In the first colab notebook why we use the first layer of each section?

# style layers of interest
style_layers = ['block1_conv1', 

# choose the content layer and put in a list
content_layers = ['block5_conv2'] 

# combine the two lists (put the style layers before the content layers)
output_layers = style_layers + content_layers 

# declare auxiliary variables holding the number of style and content layers
NUM_CONTENT_LAYERS = len(content_layers)
NUM_STYLE_LAYERS = len(style_layers)

Because these layers have better representation for the style of the image, than compared to layers which are more in depth of the block.