Dive deep into Meta AI's latest release of Llama-3 in 8B and 70B sizes in this comprehensive 24-minute video tutorial. Explore the model's statistics, benchmarks, and key information while learning how to run it yourself. Examine the Meta AI blog post, analyze the model card details for both sizes, and understand the intended use cases. Discover which cloud providers offer Llama-3 and compare its performance benchmarks. Learn about the scaling up of pre-training and how to download Llama-3 from Hugging Face, including important license conditions. Get a sneak peek at the upcoming 405B model and explore various deployment options, including Ollama, Hugging Chat, Together AI, and Google Colab. Follow along with provided code examples and resources to gain hands-on experience with this cutting-edge language model.
Overview
Syllabus
Intro
Meta AI Blog: Llama 3
Llama 3 Model Card: 8B and 70B
Intended Use Cases
Cloud Providers available for Llama 3
Llama 3 Benchmarks
Scaling up Pre-training
Downloading Llama 3 on Hugging Face
License Conditions
Llama 3 405B Model: Sneak Peek
Code Time: Ollama
Llama 3 on Hugging Chat
Different Options on Deploying Llama 3
Llama 3 on Together AI
Llama 3 on Colab
Taught by
Sam Witteveen
