SAELens
Autoencoder trainer
A tool for training and analyzing sparse autoencoders to improve the understanding of neural networks and create safer AI systems.
Training Sparse Autoencoders on Language Models
526 stars
7 watching
129 forks
Language: Jupyter Notebook
last commit: 2 months ago Related projects:
Repository | Description | Stars |
---|---|---|
| An implementation of Adversarially Regularized Autoencoders for language generation and discrete structure modeling. | 400 |
| A software framework for semi-supervised generative Autoencoder models applied to single-cell data analysis. | 18 |
| Automates large-scale deep learning training on distributed clusters, providing fault tolerance and fast recovery from failures. | 1,302 |
| An implementation of Adversarial Autoencoders using PyTorch for training neural networks on structured data. | 199 |
| Trains an autoencoder to learn generic sentence representations using convolutional neural networks | 34 |
| A tool for training neural networks using large batch sizes and analyzing the trade-offs between longer training periods and better generalization performance. | 148 |
| A framework to learn simple algorithms from examples by generating and visualizing intermediate solutions | 180 |
| A Tiny AutoEncoder implementation for Stable Diffusion's image generation model | 602 |
| Trains artificial neural networks using the genetic algorithm | 241 |
| Develops a method for pre-training language understanding models by combining masked and permuted techniques, and provides code for implementation and fine-tuning. | 288 |
| A library of autoencoders for sequential data | 435 |
| A library for training and evaluating neural networks with a focus on adversarial robustness. | 921 |
| An Android app for training speech recognition in low-resource languages, utilizing FieldDB corpora and user-defined vocabularies. | 3 |
| Trains Keras models using the Genetic Algorithm | 36 |
| Trains CNNs using a genetic algorithm for classification problems where one class per sample is allowed. | 22 |