brando90

AutoGPT

AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

nanoGPT

The simplest, fastest repository for training/finetuning medium-sized GPTs.

stanford_alpaca

Code and documentation to train Stanford's Alpaca models, and generate the data.

LLMs-from-scratch

Implementing a ChatGPT-like LLM in PyTorch from scratch, step by step

marker

Convert PDF to markdown quickly with high accuracy

ivy

Convert Machine Learning Code Between Frameworks

language-server-protocol

Defines a common protocol for language servers.

nougat

Implementation of Nougat Neural Optical Understanding for Academic Documents

open_llama

OpenLLaMA, a permissively licensed open source reproduction of Meta AI’s LLaMA 7B trained on the RedPajama dataset

x-transformers

A concise but complete full-attention transformer with a set of promising experimental features from various papers

rift

Rift: an AI-native language server for your personal AI software engineer

direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

mup

maximal update parametrization (µP)

evalplus

Rigourous evaluation of LLM-synthesized code - NeurIPS 2023

alpaca_farm

A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

LeanDojo

Tool for data extraction and interacting with Lean programmatically.

parsel

Code for Parsel 🐍 - generate complex programs with language models

SciLean

Scientific computing in Lean 4

yatima

A zero-knowledge Lean4 compiler and kernel

Portal-to-ISAbelle

https://albertqjiang.github.io/Portal-to-ISAbelle/

tacheles

a lightweight, open-source blueprint for building powerful and scalable LLM chat applications

FIMO

lm-evaluation-harness

A framework for few-shot evaluation of autoregressive language models.

ultimate-anatome

Ἀνατομή is a PyTorch library to analyze representation of neural networks

ultimate-utils

Brando's utils

beyond-scale-language-data-diversity

UnicodeSkipListTable

A library to create and use Unicode tables based on the skip list data structure.

UnicodeSkipListTableExample

A library that shows how to use the Unicode skip list tables generation tool to create a table to test if a codepoint is numeric.

Why-Has-Predicting-Downstream-Capabilities-Remained-Elusive

Code for Preprint: Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?

learning_lean