The Ultra-Scale Playbook
🌌
3.93k
The ultimate guide to training LLM on large GPU Clusters
HF Books
-
-
The Smol Training Playbook
📚3.23kThe secrets to building world-class LLMs
-
FineWeb: decanting the web for the finest text data at scale
🍷1.38kExplore and download the FineWeb web‑scale text dataset
-
Unlocking On-Policy Distillation for Any Model Family
📝115Explore on-policy distillation visualization for any model
-
Distilling 100B+ Models 40x Faster with TRL
📝91TRL distillation for 100B+ teachers, 40x faster
-
Evaluation Guidebook
📝332Explore LLM benchmark scores over time
-
The Eiffel Tower Llama
📝119Explore the Eiffel Tower Llama experiment with open-source models