Fine-tuning with PEFT using NVIDIA NeMo in Domino

Are you ready to transform your AI capabilities with generative AI and large language models (LLMs) using your proprietary data? The potential is immense, but development can be resource-intensive and time-consuming. That's where Domino, alongside NVIDIA NeMo Framework, steps in to accelerate your journey.

The NVIDIA NeMo Framework accelerates LLM training by up to 30% (models ranging from 22 billion to 1 trillion parameters) — a quick, efficient, containerized framework for model training, evaluation, and inference. In this session, we’ll use NVIDIA NeMo Megatron — a powerful transformer developed by the Applied Deep Learning Research team at NVIDIA — to fine-tune using parameter efficient fine-tuning (PEFT) on Domino’s Enterprise AI Platform.

In this episode, we’ll walk through an end-to-end model lifecycle example, covering:

1. How Domino’s AI Project Hub integrates seamlessly with NVIDIA’s NeMo Toolkit and AI solutions.
2. Customizing a data science environment for fine-tuning tasks in Domino’s platform while optimizing resource utilization and performance.
3. Using NVIDIA NeMo Megatron to encode text prompts and generate task-specific virtual tokens.