

1

Thanks for your great work. When will the code be released?