EleutherAI

the-pile

pyfra

Python Research Framework

openwebtext2

magiCARP

One stop shop for all things carp

mp_nerf

Massively-Parallel Natural Extension of Reference Frame

polyglot-data

data related codebase for polyglot project

exploring-contrastive-topology

text-generation-testing-ui

Web app for demoing the EAI models

NeMo

NeMo: a toolkit for conversational AI

pile_dedupe

Pile Deduplication Code

pilev2

architecture-objective

tagged-pile

Part-of-Speech Tagging for the Pile and RedPajama

multimodal-fid

GPTeacher

A collection of modular datasets generated by GPT-4, General-Instruct - Roleplay-Instruct - Code-Instruct - and Toolformer

architecture-experiments

Repository to host architecture experiments and development using Paxml and Praxis

FLAN

LLM-Markov-Chains

Project github for LLM Markov Chains Project

composer

Train neural networks up to 7x faster

BIG-bench

Beyond the Imitation Game collaborative benchmark for measuring and extrapolating the capabilities of language models

eleutherai-instruct-dataset

A large instruct dataset for open-source models (WIP).

examples

Mosaicml example benchmarks + LLM scripts

alignment-reader

Search and filter through alignment literature

maxtext

A simple, performant and scalable Jax LLM!

minetest_game

Minetest Game - The default game for the Minetest engine [https://github.com/minetest/minetest/]

CommonLoopUtils

[WIP] a version of CLU with WandB logging added.

gaia

Hugging Face and Pyserini interoperability

oslo-1

OSLO: Open Source for Large-scale Optimization

Plenoxels_FreeNerf

implmentation of Plenoxels radiance fields without neural networks, with free nerf strategy

t-zero

Reproduce results and replicate training fo T0 (Multitask Prompted Training Enables Zero-Shot Task Generalization)