Re:Deep Learning in JavaScript
by jsight on Friday March 29, 2024 @11:25AM
(#39865209)
If you have the option to disable the mask, isn't it then a generic nn.Block?
[ Parent
| on HN ]
The ability to disable the mask in nn.MultiHeadSelfAttention, then having nn.DecoderBlock and a nn.EncoderBlock.
Re:Deep Learning in JavaScript
by eduardoleao052 on Friday March 29, 2024 @02:07PM
(#39867149)
I think that’s it. I’ll probably add that soon
Re:Deep Learning in JavaScript
by eduardoleao052 on Friday March 29, 2024 @11:43AM
(#39865425)
It would be with some simple tweaks. For instance, the current block does not support Cross-Attention, just Self-Attention.