diff --git a/demonstrations/tutorial_liesim.py b/demonstrations/tutorial_liesim.py
index 0cb3ffabb2..6eebb7f1d0 100644
--- a/demonstrations/tutorial_liesim.py
+++ b/demonstrations/tutorial_liesim.py
@@ -66,7 +66,7 @@
     Technically, the (dynamical) Lie algebra is formed by skew-Hermitian operators :math:`\{i h_i\}`.
     We avoid this distinction here since for all practical purposes one can also look at Hermitian
     operators and explicitly add imaginary units in the exponents where appropriate.
-    For more details, see the note in the "Lie algebras" section of our :doc:`Intro to (Dynamical) Lie Algebras for quantum practitioners </demos/tutorial_liealgebra//#lie-algebras>`.
+    For more details, see the note in the "Lie algebras" section of our `Intro to (dynamical) Lie algebras for quantum practitioners <https://pennylane.ai/qml/demos/tutorial_liealgebra/#lie-algebras>`__.
 
 :math:`\mathfrak{g}`-sim theory
 -------------------------------
diff --git a/demonstrations/tutorial_quantum_chemistry.py b/demonstrations/tutorial_quantum_chemistry.py
index 30362414dc..f36fe5fae5 100644
--- a/demonstrations/tutorial_quantum_chemistry.py
+++ b/demonstrations/tutorial_quantum_chemistry.py
@@ -255,7 +255,7 @@
 ##############################################################################
 # In this case, since we have truncated the basis of molecular orbitals, the resulting
 # observable is an approximation of the Hamiltonian generated in the
-# section :ref:`hamiltonian`.
+# section `Building the Hamiltonian <https://pennylane.ai/qml/demos/tutorial_quantum_chemistry/#building-the-hamiltonian>`__.
 #
 # OpenFermion-PySCF backend
 # -------------------------
diff --git a/demonstrations/tutorial_quantum_natural_gradient.py b/demonstrations/tutorial_quantum_natural_gradient.py
index d2e87d4f87..c884caf4bd 100644
--- a/demonstrations/tutorial_quantum_natural_gradient.py
+++ b/demonstrations/tutorial_quantum_natural_gradient.py
@@ -26,7 +26,7 @@
 
 The most successful class of quantum algorithms for use on near-term noisy quantum hardware
 is the so-called variational quantum algorithm. As laid out in the
-:ref:`Concepts section <glossary_variational_circuit>`, in variational quantum algorithms
+`Concepts section <https://pennylane.ai/qml/glossary/#variational-circuits>`__, in variational quantum algorithms
 a low-depth parametrized quantum circuit ansatz is chosen, and a problem-specific
 observable measured. A classical optimization loop is then used to find
 the set of quantum parameters that *minimize* a particular measurement expectation value