Ultra-Scale-Playbook-Series

Learn how to scale LLM training efficiently on GPUs and clusters.

This is a companion repo to the YouTube video series that walks through the Hugging Face Ultra-Scale Playbook, with interactive notebooks per episode.

---

Target Audience

• Intermediate ML/LLM developers
• Comfortable with PyTorch and basic Transformer training
• New to GPU cluster-level distributed training

---

Setup Instructions

git clone https://github.com/neuralkian/ultra-scale-playbook-series.git
cd ultra-scale-playbook-series

• Use colab or local development to access notebooks!