xiaoda99's repositories
paxml
Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.
transformers
🤗 Transformers: State-of-the-art Natural Language Processing for Pytorch, TensorFlow, and JAX.
triton
Development repository for the Triton language and compiler
mesh-transformer-jax
Model parallel transformers in JAX and Haiku
rerope
Rectified Rotary Position Embeddings
blockwise-parallel-transformer
Blockwise Parallel Transformer for Long Context Large Models
seqio
Task-based datasets, preprocessing, and evaluation for sequence models.
FastChat
An open platform for training, serving, and evaluating large languages. Release repo for Vicuna and FastChat-T5.
unseal
Mechanistic Interpretability for Transformer Models
PyFunctional
Python library for creating data pipelines with chain functional programming
PySvelte
A library for bridging Python and HTML/Javascript (via Svelte) for creating interactive visualizations
category-theory-resources
Resources for learning Category Theory for an enthusiast
AI-Writer
用魔改 GPT 生成网文小说。AI 写小说。Tuned GPT for novel generation.
LARC
Language-annotated Abstraction and Reasoning Corpus
gpt-neo
An implementation of model parallel GPT-2 and GPT-3-style models using the mesh-tensorflow library.
swarm-jax
Swarm training framework using Haiku + JAX + Ray for layer parallel transformer language models on unreliable, heterogeneous nodes
ARC
The Abstraction and Reasoning Corpus
pa-transformer
The init version of experiment pipeline of the PA-Transformer model.
captum
Model interpretability and understanding for PyTorch
bypasswaf
关于安全狗和云锁的自动化绕过脚本
identidock
Identidock Example Application from Using Docker
using_docker_in_dev
Code for "Using Docker In Development" Chapter of Using Docker
pytorch-openai-transformer-lm
A PyTorch implementation of OpenAI's finetuned transformer language model with a script to import the weights pre-trained by OpenAI