Fix PR comments

tutorials/notebooks/example_keras_yolov8n.ipynb

@@ -133,9 +133,10 @@
    "cell_type": "markdown",
    "source": [
     "### Generate Yolo8-nano Keras model\n",
-    "In the following steps, we proceed to create a Keras model based on [Ultralytics](https://github.com/ultralytics/ultralytics). We then initialize the model weights to match those of the original model.\n",
+    "In the following steps, we proceed to create a Keras model based on [Ultralytics](https://github.com/ultralytics/ultralytics). This implementation includes a slightly modified version of yolov8\n",
+    "detection-head (mainly the box decoding part) that was optimized for model quantization. We then initialize the model weights to match those of the original model.\n",
     "\n",
-    "To this base model, we integrate the post-processing components, which include box decoding layers following by tensorflow [tf.image.combined_non_max_suppression](https://www.tensorflow.org/api_docs/python/tf/image/combined_non_max_suppression) layer.\n",
+    "To this base model, we integrate the post-processing components, which include tensorflow [tf.image.combined_non_max_suppression](https://www.tensorflow.org/api_docs/python/tf/image/combined_non_max_suppression) layer.\n",
     "For further insights into the model's implementation details, please refer to the tutorials [\"resources\" folder](https://github.com/sony/model_optimization/tree/main/tutorials/resources).   "
    ],
    "metadata": {