Running 73 Unlocking On-Policy Distillation for Any Model Family ๐ 73 Apply on-policy distillation to any model family
Running 85 Scaling FineWeb to 1000+ languages: Step 1: finding signal in 100s of evaluation tasks ๐ 85 Evaluate multilingual models using FineTasks
Running on CPU Upgrade Featured 2.73k The Smol Training Playbook ๐ 2.73k The secrets to building world-class LLMs
Running 3.6k The Ultra-Scale Playbook ๐ 3.6k The ultimate guide to training LLM on large GPU Clusters