diff --git a/examples/token_classification.ipynb b/examples/token_classification.ipynb
index 3fb038c3..d6ca8b64 100644
--- a/examples/token_classification.ipynb
+++ b/examples/token_classification.ipynb
@@ -22,7 +22,7 @@
    },
    "outputs": [],
    "source": [
-    "#! pip install datasets transformers seqeval"
+    "#! pip install datasets transformers seqeval evaluate"
    ]
   },
   {
@@ -171,7 +171,7 @@
     "id": "W7QYTpxXIrIl"
    },
    "source": [
-    "We will use the [🤗 Datasets](https://github.com/huggingface/datasets) library to download the data and get the metric we need to use for evaluation (to compare our model to the benchmark). This can be easily done with the functions `load_dataset` and `load_metric`.  "
+    "We will use the [🤗 Datasets](https://github.com/huggingface/datasets) library to download the data. This can be easily done with the `load_dataset` function. To get the metric we need to use for evaluation (to compare our model to the benchmark), we will use the [🤗 Evaluate](https://github.com/huggingface/evaluate) library."
    ]
   },
   {
@@ -182,7 +182,8 @@
    },
    "outputs": [],
    "source": [
-    "from datasets import load_dataset, load_metric"
+    "from datasets import load_dataset\n",
+    "import evaluate"
    ]
   },
   {
@@ -1096,7 +1097,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "The last thing to define for our `Trainer` is how to compute the metrics from the predictions. Here we will load the [`seqeval`](https://github.com/chakki-works/seqeval) metric (which is commonly used to evaluate results on the CONLL dataset) via the Datasets library."
+    "The last thing to define for our `Trainer` is how to compute the metrics from the predictions. Here we will load the [`seqeval`](https://github.com/chakki-works/seqeval) metric (which is commonly used to evaluate results on the CONLL dataset) via the Evaluate library."
    ]
   },
   {
@@ -1105,7 +1106,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "metric = load_metric(\"seqeval\")"
+    "metric = evaluate.load(\"seqeval\")"
    ]
   },
   {