Evaluating LLMs with Weights & Biases and Docker Desktop

Evaluating LLMs with Weights & Biases and Docker Desktop

Explore the process of configuring Docker to run popular open-source Large Language Models (LLMs) in this 41-minute conference talk from DockerCon 2023. Learn how to integrate Weights & Biases (W&B) to generate useful charts and automate evaluations. Discover techniques for transitioning from a local development environment to GitHub actions triggered by W&B. Presented by Chris Van Pelt, Co-Founder of Weights & Biases, this talk provides insights into leveraging Docker and W&B for effective LLM evaluation and automation. Gain practical knowledge on containerization, machine learning workflows, and DevOps integration for AI/ML projects.