Andrej, please train an llm with dilated convs, please show people we were doing it wrong this whole time, I know you can do it.     See Less

Just wanna say thank you for sharing your experience -- love this from-scratch series starting from first principles!     See Less

Awesome series!     See Less

Hi @AndrejKarpathy thanks for recording this for us. I was following the whole way through, and funnily enough, I also wasn't able to beat the 1.993 that you got from this fancy hierarch     See More rk. I actually went back and tuned the single hidden layer network you mentioned above and was able to get that one to perform even better than 1.993. To be exact, I got
train 1.7930818796157837
val 1.9838893413543701
test 1.9920368194580078
just from making the network bigger and running it for longer. But don't worry, it's not that embarrassing, as I'm sure there is a setting in this hierarchical network that will product better results. It seems like there should be.    See Less

I have challenging question ( for me :) ). I made a very simple network which takes x as an input and produce y as an output. the network looks like that (y = sin(ax + b)) where a and b are     See More variables. training data is built out from sin(3x+4312). loss function is quadratic mean. using usual approaches I couldn't make it works ! what do you think the problem is?    See Less

Um, can I find Part 6 somewhere?(RNN, LSTM, GRU..) I was under the impression that the next video in the playlist is about building GPT from skretch.     See Less

That was a very great playlist, easy to understand and very helpfull, thank you very much!!     See Less

So far THE BEST lecture series I came across on YouTube. Along side learning the neural networks in this series, I have learned the PyTorch more than learning it by waching a PyTorch video s     See More 6 hrs from a youtuber.    See Less

Thank you so much for creating this video lecture series. Your passion for this topic comes through so vividly in your lectures. I learned so much from every lecture and especially appreciat     See More lectures started from the foundational concepts and built up to the state-of-the art techniques. Thank you!    See Less

Please show how to implement conv2D layers as matrix multiplication and cover the math     See Less