From 8ea5c1e8484164e0081f47eb9327b2c16452329c Mon Sep 17 00:00:00 2001
From: murrellb <murrellb@gmail.com>
Date: Thu, 14 Nov 2024 14:44:22 +0100
Subject: [PATCH] Install instructions

---
 README.md | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/README.md b/README.md
index bdfdd3d..aa0dd29 100644
--- a/README.md
+++ b/README.md
@@ -5,6 +5,12 @@
 [![Build Status](https://github.com/MurrellGroup/Jjama3.jl/actions/workflows/CI.yml/badge.svg?branch=main)](https://github.com/MurrellGroup/Jjama3.jl/actions/workflows/CI.yml?query=branch%3Amain)
 [![Coverage](https://codecov.io/gh/MurrellGroup/Jjama3.jl/branch/main/graph/badge.svg)](https://codecov.io/gh/MurrellGroup/Jjama3.jl)
 
+# Installation
+
+```julia
+] add https://github.com/MurrellGroup/Jjama3.jl
+```
+
 # Quickstart
 
 Download a Llama3 model `config.json` and safetensor weights from Huggingface. Eg. [Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct). You might need access permissions for this. Note: Huggingface use a different RoPE convention to the original Meta implementation, and their weights have been permuted. This package works with the Huggingface convention, so if you load from the original weights you'll need to permute them.