Running on CPU Upgrade 208 The Synthetic Data Playbook: Generating Trillions of the Finest Tokens π 208 Explore synthetic data experiments as an interactive bookshelf
Building Featured 69 QED-Nano: Teaching a Tiny Model to Prove Hard Theorems π 69 Who needs 1T parameters? Olympiad proofs with a 4B model
Running 56 Bringing paper to life: A modern template for scientific writing π 56 Explore and download a modern scientific paper template
Running 3.75k The Ultra-Scale Playbook π 3.75k The ultimate guide to training LLM on large GPU Clusters
Running 91 Scaling FineWeb to 1000+ languages: Step 1: finding signal in 100s of evaluation tasks π 91 Evaluate multilingual models using FineTasks