Decoding base16 sequences quickly(lemire.me)

39 points by greghn 3 years ago | 11 comments

Graviton 3, Apple M2 and Qualcomm 8cx 3rd gen: a URL parsing benchmark(lemire.me)

146 points by ibobev 3 years ago | 86 comments

Parsing integers quickly with AVX-512(lemire.me)

156 points by usdogu 3 years ago | 30 comments

Random digits is not enough to uniquely identify all human beings(lemire.me)

39 points by adunk 3 years ago | 51 comments

Modern Optimizers – An Alchemist's Notes on Deep Learning(notes.kvfrans.com)

27 points by maxall4 11 months ago | 1 comments

Fast and concise probabilistic filters in Python(lemire.me)

5 points by ingve 2 years ago | 0 comments

What are transformer models and how do they work?(txt.cohere.ai)

380 points by tomcam 3 years ago | 91 comments

Cohere blog post explaining what transformer models are and how they work in natural language processing.

Deep learning gets the glory, deep fact checking gets ignored(rachel.fast.ai)

549 points by chmaynard 1 year ago | 134 comments

Rachel Thomas blog post on how a high-profile Nature paper using a Transformer to predict enzyme functions contained many errors that received far less attention than the original result.

You could have invented transformers from memory(alexwang.ai)

602 points by speckx 10 days ago | 133 comments

Blog post by Alex Wang explaining transformer layers (MLPs, gated MLPs, self-attention) as associative memory stores, with code and an interactive app.

How we hacked McKinsey's AI platform(codewall.ai)

420 points by mycroft_4221 7 months ago | 168 comments

CodeWall blog post describing how its autonomous offensive agent found vulnerabilities in McKinsey's internal AI platform Lilli.

Transformers from Scratch (2021)(e2eml.school)

594 points by jasim 3 years ago | 46 comments

A long-form tutorial explaining how transformer neural networks work from the ground up, covering matrix multiplication, attention, and backpropagation.

I built a programming language that LLMs can't use(blog.picheta.me)

244 points by dom96 8 days ago | 179 comments

Blog post by Dominik Picheta introducing KillSwitch, a programming language he designed so that large language models cannot use it.