2 papers
cs.LG2025
MEIcoder: Decoding Visual Stimuli from Neural Activity by Leveraging Most Exciting Inputs
Jan Sobotka, Luca Baroni, Ján AntolÃk
Decoding visual stimuli from neural population activity is crucial for understanding the brain and for applications in brain-machine interfaces. However, such biological data is of…
cs.LG2025
Transformers Don't Need LayerNorm at Inference Time: Scaling LayerNorm Removal to GPT-2 XL and the Implications for Mechanistic Interpretability
Luca Baroni, Galvin Khara, Joachim Schaeffer +2
Layer-wise normalization (LN) is an essential component of virtually all transformer-based large language models. While its effects on training stability are well documented, its r…