stephenroller

followers

following

stars

@facebookresearch

NYC

http://stephenroller.com/

Stephen Roller's starred repositories

evals

Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

Language:PythonNOASSERTION13887 258 197

triton

Development repository for the Triton language and compiler

Language:C++MIT10965 177 1154

flash-attention

Fast and memory-efficient exact attention

Language:PythonBSD-3-Clause10775 104 781

ParlAI

A framework for training and evaluating AI models on a variety of openly available dialogue datasets.

Language:PythonMIT10427 284 1544

FlexGen

Running large language models on a single GPU for throughput-oriented scenarios.

Language:PythonApache-2.09002 105 79

Megatron-LM

Ongoing research training transformer models at scale

Language:PythonNOASSERTION8553 151 500

tokenizers

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

Language:RustApache-2.08405 118 916

metaseq

Repo for external large-scale work

Language:PythonMIT6386 109 292

fairscale

PyTorch extensions for high performance and large scale training.

Language:PythonNOASSERTION2903 43 357

slurm

Slurm: A Highly Scalable Workload Manager

Language:CNOASSERTION2334 1250

The-NLP-Pandect

A comprehensive reference for all topics related to Natural Language Processing

Language:PythonCC0-1.01995 129 2

longformer

Longformer: The Long-Document Transformer

Language:PythonApache-2.01973 41 227

mkdocstrings

:blue_book: Automatic documentation from sources, for MkDocs.

Language:PythonISC1568 14 386

gpu-burn

Multi-GPU CUDA stress test

Language:C++BSD-2-Clause1156 18 67

bigscience

Central place for the engineering/scaling WG: documentation, SLURM scripts and logs, compute environment and data.

Language:ShellNOASSERTION937 36 19

filesystem_spec

A specification that python filesystems should adhere to.

Language:PythonBSD-3-Clause892 22 657

PrefixTuning

Prefix-Tuning: Optimizing Continuous Prompts for Generation

Language:Python841 8 47

madgrad

MADGRAD Optimization Method

Language:PythonMIT797 18 10

MyST-Parser

An extended commonmark compliant parser, with bridges to docutils/sphinx

Language:PythonMIT689 24 417

hck

A sharp cut(1) clone.

Language:RustUnlicense678 7 28

ConvLab-2

ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems

Language:PythonApache-2.0442 21 119

openchat

OpenChat: Easy to use opensource chatting framework via neural networks

Language:PythonApache-2.0440 16 25

lambdaprompt

λprompt - A functional programming interface for building AI systems

Language:PythonMIT367 6 5

Mephisto

A suite of tools for managing crowdsourcing tasks from the inception through to data packaging for research use.

Language:PythonMIT294 16 254

cascades

Python library which enables complex compositions of language models such as scratchpads, chain of thought, tool use, selection-inference, and more.

Language:PythonApache-2.0181 110

ParlAI_SearchEngine

A search engine for ParlAI's BlenderBot project (and probably other ones as well)

Language:PythonCC-BY-4.0132 4 11

simmc

With the aim of building next generation virtual assistants that can handle multimodal inputs and perform multimodal actions, we introduce two new datasets (both in the virtual shopping domain), the annotation schema, the core technical tasks, and the baseline models. The code for the baselines and the datasets will be opensourced.

Language:PythonNOASSERTION130 20 27

self_talk

Code and data for the paper: "Unsupervised Common Sense Question Answering with Self-Talk"

Language:PythonApache-2.078 3 2

forked-pdb

Python pdb for multiple processes

Language:PythonApache-2.028 50

dotfiles

My dotfiles

Language:Vim Script8 30