scottlogic-alex Goto Github PK

accelerate

🚀 A simple way to train and use PyTorch models with multi-GPU, TPU, mixed-precision

axolotl

Go ahead and axolotl questions

blog-2

The Scott Logic blog

chart-play

scripts to help with exporting charts

create-react-app

Create React apps with no build configuration.

dagre-d3

A D3-based renderer for Dagre

diffusers-play

Repository with which to explore k-diffusion and diffusers, and within which changes to said packages may be tested.

graal-play

llm-amp-f8

Train an LLM in fp8 with MS-AMP

llm-memory-calculator

Compute theoretical memory requirements for finetuning LLM

lm-evaluation-harness

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

megatron-deepspeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2

mini-grad

Learn basics of Neural Net backpropagation by implementing a wee tensor library

ms-amp-fp8

Let's try and run MS-AMP training in fp8 and measure the differences

panda-play

prm800k-denorm

Script for processing OpenAI's PRM800K process supervision dataset into an Alpaca-style instruction-response format

qlora

Efficient methods for LLMs finetuning and inference.

react-piano

reactive-game-angular

:tada: An Angular Starter kit featuring Angular (Router, Http, Forms, Services, Tests, E2E, Dev/Prod, HMR, Async/Lazy Routes, AoT via ngc), Karma, Protractor, Jasmine, Istanbul, TypeScript, TsLint, Codelyzer, Hot Module Replacement, @types, and Webpack

reactivegame

tns-template-blank-react

Starter boilerplate for a React NativeScript app