lukebechtel

8 ms·

91.
by lukebechtel 1y ago
It's meant to replace the BPE tokenizer piece, so it isn't a full Language Model by itself. In fact in Gu's blog post (linked in a post below) it's mentioned that they created a Mamba model that used this in place of the
92.
by lukebechtel 1y ago
No problem! I'm still parsing it myself, but it seems promising in theory, and the result curves are impressive.
93.
by lukebechtel 1y ago
https://arxiv.org/pdf/2507.07955 paper
94.
by lukebechtel 1y ago
> H-Net demonstrates three important results on language modeling: > 1. H-Nets scale better with data than state-of-the-art Transformers with BPE tokenization, while learning directly from raw bytes. This improved scaling is even more
95.
Hierarchical Modeling (H-Nets)(cartesia.ai)
93 points by lukebechtel 1y ago|21 comments
96.
by lukebechtel 1y ago
Identifying a problem is the first step to solving it. Changing these buttons to "Rent" would force people to acknowledge what they're getting for their $60 bucks -- effectively a rental. Once the problem is identified en mas
97.
by lukebechtel 1y ago
don't mind me I'm just moloch posting https://slatestarcodex.com/2014/07/30/meditations-on-moloch/
98.
Relationships: Start with Several(lukebechtel.com)
57 points by lukebechtel 3y ago|34 comments
99.
Progressivism increases with population density (2021)(oxfordpoliticalreview.com)
28 points by lukebechtel 3y ago|72 comments
100.
GPT-4 Designed a Programming Language(lukebechtel.com)
227 points by lukebechtel 4y ago|328 comments
101.
by lukebechtel 4y ago
I have taken several masters-level courses in Machine Learning -- and even with those credentials, I cannot recommend enough Andrej's youtube series, "Neural Networks: Zero to Hero". There, he teaches you, from scratch, how