diff --git a/README.md b/README.md
index 512e46d..8c78d6e 100644
--- a/README.md
+++ b/README.md
@@ -64,7 +64,7 @@ To see the TensorBoard visualization of the training progress:
 
 Optionally edit the `hparams.yaml` configuration file in the folder `egs/covers80/config` before starting a training run.
 
-This fork added an hparam.yaml setting of `early_stopping_patience` to support the added feature of early stopping (original CoverHunter defaulted to 10,000 epochs!).
+This fork added the hyperparameter `early_stopping_patience` to support the added feature of early stopping (original CoverHunter defaulted to 10,000 epochs!).
 
 Note: Don't use the `torchrun` launch command offered in original CoverHunter. In the single-computer Apple Silicon context, it is not only irrelevant, it actually slows down performance. In my tests it slowed down tools.train performance by about 20%.
 
@@ -90,7 +90,6 @@ This figure shows the results of training from scratch on the covers80 dataset w
 ![t-SNE plot for Covers80](tSNE-example.png)
 
 The optional `dist_name` argument is a path where you want to save the distance matrix and ref labels so that you can study the results separately, such as perhaps doing custom t-SNE plots, etc.
-
  
 See the "Training checkpoint output" section below for a description of the embeddings saved by the `eval_for_map_with_feat()` function called in this script. They are saved in a new subfolder of the `pretrained_model` folder named `embed_NN_tmp` where NN is the highest-numbered epoch subfolder in the `pretrained_model` folder.
 
diff --git a/src/eval_testset.py b/src/eval_testset.py
index a5a4196..9f1fffd 100644
--- a/src/eval_testset.py
+++ b/src/eval_testset.py
@@ -40,7 +40,6 @@ def _cluster_plot(dist_matrix, ref_labels, output_path, test_only_labels=[],logg
     marker_styles = ['o', 's', '^', 'p', 'x', 'D']
     num_colors = len(unique_labels) // len(marker_styles)
     colors = plt.get_cmap(cmap_name, num_colors)(range(num_colors))
-
     plt.figure(figsize=(15, 15))
 
     color_dict = {}  # Dictionary to store color for each label
@@ -84,6 +83,7 @@ def _cluster_plot(dist_matrix, ref_labels, output_path, test_only_labels=[],logg
     if test_only_labels:
         plt.text(1, 1.02, "Circles = song_ids not seen in training", ha='right', va='bottom', transform=plt.gca().transAxes)
 
+    plt.tight_layout()
     plt.savefig(output_path)
     plt.close()
 
diff --git a/tSNE.png b/tSNE.png
new file mode 100644
index 0000000..736b3dd
Binary files /dev/null and b/tSNE.png differ
diff --git a/tools/eval_testset.py b/tools/eval_testset.py
index 96bf618..49dc814 100644
--- a/tools/eval_testset.py
+++ b/tools/eval_testset.py
@@ -23,8 +23,8 @@ def _main():
   parser.add_argument('query_path')
   parser.add_argument('ref_path')
   parser.add_argument('-query_in_ref_path', default='', type=str)
-  parser.add_argument('-test_only_labels', default='', type=str, help='Path to list of song_ids reserved for test dataset for use in t-SNE plot.')
   parser.add_argument('-plot_name', default='', type=str, help='Save a t-SNE plot of the distance matrix to this path')
+  parser.add_argument('-test_only_labels', default='', type=str, help='Path to list of song_ids reserved for test dataset for use in t-SNE plot.')
   parser.add_argument('-dist_name', default='', type=str, help='Save the distance matrix to this path')
 
   args = parser.parse_args()