diff --git a/README.md b/README.md
index 78abd56..a1403e0 100644
--- a/README.md
+++ b/README.md
@@ -103,7 +103,7 @@ As a baseline to compare `xlm-roberta-base-language-detection` against, we have
 
 ## Training procedure
 
-Fine-tuning was done via the `Trainer` API.
+Fine-tuning was done via the `Trainer` API. Here is the [Colab notebook](https://colab.research.google.com/drive/15LJTckS6gU3RQOmjLqxVNBmbsBdnUEvl?usp=sharing) with the training code.
 
 ### Training hyperparameters