Update bert.md (#1054)

chapter_natural-language-processing-pretraining/bert.md

@@ -203,7 +203,7 @@ class MaskLM(nn.Block):
         batch_size = X.shape[0]
         batch_idx = np.arange(0, batch_size)
         # 假设batch_size=2，num_pred_positions=3
-        # 那么batch_idx是np.array（[0,0,0,1,1]）
+        # 那么batch_idx是np.array（[0,0,0,1,1,1]）
         batch_idx = np.repeat(batch_idx, num_pred_positions)
         masked_X = X[batch_idx, pred_positions]
         masked_X = masked_X.reshape((batch_size, num_pred_positions, -1))