Type a line. Watch it become verse.
Generated by a 2.7M-parameter transformer, trained from scratch on public-domain poetry, Tagore, Naidu, Dickinson, Poe, and more. It runs on a small server now, so there's nothing to download; just start typing.
What's actually happening
Runs on a small server
The model runs server-side, a few million parameters, milliseconds per request. No download, no setup, always ready.
Architecture
Embeddings, attention, softmax, sampling. See my article and code using the link below. The model architecture follows Andrej Karpathy's "Let's Build GPT", a well established way to build a transformer.
Public-domain training data
The poetry it learned from, Tagore, Naidu, Kabir, Dickinson, Poe, and more, comes from Project Gutenberg. Every book is in the public domain, royalty free, and free of copyright.
LyricsLM: Build and Train a GPT-Style Language Model From Scratch
Explained so clearly that even a high school student can build a new Language Model from scratch while understanding the core concepts behind how it works. We will use the same core methodology behind frontier models, at micro scale.
