tensorli
Learning and science
educational machine learning implementation
SHOW HN2023
90points
18comments
#714of 3,109 in 2023
Top 23%of 2023 by points
How HN took it
CuriousMostly questions and neutral comments
0 positive, 5 neutral or asking, 0 mixed, 0 critical, out of 5 comments about the project. Thread score 0.00.
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Interesting demo of how small a GPT can be
From 1 comment
Mo Gawdat has famously said that GPT-4 was something like "4300 lines of code," and that he could have written that when he was a kid. He's clearly a smart man, so, I think we…
Paul-Craft on HN
Objections
Title uses the branded name GPT for generic transformer architecture
From 2 comments
I think people are forgetting that transformer architectures are a wider field from GPT and predate GPT3 by 3+ years. Referring to transformer architectures using a branded…
cuuupid on HNFWIW: the GitHub project description says “GPT-like”. It’s the title here that dropped the “like”.
tverbeure on HN
LayerNorm is listed in README but not visible as a separate implementation
From 2 comments
Nice! The README mentions `LayerNorm` is implemented here, but while it's in the equivalence tests with PyTorch, I don't see it in the implementation.
gfaure on HNIt's part of the TensorLi definition where all the magic happens.
dauertewigkeit on HN
Requests
Port to CuPy or cuNumeric for GPU speed
From 2 comments
I wonder how easy it would be to port this library from numpy to cupy.
p1esk on HNOr cuNumeric: https://developer.nvidia.com/cunumeric
eslaught on HN
Questions
Whether GPT is subject to trademark
From 1 comment
Is GPT subject to trademark. It stands for Generative Pre-training Transformer?
quickthrower2 on HN