Giordano B. Ferreira
I'm a computer scientist working at the intersection of deep learning systems, interpretability, and the philosophy of mind. My PhD at Tufts focused on spatial agent-based models across different scales. I now build and train large-scale neural networks, and I think and write about what it means for them to have representations, intentions, and perhaps something like experience.
Latest Posts
16 Apr 2026 - Do Small Models Feel Desperate? Replicating Anthropic's Emotion Vectors on Qwen3-4B
05 May 2025 - How GPUs Communicate: Fundamentals of Distributed Training with PyTorch
19 Feb 2025 - Representation Engineering: Steering Language Models Beyond Prompting