Low-rank Inference from Neural Trajectories (LINT) Code for the paper "Extracting computational mechanisms from neural data using low-rank RNNs", accepted at NeurIPS 2022. Networks are fitted by the scripts in training_scripts (all weights already in models/), and analyzed in the notebooks.