A playful project that cleverly transforms everyday sentences into their mischievous "toxic" counterparts. 😈

Finetune LLM using Torchtune on Windows

This repository contains a comprehensive pipeline for fine-tuning Large Language Models (LLMs) to optimize their performance for specialized healthcare applications. The project focuses on transforming a base Llama model into a specialized medical assistant capable of classifying patient intent and answering clinical queries.

This repository provides resources for fine-tuning various types of models using different techniques and frameworks.

Internet-scale data-parallel fine-tuning of Qwen2-0.5B-Instruct using Hivemind + TorchTune. Initial peer on public IP; second peers on free GPUs (e.g., Kaggle).