diff --git a/README.md b/README.md
index 4f819fa61..d3f44b880 100644
--- a/README.md
+++ b/README.md
@@ -98,14 +98,14 @@ to another project.
   </tr>
   <tr>
     <td><a href="./date-conversion-attention">date-conversion-attention</a></td>
-    <td></td>
-    <td></td>
-    <td>Attention RNN for text-to-text conversion</td>
-    <td></td>
-    <td></td>
-    <td></td>
-    <td></td>
-    <td></td>
+    <td><a href="https://storage.googleapis.com/tfjs-examples/date-conversion-attention/dist/index.html">🔗</a></td>
+    <td>Text</td>
+    <td>Text-to-text conversion</td>
+    <td>Attention mechanism, RNN</td>
+    <td>Node</td>
+    <td>Browser and Node</td>
+    <td>Layers</td>
+    <td>Saving to filesystem and loading in browser</td>
   </tr>
   <tr>
     <td><a href="./iris">iris</a></td>
@@ -144,7 +144,7 @@ to another project.
     <td><a href="./lstm-text-generation">lstm-text-generation</a></td>
     <td><a href="https://storage.googleapis.com/tfjs-examples/lstm-text-generation/dist/index.html">🔗</a></td>
     <td>Text</td>
-    <td>Sequence-to-prediction</td>
+    <td>Sequence prediction</td>
     <td>RNN: LSTM</td>
     <td>Browser</td>
     <td>Browser</td>
diff --git a/sentiment/data.js b/sentiment/data.js
index c9a8352b3..6db1caaae 100644
--- a/sentiment/data.js
+++ b/sentiment/data.js
@@ -21,7 +21,7 @@ import * as https from 'https';
 import * as os from 'os';
 import * as path from 'path';
 
-import {OOV_CHAR, PAD_CHAR, padSequences} from './sequence_utils';
+import {OOV_INDEX, PAD_INDEX, padSequences} from './sequence_utils';
 
 // `import` doesn't seem to work with extract-zip.
 const extract = require('extract-zip');
@@ -36,7 +36,7 @@ const METADATA_TEMPLATE_URL =
  *
  * @param {string} filePath Data file on local filesystem.
  * @param {string} numWords Number of words in the vocabulary. Word indices
- *   that exceed this limit will be marked as `OOV_CHAR`.
+ *   that exceed this limit will be marked as `OOV_INDEX`.
  * @param {string} maxLen Length of each sequence. Longer sequences will be
  *   pre-truncated; shorter ones will be pre-padded.
  * @return {tf.Tensor} The dataset represented as a 2D `tf.Tensor` of shape
@@ -60,7 +60,7 @@ function loadFeatures(filePath, numWords, maxLen) {
       seq = [];
     } else {
       // Sequence continues.
-      seq.push(value >= numWords ? OOV_CHAR : value);
+      seq.push(value >= numWords ? OOV_INDEX : value);
     }
     index += 4;
   }
diff --git a/sentiment/index.js b/sentiment/index.js
index 430fdd70f..05a6559e1 100644
--- a/sentiment/index.js
+++ b/sentiment/index.js
@@ -18,7 +18,7 @@
 import * as tf from '@tensorflow/tfjs';
 import * as loader from './loader';
 import * as ui from './ui';
-import {OOV_CHAR, padSequences} from './sequence_utils';
+import {OOV_INDEX, padSequences} from './sequence_utils';
 
 const HOSTED_URLS = {
   model:
@@ -65,7 +65,7 @@ class SentimentPredictor {
     const sequence = inputText.map(word => {
       let wordIndex = this.wordIndex[word] + this.indexFrom;
       if (wordIndex > this.vocabularySize) {
-        wordIndex = OOV_CHAR;
+        wordIndex = OOV_INDEX;
       }
       return wordIndex;
     });
diff --git a/sentiment/sequence_utils.js b/sentiment/sequence_utils.js
index 54d6169a0..8ac41d446 100644
--- a/sentiment/sequence_utils.js
+++ b/sentiment/sequence_utils.js
@@ -19,8 +19,8 @@
  * Utilities for sequential data.
  */
 
-export const PAD_CHAR = 0;
-export const OOV_CHAR = 2;
+export const PAD_INDEX = 0;  // Index of the padding character.
+export const OOV_INDEX = 2;  // Index fo the OOV character.
 
 /**
  * Pad and truncate all sequences to the same length
@@ -34,7 +34,7 @@ export const OOV_CHAR = 2;
  * @param {number} value Padding value.
  */
 export function padSequences(
-    sequences, maxLen, padding = 'pre', truncating = 'pre', value = PAD_CHAR) {
+    sequences, maxLen, padding = 'pre', truncating = 'pre', value = PAD_INDEX) {
   // TODO(cais): This perhaps should be refined and moved into tfjs-preproc.
   return sequences.map(seq => {
     // Perform truncation.