benchmarks for LLM tokenizers

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

Agentic Research and Evaluation Suite

Starter TS library template. Vitest, Biome, tsdown, CI publishing, JSR, Deno, etc.

SQLSync is a collaborative offline-first wrapper around SQLite. It is designed to synchronize web application state between users, devices, and the edge.

LLaMA 2-based chat for ultra-low-memory hardware with a 60 MB footprint

Replication of paper "How much do language models memorize?"

Our library for RL environments + evals

PyTorch Single Controller

pytorch-like webgpu-backed computations for ts/js

RASP-L in Haskell for my fellow rascals

Code for EMNLP-IJCNLP 2019 MRQA Workshop Paper: "Domain-agnostic Question-Answering with Adversarial Training"

📈 A small, fast chart for time series, lines, areas, ohlc & bars

Official code for "Algorithmic Capabilities of Random Transformers" (NeurIPS 2024)

Deep Learning in Javascript. Train Convolutional Neural Networks (or ordinary ones) in your browser.

Train small sequence models in your browser with WebGPU.

Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

A PyTorch native platform for training generative AI models

SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

NanoGPT (124M) in 2 minutes

🧩 The Browser Extension Framework

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