1	‘input'	Image Input	227x227x3 images with ‘zerocenter’ normalization	Input layer
2	‘conv1′	Convolution	96 11x11x3 convolutions with stride [44] and padding [0 0 0 0]	The 1st layer
3	‘relu1′	ReLU	ReLU
4	‘norm1′	Cross Channel Normalization	cross channel normalization with 5 channels per element
5	‘pool1′	Max Pooling	3x3 max pooling with stride [22] and padding [0 0 0 0]
6	‘conv'	Convolution	256 5x5x48 convolutions with stride [11] and padding [2 2 2 2]	The 2nd layer
7	‘relu2′	ReLU	ReLU
8	‘norm2′	Cross Channel Normalization	cross channel normalization with 5 channels per element
9	‘pool2′	Max Pooling	3x3 max pooling with stride [22] and padding [0 0 0 0]
10	‘conv3′	Convolution	384 3x3x256 convolutions with stride [11] and padding [1 1 1 1]	The 3rd layer
11	‘relu3′	ReLU	ReLU
12	‘conv4′	Convolution	384 3x3x192 convolutions with stride [11] and padding [1 1 1 1]	The 4th layer
13	‘relu4′	ReLU	ReLU
14	‘conv5′	Convolution	256 3x3x192 convolutions with stride [11] and padding [1 1 1 1]	The 5th layer
15	‘relu5′	ReLU	ReLU
16	‘pool5′	Max Pooling	3x3 max pooling with stride [22] and padding [0 0 0 0]
17	‘conv6′	Convolution	512 3x3 convolutions with stride [11] and padding [1 1 1 1]	The 6th layer
18	‘relu6′	ReLU	ReLU
19	‘conv7′	Convolution	512 3x3 convolutions with stride [11] and padding [1 1 1 1]	The 7th layer
20	‘relu7′	ReLU	ReLU