Doing Good Better for Technologists
An unofficial list of personal and open source projects that are relevant to EA Effective Altruism (or are at least deemed EA-adjacent by the contributors).
Created by EA Software Engineers
- Item – short description goes here
Recommended by EA Software Engineers
- AI Safety Info / Stampy - "a single-point-of-access to AI Safety, designed to route people to the information they need whatever stage of exploration they are at"
- You can interact with the wiki here
- Stampy is a chatbot written in Python which can serve as a conversational interface between a human and AI Safety Info (and more!). Currently, Stampy runs only on Rob Miles' Discord server but he's probably going to invade other online spaces too.
- DecisionTransformerInterpretability – mechanistic interpretability of Reinforcement Learning project (read more) (POC: Joseph Bloom)
- Transformer Lens – library for doing mechanistic interpretability of GPT-2 Style language models. The goal of mechanistic interpretability is to take a trained model and reverse engineer the algorithms the model learned during training from its weights (POC: Joseph Bloom)
- Tuned Lens - tools for understanding how transformer predictions are built layer-by-layer
- The Nonlinear Library - automatically turns posts from the EA Forum, Alignment Forum, and LessWrong into a podcast you can listen to. Read more here.
Contributions welcome! Read the contribution guidelines first.
This project was initially created with Cookiecutter and the custom cookiecutter-awesome 🍪
To the extent possible under law, Nicole Janeway has waived all copyright and related or neighboring rights to this work. See LICENSE.