As time progresses, the importance of context, tokens and massive cleaned data seems to have the most potential for new developments. Gone are the times, where companies would just pump parameter numbers. Everything will probably scale linearly.
Thanks for breaking down the paper Jay. I like the performance based on multi input data types and Gato being able to perform beta. This is a cool use case towards AGI
Hello Jay As you explained and taught word2vec well in the article "The Illustrated Word2vec". Please prepare an educational clip about attention and transformer that accurately defines the concepts of attention and transformer and tells why the transformer works?
This feels like sticking together pieces of narrow models and attach them with duct tape and see how it fares... I'm afraid a model with 100 billion parameters won't perform as well as a system that is a bit better designed and that uses high level symbolic representations. It's high time we concentrate on a different architecture instead of doing more of the same.
As time progresses, the importance of context, tokens and massive cleaned data seems to have the most potential for new developments. Gone are the times, where companies would just pump parameter numbers. Everything will probably scale linearly.
Thanks for breaking down the paper Jay. I like the performance based on multi input data types and Gato being able to perform beta. This is a cool use case towards AGI
Hello Jay
As you explained and taught word2vec well in the article "The Illustrated Word2vec". Please prepare an educational clip about attention and transformer that accurately defines the concepts of attention and transformer and tells why the transformer works?
awesome video! what software do you use to edit and record your screen?
Thanks Jay for this explanation. Can you create some videos in which you teach us how to create practical applications
if i want to use Gato how do I go about that?🤔
i want to start making soft and rapping i have an acer laptop. does soft soft co with good softs already?
It's the mic
This feels like sticking together pieces of narrow models and attach them with duct tape and see how it fares...
I'm afraid a model with 100 billion parameters won't perform as well as a system that is a bit better designed and that uses high level symbolic representations.
It's high time we concentrate on a different architecture instead of doing more of the same.
save... ( how can image-line call their software a professional DAW... when the programming belongs to kindergarden... ) Last ti I ever