HealthCards: Exploring Text-to-Image Generation as Visual Aids for Healthcare Knowledge Democratizing and Education

Official repository for the paper: Can Large Language Models Capture Human Annotator Disagreements?

Implementation of "BitNet: Scaling 1-bit Transformers for Large Language Models" in pytorch

绝区零 | ZenlessZoneZero | 零号空洞 | 自动战斗 | 自动化 | 图片分类 | OCR识别

STREET: a multi-task and multi-step reasoning dataset

arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

Benchmarking large language models' complex reasoning ability with chain-of-thought prompting

Implementation of the paper "Multi-Agent Exploration via Self-Learning and Social Learning"

Implementation of the paper "WToE: Learning When to Explore in Multi-Agent Reinforcement Learning"

📖 Paper reading list in conversational AI.

Code for CRATE (Coding RAte reduction TransformEr).

A library for mechanistic interpretability of GPT-style language models

Official Code for Paper: RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Fastest random walks generator on networkx graphs

Karate Club: An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs (CIKM 2020)

A tiny library for larger graphs

The hub for EleutherAI's work on interpretability and learning dynamics

Instruct-tune LLaMA on consumer hardware

Code and documentation to train Stanford's Alpaca models, and generate the data.

Open Academic Research on Improving LLaMA to SOTA LLM

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

The source code and the data for ACL 2022 paper "Show Me More Details: Discovering Hierarchies of Procedures from Semi-structured Web Data"

"One cannot conceive anything so strange and so implausible that it has not already been said by one philosopher or another." ― René Descartes

A Pythonic wrapper for the Wikipedia API

🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support