NanoGPT

NanoGPT is a small-scale attempt to build a GPT-like language model from scratch. This project explores the fundamentals of transformer-based architectures and language modeling using a tiny dataset.

Overview

This project implements a minimalistic version of a GPT-style model. It is trained using a tiny Shakespeare dataset and aims to generate text in a similar style. The training process follows standard deep learning techniques for language modeling, leveraging a simplified transformer architecture.

Features

• Small-scale GPT-style language model
• Trained on a tiny Shakespeare dataset
• Minimal implementation for educational purposes
• Outputs sample generated text in out.txt

Files

• nanogpt.py - The script implementation of the NanoGPT model
• nanogpt.py - The notebook implementation of the NanoGPT model
• out.txt - A sample output of the trained model
• screenshot.png - Screenshot of the output
• README.md - Project documentation

Output Screenshot