Learn how to push models and adapters to the Hugging Face Hub in this comprehensive 45-minute tutorial video. Explore the process of setting up LLM notebooks, configuring fine-tuning environments on Runpod and Vast AI, and downloading and loading models. Master techniques for pushing 16-bit models, merging and pushing LoRA adapters, and handling QLora (quantized) models on the Hugging Face platform. Discover best practices for model formats in training and inference scenarios. Access free resources, including a detailed notebook and presentation slides, to enhance your understanding. Gain insights into advanced fine-tuning techniques and stay updated with the latest model developments through provided links.
Overview
Syllabus
How to push models to hugging face?
Video overview
LLM Notebook Setup
Runpod Fine-tuning Setup
Vast AI Fine-tuning Setup
Downloading and loading models
Push 16-bit models to hugging face
Merging and Pushing LoRA adapters to HuggingFace
Merging and Pushing QLora quantised models to HuggingFace
Best model format for training
Best model formats for inference
Free and advanced resources
How to get video/model updates
Taught by
Trelis Research
