Re:Simplifying Transformer Blocks
by Bayes7 on Tuesday November 28, 2023 @06:47AM
(#38444747)
Okay, I see that for inference. But for training it shouldn't matter because I need to hold on to all my activations for my backwards pass anyways?
But yeah, fair point!
[ Parent
| on HN ]
There are no comments on this comment yet.