dabochen/spreadsheet-is-all-you-need
A spreadsheet implementation of the full nanoGPT inference pipeline for visualizing transformer architecture.
Not currently ranked — collecting fresh signals.
star history
This project implements Andrej Karpathy’s nanoGPT architecture entirely within a spreadsheet, replacing Python with Excel/Google Sheets formulas. It encodes all transformer components—embedding, layer normalization, self-attention (KQV), projection, MLP, softmax, and logits—as interactive matrix calculations. The project serves as a visual learning tool, allowing users to manipulate and trace data flow through the entire character-based prediction system, which uses roughly 85,000 parameters configured for characters A/B/C.
Frequently asked
- What is dabochen/spreadsheet-is-all-you-need?
- A spreadsheet implementation of the full nanoGPT inference pipeline for visualizing transformer architecture.
- Is spreadsheet-is-all-you-need open source?
- Yes — dabochen/spreadsheet-is-all-you-need is open source, released under the MIT license.
- How popular is spreadsheet-is-all-you-need?
- dabochen/spreadsheet-is-all-you-need has 2.2k stars on GitHub.
- Where can I find spreadsheet-is-all-you-need?
- dabochen/spreadsheet-is-all-you-need is on GitHub at https://github.com/dabochen/spreadsheet-is-all-you-need.