diff --git a/fig/distreps-as-features.png b/fig/distreps-as-features.png
new file mode 100644
index 0000000..1f2764b
Binary files /dev/null and b/fig/distreps-as-features.png differ
diff --git a/sst_03_neural_networks.ipynb b/sst_03_neural_networks.ipynb
index 6cedbf5..9183d65 100644
--- a/sst_03_neural_networks.ipynb
+++ b/sst_03_neural_networks.ipynb
@@ -126,6 +126,8 @@
     "\n",
     "As a first step in the direction of neural networks for sentiment, we can connect with our previous unit on distributed representations. Arguably, more than any specific model architecture, this is the major innovation of deep learning: __rather than designing feature functions by hand, we use dense, distributed representations, often derived from unsupervised models__.\n",
     "\n",
+    "<img src=\"fig/distreps-as-features.png\" width=500 alt=\"distreps-as-features.png\" />\n",
+    "\n",
     "Our model will just be `LogisticRegression`, and we'll continue with the experiment framework from the previous notebook. Here is `fit_maxent_classifier` again:"
    ]
   },
@@ -471,7 +473,7 @@
     "\n",
     "* At the same time, we might want to collapse infrequent tokens into `$UNK` to make optimization easier.\n",
     "\n",
-    "In `sst`, the function `get_vocab` implements these strategies. Now we can extract the training vocab and use it for the model embedding, secure in the knowledge that we will be able to process tokens outside of this set (by mapping them to `$UNK`)."
+    "In `utils`, the function `get_vocab` implements these strategies. Now we can extract the training vocab and use it for the model embedding, secure in the knowledge that we will be able to process tokens outside of this set (by mapping them to `$UNK`)."
    ]
   },
   {