diff --git a/notebooks/feature_engineering/solutions/4_keras_adv_feat_eng.ipynb b/notebooks/feature_engineering/solutions/4_keras_adv_feat_eng.ipynb
index 4c86e9b5..170eb160 100644
--- a/notebooks/feature_engineering/solutions/4_keras_adv_feat_eng.ipynb
+++ b/notebooks/feature_engineering/solutions/4_keras_adv_feat_eng.ipynb
@@ -163,7 +163,7 @@
    "source": [
     "## Create a Baseline DNN Model in Keras\n",
     "\n",
-    "Now let's build the Deep Neural Network (DNN) model in Keras using the functional API. Unlike the sequential API, we will need to specify the input and hidden layers.  Note that we are creating a linear regression baseline model with no feature engineering. Recall that a baseline model is a solution to a problem without applying any machine learning techniques."
+    "Now let's build the Deep Neural Network (DNN) model in Keras using the functional API. Unlike the sequential API, we will need to specify the input and hidden layers.  Note that we are creating a linear regression baseline model with no feature engineering. A baseline model is a simple, basic model that acts as a reference point for evaluating the performance of more complex models."
    ]
   },
   {
@@ -649,15 +649,15 @@
  ],
  "metadata": {
   "environment": {
-   "kernel": "conda-base-py",
-   "name": "workbench-notebooks.m121",
+   "kernel": "python3",
+   "name": "tf2-gpu.2-17.m126",
    "type": "gcloud",
-   "uri": "us-docker.pkg.dev/deeplearning-platform-release/gcr.io/workbench-notebooks:m121"
+   "uri": "us-docker.pkg.dev/deeplearning-platform-release/gcr.io/tf2-gpu.2-17:m126"
   },
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel) (Local)",
+   "display_name": "Python 3 (Local)",
    "language": "python",
-   "name": "conda-base-py"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -669,7 +669,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.14"
+   "version": "3.10.15"
   }
  },
  "nbformat": 4,