Exploring cutting-edge LLMs: theory and practice | DSPT #116 @ Porto

Hello DSPT community we are back with another DSPT Meetup in Porto, this time at DevScope’s office!

Dive deep into the fascinating world of Large Language Models (LLMs) and Transformers in this session by Rafael Guedes, Senior ML Engineer @ Marley Spoon | Data Scientist @ ZAAI. It includes an interactive hands-on activity where you can explore one of these groundbreaking LLMs using the powerful DSPy library in Python. 🐍💻

Book your spot today and don’t forget to add the event to your calendar!

=== SCHEDULE ===
The preliminary agenda for the meetup is the following:

• 18:15 - 18:30: Get together
• 18:30 - 18:40: Welcome message
• 18:40 - 19:30: Talk + Q&A: “LLMs: A Theoretical Overview of the Transformer Architecture and the Novel Concepts of LLaMA 3, Gemma, and Mixtral” by Rafael Guedes
• 19:30 - 20:30: Networking.
• 20:30 onward: Optional Dinner

=== Talk Abstract ===

• LLMs: A Theoretical Overview of the Transformer Architecture and the Novel Concepts of LLaMA 3, Gemma, and Mixtral

This presentation focuses on providing the basic and advanced concepts of the backbone architecture of Large Language Models, theTransformer. Apart from that, it also compares and exposes the novel concepts that LLaMA 3, Mixtral and Gemma brought to the original Transformer architecture. It finishes with a hands-on activity where we explore one of these LLMs using the python library DSPy.

=== About the speaker ===

• Rafael Guedes, Senior ML Engineer @ Marley Spoon | Data Scientist @ ZAAI

Rafael is a FEUP alumni with 5 years of experience working in Data Science. Started his journey at Farfetch in 2019 solving time series forecasting problems and he is currently Senior Machine Learning Engineer at Marley Spoon working in several domains such as forecasting, recommender systems and marketing models. Lately, he has been writing AI articles focused on LLMs for ZAAI.