Reading the Mind of an LLM

This talk distills the latest research from Anthropic, DeepMind, and OpenAI to present the current state of the art in LLM interpretability.
We’ll start with the modern interpretation of embeddings as sparse, monosemantic features living in high-dimensional space.
From there, we’ll explore emerging techniques such as circuit tracing and attribution graphs, and see how researchers reconstruct the computational pathways behind behaviors like multilingual reasoning, refusals, and hallucinations.
We’ll also look at new evidence suggesting that models may have limited forms of introspection—clarifying what they can, and crucially cannot, reliably report about their internal processes.
Finally, we’ll connect these “microscopic” insights to real engineering practice: how feature-level understanding can improve debugging, safety, and robustness in deployed AI systems, and where current methods still fall short.
Speaker: Emanuele Fabbiani
Co-founder and head of AI at xtream (acquired by TeamSystem), Professor at Catholic University of Milan
Milan, Italy
Emanuele is an engineer, researcher, and entrepreneur passionate about AI.
After earning his PhD in time series forecasting and working as a guest researcher at EPFL in Lausanne, he is now co-founder and Head of AI at xtream, a company that applies AI to solve complex business challenges. xtream was acquired by TeamSystem in September 2025.
Emanuele is also a professor of AI at the Catholic University of Milan and has been a guest lecturer at Italian, Swiss, and Polish universities.
He has published 8 papers in international journals and contributed to over 30 international conferences, including AMLD Lausanne, ODSC London, WeAreDevelopers Berlin, PyData Berlin, PyData Paris, PyCon Italy, the Swiss Python Summit in Zurich, and Codemotion Milan.

Know what’s Happening Next — before everyone else does.

Get App

Host or Publisher Data Science Milan