How to create a LLM on a custom dataset

language Danish English

Build and fine-tune LLMs using your own datasets. This training provides EDIH staff with the practical skills to identify AI opportunities and implement custom models for improved hub efficiency.

How to create a LLM on a custom dataset

Purpose & Scope

This on-demand training provides European Digital Innovation Hub (EDIH) staff and stakeholders with practical knowledge on building and fine-tuning LLMs tailored to specific datasets. 

Participants will learn the fundamental concepts behind LLMs, the workflow for preparing and training models, and the practical considerations for applying them in real-world scenarios. 

The training also explores real-world use cases where customised LLMs can drive innovation and improve workflow efficiency, enabling participants to identify opportunities for AI integration within their hubs.

If you want to get more information about the training, please get in touch with asco@teknologisk.dk

Learning objectives

By the end of the training, participants will be able to:

  • Describe the foundational principles behind Large Language Models (LLMs).
  • Understand the practical steps involved in training and fine-tuning LLMs on custom datasets.
  • Explore real-world use cases of customised LLMs in different sectors.

Learning outcomes

Participants completing this training will be able to:

  • Explain what a Large Language Model is, including its key components and operational principles.
  • Outline the workflow to create and fine-tune a customised LLM.
  • Identify potential applications of LLMs in their own EDIH environment.

Who should attend?

EDIH staff, AI practitioners, and professionals interested in developing or implementing custom LLM solutions for their organisation or stakeholders.

Type of Topic

Software / AI model

Type of Sector
Arable farming
Food processing
Greenhouse
Horticulture
Livestock farming
Tree Crops
Viticulture