conceptofmind

PaLM

An open-source implementation of Google's PaLM models

LaMDA-rlhf-pytorch

Open-source pre-training implementation of Google's LaMDA in PyTorch. Adding RLHF similar to ChatGPT.

toolformer

t5-pytorch

Implementation of Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer in PyTorch.

hf_fsdp

A very basic fsdp wrapper for HF.

mpire

A Python package for easy multiprocessing, but faster than multiprocessing

PaLM-rlhf-pytorch

Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM

conceptofmind

Config files for my GitHub profile.

edgar-crawler

Download financial reports from SEC's EDGAR quickly. Extract clean textual data from specific item sections and bootstrap your financial research. Software from the research paper published in ECONLP 2021.

fastText

Library for fast text representation and classification.

GPTQ-for-LLaMa

4 bits quantization of LLaMa using GPTQ

magic-wormhole

get things from one computer to another, safely

pynecone

🕸 Web apps in pure Python 🐍

scaled-rope

unsloth-bert

attention_sinks

Extend existing LLMs way beyond the original training length with constant memory usage, without retraining

cc_net

Tools to download and cleanup Common Crawl data

CoH

common-pile

Repo to hold code and track issues for the collection of permissively licensed data

fastRAG

Efficient Retrieval Augmentation and Generation Framework

hidet

An open-source efficient deep learning framework/compiler, written in python.

HojiChar

The robust text processing pipeline framework enabling customizable, efficient, and metric-logged text preprocessing.

jax-dataloader

Pytorch-like dataloaders in JAX.

langchain

⚡ Building applications with LLMs through composability ⚡

pytriton

PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.

self-instruct

Aligning pretrained language models with instruction data generated by themselves.

streaming-llm

Efficient Streaming Language Models with Attention Sinks

torchscale

Transformers at any scale

vocos

Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis

wikipedia-downloader

Downloads 2020 English Wikipedia articles as plaintext