Skip to content
conftrace_

neural network activation

7 papers

Explore in graph

Co-occurring keywords · papers mentioning both

activation function (2) sparse autoencoder (2) neural network (2) feature learning (2) activation mechanism (1) end-to-end training (1) pre-trained language model (1) causal analysis (1) pre-training strategy (1) feature interpretability (1)

Papers

Efficient Dictionary Learning with Switch Sparse Autoencoders ICLR 2025 Repetition Neurons: How Do Language Models Produce Repetitions? NAACL 2025 Does Large Language Model Contain Task-Specific Neurons? EMNLP 2024 Dual-Perspective Activation: Efficient Channel Denoising via Joint Forward-Backward Criterion for Artificial Neural Networks NeurIPS 2024 Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning NeurIPS 2024 EigenGame: PCA as a Nash Equilibrium ICLR 2021 Slow, Decorrelated Features for Pretraining Complex Cell-like Networks NeurIPS 2009