diff --git a/chapters/en/chapter12/5.mdx b/chapters/en/chapter12/5.mdx
index 8c42ad793..e27368ca0 100644
--- a/chapters/en/chapter12/5.mdx
+++ b/chapters/en/chapter12/5.mdx
@@ -9,7 +9,7 @@
 Now that you've seen the theory, let's put it into practice! In this exercise, you'll fine-tune a model with GRPO.
 
 <Tip>
-This exercise was written by LLM fine-tuning expert [@mlabonne](https://huggingface.co/mlabonne).
+This exercise was written by LLM fine-tuning expert <a href="https://huggingface.co/mlabonne">@mlabonne</a>.
 </Tip>
 
 ## Install dependencies
@@ -56,7 +56,7 @@ print(dataset)
 
 Now, let's load the model.
 
-For this exercise, we'll use the [`SmolLM2-135M`](hhttps://huggingface.co/HuggingFaceTB/SmolLM2-135M) model. 
+For this exercise, we'll use the [`SmolLM2-135M`](https://huggingface.co/HuggingFaceTB/SmolLM2-135M) model. 
 
 This is a small 135M parameter model that runs on limited hardware. This makes the model ideal for learning, but it's not the most powerful model out there. If you have access to more powerful hardware, you can try to fine-tune a larger model like [`SmolLM2-1.7B`](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B).