Deprecated: Using null as an array offset is deprecated, use an empty string instead in /home/u876752588/domains/capria.vc/public_html/wp-content/plugins/jet-engine/includes/components/blocks-views/dynamic-content/manager.php on line 113
Access to the video and presentation deck.
Abstract:
Capria’s lead GenAI developer Pranav Chellagurki joined our AI guru Vijay Mital, Microsoft’s Corporate VP of AI Architecture & Strategy, for a GenAI Network (GAIN) workshop. The session explored the technical intricacies of LLMs, fine-tuning, and how you can harness their power to address your business challenges. The workshop catered to both technical and non-technical founders / CxOs, offering a technically oriented discussion with examples for people to comprehend. He also introduced Retrieval Augmented Generation (RAG) and what it can do vs. fine-tuning, and vice versa.
Key topics covered in the workshop:
- Understanding Large Language Models (Language Models 101):
-
- Why do language models work?
- How are language models trained?
- The Significance of Fine Tuning:
-
- Defining fine-tuning and its purpose
- Distinguishing between pre-training and fine-tuning
- Exploring the reasons for fine-tuning, especially in the context of moving beyond auto-complete to create accurate Question and Answering systems
- The Mechanics of Model Fine Tuning:
-
- A deeper dive into the technical aspects of fine-tuning
- Practical insights into how to fine-tune a model
- Considerations for RAG and Fine Tuning:
-
- A quick introduction to Retrieval Augmented Generation
- Identifying scenarios where fine-tuning may not be advisable (RAG vs prompting vs Fine Tuning)
- Understanding when and why you should opt for fine-tuning
