diff --git a/5_gemma_2b_axolotl/gemma_2b_axolotl.ipynb b/5_gemma_2b_axolotl/gemma_2b_axolotl.ipynb
new file mode 100644
index 0000000..6c74792
--- /dev/null
+++ b/5_gemma_2b_axolotl/gemma_2b_axolotl.ipynb
@@ -0,0 +1,22 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Getting Started with Axolotl: Fine-Tuning Gemma 2B\n",
+    "\n",
+    "Axolotl is \"a tool designed to streamline the fine-tuning of various AI models.\" It is primarily for training Hugging Face models via full fine-tuning, lora, qlora, relora, gptq. Configurations are specified in yaml files. It supports a variety of different dataset formats. It supports additional libraries such as xformer and flash attention. It is compatible with FSDP and deepspeed for multi-gpu training. It supports logging to MLflow or WandB.\n",
+    "\n",
+    "# The recommended workflow is to pick a quickstart notebook from the [examples](https://github.com/OpenAccess-AI-Collective/axolotl/tree/main/examples) directory and modify it as needed.\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}