The bottleneck was never the code(thetypicalset.com)

538 points by Anon84 5 months ago | 349 comments

Blog post reflecting on how coding agents like Codex quickly built a long-postponed structured-generation experiment, arguing code was never the real bottleneck.

Show HN: A plain-text cognitive architecture for Claude Code(lab.puga.com.br)

114 points by marciopuga 7 months ago | 34 comments

Towards Monosemanticity: Decomposing Language Models with Dictionary Learning(transformer-circuits.pub)

56 points by goodmachine 3 years ago | 3 comments

WorstFit: Unveiling Hidden Transformers in Windows ANSI(blog.orange.tw)

342 points by notmine1337 1 year ago | 126 comments

Orange Tsai's blog post presenting WorstFit, a class of Windows vulnerabilities arising from Best-Fit character conversion between UTF-16 and ANSI code pages.

Diffusion models explained simply(seangoedecke.com)

123 points by onnnon 1 year ago | 25 comments

Transformer^2: Self-Adaptive LLMs(sakana.ai)

130 points by hardmaru 1 year ago | 43 comments

Understand how transformers work by demystifying the math behind them(osanseviero.github.io)

408 points by LaserPineapple 2 years ago | 94 comments

A hackerllama blog post walking through an end-to-end worked example of transformer math using small random vectors and matrices.

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.

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.

Transformers Explained Visually(poloclub.github.io)

576 points by aray07 1 month ago | 85 comments

Transformer Explainer is an interactive in-browser visualization that runs GPT-2 and shows how embeddings, multi-head self-attention, and MLP layers process text.

RustGPT: A pure-Rust transformer LLM built from scratch(github.com)

357 points by amazonhut 1 year ago | 170 comments

GitHub repository for RustGPT, a transformer-based large language model written entirely in Rust from scratch.

I built a neural network using DOOM games(github.com)

4 points by falperc 4 days ago | 1 comments

GitHub repo for Doomformer, which computes neural networks including transformers using DOOM game mechanics like voodoo dolls, conveyor belts, lava and doors, verified bit-exact.