Picture for Sarah E. Marzen

Sarah E. Marzen

Transformers represent belief state geometry in their residual stream

Add code
May 24, 2024
Viaarxiv icon

Complexity-calibrated Benchmarks for Machine Learning Reveal When Next-Generation Reservoir Computer Predictions Succeed and Mislead

Add code
Mar 25, 2023
Viaarxiv icon

Nearly Maximally Predictive Features and Their Dimensions

Add code
Feb 27, 2017
Figure 1 for Nearly Maximally Predictive Features and Their Dimensions
Figure 2 for Nearly Maximally Predictive Features and Their Dimensions
Figure 3 for Nearly Maximally Predictive Features and Their Dimensions
Figure 4 for Nearly Maximally Predictive Features and Their Dimensions
Viaarxiv icon

Time Resolution Dependence of Information Measures for Spiking Neurons: Atoms, Scaling, and Universality

Add code
Apr 18, 2015
Figure 1 for Time Resolution Dependence of Information Measures for Spiking Neurons: Atoms, Scaling, and Universality
Figure 2 for Time Resolution Dependence of Information Measures for Spiking Neurons: Atoms, Scaling, and Universality
Figure 3 for Time Resolution Dependence of Information Measures for Spiking Neurons: Atoms, Scaling, and Universality
Figure 4 for Time Resolution Dependence of Information Measures for Spiking Neurons: Atoms, Scaling, and Universality
Viaarxiv icon