[HN Gopher] PyTorch 2.2: FlashAttention-v2 integration, AOTInductor
___________________________________________________________________
PyTorch 2.2: FlashAttention-v2 integration, AOTInductor
Author : egnehots
Score : 10 points
Date : 2024-01-30 19:42 UTC (3 hours ago)
(HTM) web link (pytorch.org)
(TXT) w3m dump (pytorch.org)
| two_in_one wrote:
| > now supports FlashAttention-2, yielding around 2x speedups
|
| > torch.compile improvements
|
| so far 2.1 didn't work well with MoE GPT, at least in my
| implementation, due to dynamism in data flow. will check how 2.2
| does
___________________________________________________________________
(page generated 2024-01-30 23:02 UTC)