diff --git a/grants_tagger_light/evaluation/evaluate_model.py b/grants_tagger_light/evaluation/evaluate_model.py
index 93356aa8..edfd749b 100644
--- a/grants_tagger_light/evaluation/evaluate_model.py
+++ b/grants_tagger_light/evaluation/evaluate_model.py
@@ -37,7 +37,7 @@ def evaluate_model(
 ):
     model = BertMesh.from_pretrained(model_path)
 
-    label_binarizer = MultiLabelBinarizer()
+    label_binarizer = MultiLabelBinarizer(classes=list(model.id2label.keys()))
     label_binarizer.fit([list(model.id2label.keys())])
     model.label2id = {value: key for key, value in model.id2label.items()}
 
@@ -57,8 +57,8 @@ def evaluate_model(
         X_test, Y_test, _ = load_data(data_path, label_binarizer, model_label2id=model.label2id)
 
     logging.info('data loaded')
-    X_test = X_test[:100]
-    Y_test = Y_test[:100]
+    X_test = X_test[:10]
+    Y_test = Y_test[:10]
 
     top_10_index = np.argsort(np.sum(Y_test, axis=0))[::-1][:10]
     print(top_10_index)
@@ -74,9 +74,8 @@ def evaluate_model(
         print(f'argmax: {argmax}')
         print(argmax, output[0][argmax])
     Y_pred_proba = pipe(X_test, return_labels=False)
-    #Y_pred_proba = [torch.sigmoid(proba) for proba in Y_pred_proba]
     
-    #Y_pred_proba = [torch.sigmoid(proba) for proba in Y_pred_proba]
+    Y_pred_proba = [torch.sigmoid(proba) for proba in Y_pred_proba]
     print(Y_pred_proba)
     Y_pred_proba = torch.vstack(Y_pred_proba)
     print('loss')