diff --git a/example/README.md b/example/README.md
index 2123104a1487..dea7e289e6cd 100644
--- a/example/README.md
+++ b/example/README.md
@@ -95,7 +95,7 @@ If your tutorial depends on specific packages, simply add them to this provision
 * [Gluon Examples](gluon) - several examples using the Gluon API
   * [Style Transfer](gluon/style_transfer) - a style transfer example using gluon
   * [Word Language Model](gluon/word_language_model) - an example that trains a multi-layer RNN on the Penn Treebank language modeling benchmark
-  * [SN-GAN](gluon/sn-gan) - an example that utilizes spectral normalization to train GAN(Generative adversarial network) using Gluon API
+  * [SN-GAN](gluon/sn_gan) - an example that utilizes spectral normalization to train GAN(Generative adversarial network) using Gluon API
 * [Image Classification with R](image-classification) - image classification on MNIST,CIFAR,ImageNet-1k,ImageNet-Full, with multiple GPU and distributed training.
 * [Kaggle 1st national data science bowl](kaggle-ndsb1) - a MXnet example for Kaggle Nation Data Science Bowl 1
 * [Kaggle 2nd national data science bowl](kaggle-ndsb2) - a tutorial for Kaggle Second Nation Data Science Bowl
diff --git a/example/gluon/sn_gan/train.py b/example/gluon/sn_gan/train.py
index 5faf3a2a02a8..46e44791cebd 100644
--- a/example/gluon/sn_gan/train.py
+++ b/example/gluon/sn_gan/train.py
@@ -50,7 +50,7 @@
                     help='use gpu for training.')
 parser.add_argument('--clip_gr', type=float, default=10.0,
                     help='Clip the gradient by projecting onto the box. default is 10.0.')
-parser.add_argument('--z-dim', type=int, default=10,
+parser.add_argument('--z-dim', type=int, default=100,
                     help='dimension of the latent z vector. default is 100.')
 opt = parser.parse_args()