nanoGPT Shakespeare character model

A small character-level GPT trained on Shakespeare using Andrej Karpathy's nanoGPT implementation.

Architecture

This is a nanoGPT decoder-only Transformer trained from scratch on the Shakespeare character-level dataset.

The exact architecture parameters are stored in ckpt.pt under model_args.

Files

  • ckpt.pt - trained model checkpoint
  • model.py - nanoGPT model architecture
  • meta.pkl - character vocabulary
  • run.py - minimal inference example
  • train_shakespeare_char.py - training configuration

Run locally

pip install torch numpy

python run.py

Training

Based on:

https://github.com/karpathy/nanoGPT

The model was trained using the Shakespeare character-level dataset.

Notes

This repository contains a native nanoGPT checkpoint rather than a Hugging Face Transformers PreTrainedModel.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support