From the course: Learn Databricks GenAI
Join today to access over 24,700 courses taught by industry experts.
Understand foundation model training - Databricks Tutorial
From the course: Learn Databricks GenAI
Understand foundation model training
- [Instructor] Another capability in LLM application lifecycle is foundation model fine-tuning. So how this is different than RAG is RAG is you're bringing your own data, you're creating embeddings, putting that data in a vector database, querying it, and augmenting the prompt. You're then passing that data to the LLM at the time that the prompt is executed. Foundation model training is similar in that you bring your own data, but it differs in that you use a training process to integrate that data into a publicly available model. So this is downstream in the Gen AI application lifecycle after you've fully explored RAG. So as with the other patterns, Databricks has a set of APIs and services that make this much easier than if you are building your own. At the time of this recording, this feature set is in public preview on AWS implementations of Databricks only. I'm sure it'll be more widely available when you're watching. So as it says here, you can use your own data to customize a…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.