SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2…

NanoGPT (124M) quality in 2.67B tokens

🧩 The Browser Extension Framework

Code to reproduce "Transformers Can Do Arithmetic with the Right Embeddings", McLeish et al (2024)

An attempt to migrate Karpathy's llm.c to safe rust.

Implementation code for the paper "Parallel Structures in Pre-training Data Yield In-Context Learning"

A cross-platform browser ML framework.

Pure Typescript, dependency free, ridiculously slow implementation of GPT2 for educational purposes

Tensor computation with WebGPU acceleration

High-performance In-browser LLM Inference Engine

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

Modeling, training, eval, and inference code for OLMo

Mistral: A strong, northwesterly wind: Framework for transparent and accessible large-scale language model training, built with Hugging Face 🤗 Transformers.

Repo for the EMNLP 2023 Findings paper "Transparency at the Source"

🥉 useful helpers for react-three-fiber

Algorithms for explaining machine learning models

LLM training in simple, raw C/CUDA

Model interpretability and understanding for PyTorch

Stanford NLP Python Library for Understanding and Improving PyTorch Models via Interventions

Code and Data Repo for the CoNLL Paper -- Future Lens: Anticipating Subsequent Tokens from a Single Hidden State

Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax

A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

Run GPT model on the browser with WebGPU. An implementation of GPT inference in less than ~1500 lines of vanilla Javascript.

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

Code for the paper "Efficient Training of Language Models to Fill in the Middle"