The Predictions view has two boxes that show how close the predicted value for each example output is to the true value. The training and validating examples are shown in separate boxes. The position of the values on both axis are scaled to be from a minimum of 0 to a maximum of 1. Predicted outputs for training examples get closer to the true values as training progresses. If the predicted values are very close to the true values then the dots will be on the diagonal line. If most of the dots are close to the diagonal line the network may be over-trained and it may not generalize well.