11 dependents
| Package | Description | Downloads/month |
|---|---|---|
| This is an open-source version of the representation engineering framework for s... | 175K | |
| Open-source SAE visualizer, based on Anthropic's published visualizer. Forked / ... | 2K | |
| A framework for evaluating sparse autoencoders | 1K | |
| Sparse probing benchmark for Sparse Autoencoders derived from the paper "Are Spa... | 1K | |
| Sorbonne University Master MIND - Large Language Models course plugin | 579 | |
| A Python client for the Axionic API. | 574 | |
| A versatile kit for training and using linear probes on neural network activatio... | 406 | |
| Transformer token flow visualizer | 204 | |
| A package for mechanistic interpretability in Neural IR | 160 | |
| Toolkit for analyzing unstructured datasets with sparse autoencoders | 155 | |
| In-depth visualizations for SAE features | 136 |
