From 709d26572eb0fc29ffee15c7c1618385a1860312 Mon Sep 17 00:00:00 2001 From: Felix Andreas Date: Fri, 17 May 2024 18:10:43 +0200 Subject: [PATCH] docs: Update readme --- README.md | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/README.md b/README.md index 584c0c8..f4c7afd 100644 --- a/README.md +++ b/README.md @@ -6,6 +6,10 @@ +This project aims to be a clean and concise re-implementation of [GPT-2](https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf). The model implementation, contained in [`src/model.rs`](src/model.rs), is under 300 lines of code. While this was a fun exercise mostly for (my own) educational purposes, it demonstrates the utility of Rust and Burn in the machine learning domain: The entire project compiles into a single binary, making deployment relatively straightforward. + +The project also includes a simple CLI for training and inference. At the moment, only a character-level tokenizer is supported, so official weights requiring a BPE tokenizer cannot be used yet. However, for fun, you can try out the small toy model I trained ([see inference](#inference)). + ## Installation You can install `gpt-burn` with [Nix](https://nixos.org/): @@ -14,6 +18,12 @@ You can install `gpt-burn` with [Nix](https://nixos.org/): nix run github:felix-andreas/gpt-burn ``` +Or, install with `cargo`: + +``` +cargo install --git https://github.com/felix-andreas/gpt-burn +``` + Alternatively, clone the repo and build from source: ```sh