machine language

Andrej Karpathy builds a bigram character-level language model from scratch, introducing torch.Tensor and the framework of training, sampling and loss evaluation. It is the first step toward a GPT-style Transformer in his makemore series.

Jeremy Howard's 90-minute practical tour of language models. It covers how they work, GPT-4's capabilities and limits, using the OpenAI API, code writing, data analysis and fine-tuning.

The first part of a series on programming in Jai, Jonathan Blow's in-development language. It covers compiling, writing hello world and using code from other files.

Mitchell Hashimoto walks through his setup for running a NixOS virtual machine on a Mac as his development environment, with a config repo on GitHub.

In a long live-coding session, Andreas Kling works on making the Cookie Clicker game playable in the Ladybird browser from SerenityOS. He tracks down bugs in the browser engine and fixes them.

Andreas Kling live-codes a prototype JavaScript JIT compiler for the LibJS engine used in SerenityOS and Ladybird. He builds a standalone bytecode VM, a simple interpreter, and then a simple JIT in about 80 minutes.

Andrej Karpathy builds a GPT from scratch in code, following Attention Is All You Need and GPT-2/3, in about two hours using PyTorch. He connects it to ChatGPT and assumes familiarity with his earlier makemore videos.

A reupload of Joel Spolsky's talk teaching spreadsheet users the Excel features they are probably missing, such as keyboard navigation, named ranges, tables and lookups. The original disappeared and this copy comes from the Wayback Machine.

Markus Triska tours Prolog's most distinctive features, including its use as a database and query language, rule processing, constraint solving and compile-time rewriting. It also covers concise meta-interpreters.