jaymody/picoGPT

An unnecessarily tiny implementation of GPT-2 in NumPy.

48
/ 100
Emerging

This project offers a bare-bones version of the GPT-2 language model, allowing developers to see its core mechanics with minimal code. You input a text prompt, and it generates a text completion. This is designed for software developers or machine learning engineers who want to understand the foundational structure of a transformer-based language model without the complexities of a full-scale implementation.

3,454 stars. No commits in the last 6 months.

Use this if you are a developer looking for an extremely simplified, readable implementation of GPT-2's forward pass to learn its architecture.

Not ideal if you need to train a model, perform batch inference, use advanced sampling techniques, or require fast text generation for real-world applications.

Machine Learning Development Natural Language Processing Model Architecture Educational Resource Deep Learning
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

3,454

Forks

456

Language

Python

License

MIT

Last pushed

Apr 24, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/jaymody/picoGPT"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.