Running Featured 1.44k FineWeb: decanting the web for the finest text data at scale π· 1.44k Explore and download the FineWeb webβscale text dataset
Running 119 The Eiffel Tower Llama π 119 Explore the Eiffel Tower Llama experiment with open-source models
Running 122 Unlocking On-Policy Distillation for Any Model Family π 122 Explore on-policy distillation visualization for any model
Running Featured 94 Distilling 100B+ Models 40x Faster with TRL π 94 TRL distillation for 100B+ teachers, 40x faster
VibeVoice Collection Frontier Text-to-Speech Models https://microsoft.github.io/VibeVoice/ β’ 11 items β’ Updated 25 days ago β’ 264
Running on CPU Upgrade Featured 3.31k The Smol Training Playbook π 3.31k The secrets to building world-class LLMs
Running 4.05k The Ultra-Scale Playbook π 4.05k The ultimate guide to training LLM on large GPU Clusters