Simple, fast primitives for building pdf viewers. maintained by @unriddle-ai

A powerful and simple asynchronous task management system that divides complex tasks into subtasks, processes them concurrently using o1 models, and synthesizes the results.

Talk with your notes in Claude. RAG over your Apple Notes using Model Context Protocol.

Sidecar is the AI brains for the Aide editor and works alongside it, locally on your machine

Light-weight training code for saprmarks/dictionary_learning library

This is the template I use to start new full-stack projects.

Code for Preventing Language Models From Hiding Their Reasoning, which evaluates defenses against LLM steganography.

[ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs.

Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.

Training Sparse Autoencoders on Language Models

This is the official repository for Auto-RAG.

A LLM-based Agent that predict its tasks proactively.

Auto Interpretation Pipeline and many other functionalities for Multimodal SAE Analysis.

Vivaria is METR's tool for running evaluations and conducting agent elicitation research.

hide text in text

Steganographic text via language models

Let your Claude able to think

Generate a comprehensive review from an arXiv paper, then turn it into a blog post. This project powers the website below for the HuggingFace's Daily Papers (https://huggingface.co/papers).

Direct Preference Optimization from scratch in PyTorch