Leo Gao
leogao2
Planetary Structural Integrity Engineer
Languages
Repos
45
Stars
127
Forks
60
Top Language
Python
Loading contributions...
Top Repositories
The goal of this project is to collect beginner friendly notebooks for demonstrating alignment concepts in toy environments and make them feel more tangible.
Convert TF 2.x/Pytorch GPT2 model to original OpenAI TF 1.x GPT2 format
Repositories
45No description provided.
No description provided.
No description provided.
No description provided.
Cost aware hyperparameter tuning algorithm
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
APPS: Automated Programming Progress Standard (NeurIPS 2021)
The goal of this project is to collect beginner friendly notebooks for demonstrating alignment concepts in toy environments and make them feel more tangible.
No description provided.
No description provided.
No description provided.
No description provided.
Convert TF 2.x/Pytorch GPT2 model to original OpenAI TF 1.x GPT2 format
deep learning sandbox
No description provided.
A dataset of alignment research and code to reproduce it
Parallel computing with task scheduling
No description provided.
Academia sim
A Python library for integrating model-based and judgmental forecasting
Download, parse, and filter data from Court Listener, part of the FreeLaw projects. Data-ready for The-Pile.
Public rankings of extreme-scale models
Uses learn2learn to train a fully connected network that implements the constant function and resists SGD
An open source framework that provides a simple, universal API for building distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library.
Experiments related to scaling laws for language models.
a single source of application-specific scripts
An implementation of model parallel GPT-2 and GPT-3-style models using the mesh-tensorflow library.
Model parallel transformers in JAX and Haiku
Contrastive Language-Image Pretraining
Visually ground GPT-Neo 1.3b and 2.7b