ST-LE-2024-16
Curriculum: Starter TrackWhat does ChatGPT actually know? – An application-oriented introduction into Large Language Models
Learning contents
- Explain how the fundamental architectural building blocks of LLMs work, including transformer blocks and attention mechanism
- Explain the principles of training such networks with completely unlabeled data as well as different domain- and task-adaptation approaches
- Discuss application areas and potential pitfalls of large language models
- Describe methods to set up, fine-tune, and apply LLMs to own applications
Learning outcomes
In this workshop, we want to convey a basic understanding of the functional building blocks of those deep learning networks that underlie recent chat bots, conversational agents and Large Language Models (LLM) like ChatGPT or Llama. We will use a mix of presentations, interactive quizzes and group work to understand the prerequisites regarding data and hardware, and prepared programming exercises in Python Jupyter Notebooks to deepen the understanding hands-on.
Prior knowledge
- Basic Python programming skills
- Basic knowledge of deep learning
Further reading
---
Learning contents
- Explain how the fundamental architectural building blocks of LLMs work, including transformer blocks and attention mechanism
- Explain the principles of training such networks with completely unlabeled data as well as different domain- and task-adaptation approaches
- Discuss application areas and potential pitfalls of large language models
- Describe methods to set up, fine-tune, and apply LLMs to own applications
Learning outcomes
In this workshop, we want to convey a basic understanding of the functional building blocks of those deep learning networks that underlie recent chat bots, conversational agents and Large Language Models (LLM) like ChatGPT or Llama. We will use a mix of presentations, interactive quizzes and group work to understand the prerequisites regarding data and hardware, and prepared programming exercises in Python Jupyter Notebooks to deepen the understanding hands-on.
Prior knowledge
- Basic Python programming skills
- Basic knowledge of deep learning
Further reading
---
When?
April 4, 2024, 9:00 AM - 12:00 PM
Where?
UNICOM 2
Haus 2 (Oxford)
Small seminar room, 4.3250 (3rd floor)
Mary-Somerville-Str. 2
28359 Bremen
and
Online via Zoom
Registration closed
Scholz, Karen
Email: karen.scholz@mevis.fraunhofer.de
Wenzel, Markus
Cooperating in U Bremen Research Alliance:
