Fix various [friction points](https://docs.google.com/document/d/1nQeyjtNKig-TmbL2mImYfGaOIJQyFbNoISPdxalQm44/edit?usp=sharing&resourcekey=0-LleDJBkn_pWvC4pQaDz9Yg) in the TF beginner quickstart.

PiperOrigin-RevId: 499589900

Author: tensorflower-gardener

site/en/tutorials/quickstart/beginner.ipynb

@@ -125,7 +125,7 @@
         "\n",
         "## Load a dataset\n",
         "\n",
-        "Load and prepare the [MNIST dataset](http://yann.lecun.com/exdb/mnist/). Convert the sample data from integers to floating-point numbers:"
+        "Load and prepare the [MNIST dataset](http://yann.lecun.com/exdb/mnist/). The pixel values of the images range from 0 through 255. Scale these values to a range of 0 to 1 by dividing the values by `255.0`. This also converts the sample data from integers to floating-point numbers:"
       ]
     },
     {
@@ -150,7 +150,7 @@
       "source": [
         "## Build a machine learning model\n",
         "\n",
-        "Build a `tf.keras.Sequential` model by stacking layers."
+        "Build a `tf.keras.Sequential` model:"
       ]
     },
     {
@@ -175,6 +175,8 @@
         "id": "l2hiez2eIUz8"
       },
       "source": [
+        "[`Sequential`](https://www.tensorflow.org/guide/keras/sequential_model) is useful for stacking layers where each layer has one input [tensor](https://www.tensorflow.org/guide/tensor) and one output tensor. Layers are functions with a known mathematical structure that can be reused and have trainable variables. Most TensorFlow models are composed of layers. This model uses the [`Flatten`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Flatten), [`Dense`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense), and [`Dropout`](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dropout) layers.\n",
+        "\n",
         "For each example, the model returns a vector of [logits](https://developers.google.com/machine-learning/glossary#logits) or [log-odds](https://developers.google.com/machine-learning/glossary#log-odds) scores, one for each class."
       ]
     },
@@ -225,7 +227,7 @@
         "id": "hQyugpgRIyrA"
       },
       "source": [
-        "Define a loss function for training using `losses.SparseCategoricalCrossentropy`, which takes a vector of logits and a `True` index and returns a scalar loss for each example."
+        "Define a loss function for training using `losses.SparseCategoricalCrossentropy`:"
       ]
     },
     {
@@ -245,7 +247,7 @@
         "id": "SfR4MsSDU880"
       },
       "source": [
-        "This loss is equal to the negative log probability of the true class: The loss is zero if the model is sure of the correct class.\n",
+        "The loss function takes a vector of ground truth values and a vector of logits and returns a scalar loss for each example. This loss is equal to the negative log probability of the true class: The loss is zero if the model is sure of the correct class.\n",
         "\n",
         "This untrained model gives probabilities close to random (1/10 for each class), so the initial loss should be close to `-tf.math.log(1/10) ~= 2.3`."
       ]