Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the significance and potential of open Large Language Models (LLMs) in this 25-minute conference talk from Scala Days 2023 Seattle. Begin with a primer on the purpose and importance of LLMs, contextualized by the impact of OpenAI's ChatGPT. Delve into the historical development of increasingly large generalized LLM models. Examine the Dolly project, showcasing how it democratized ChatGPT-like capabilities with open models. Learn about Dolly 2.0, the world's first truly open instruction-tuned LLM, and discover how it was fine-tuned for just $100 using a small 15K dataset. Explore other significant open LLM models like Hugging Face's Open Assistant and Mosaic MPT. Understand the balance between powerful proprietary models like GPT-4 and the advantages of fine-tuning smaller open models on your own data, potentially creating a competitive edge where your data becomes your intellectual property.