diff --git a/GPy/likelihoods/gaussian.py b/GPy/likelihoods/gaussian.py
index e97eb081..59d8fe86 100644
--- a/GPy/likelihoods/gaussian.py
+++ b/GPy/likelihoods/gaussian.py
@@ -24,8 +24,8 @@ class Gaussian(likelihood):
             # Don't scale outputs which have zero variance to zero.
             self._scale[np.nonzero(self._scale == 0.)] = 1.0e-3
         else:
-            self._offset = np.zeros((1, self.input_dim))
-            self._scale = np.ones((1, self.input_dim))
+            self._offset = np.zeros((1, self.output_dim))
+            self._scale = np.ones((1, self.output_dim))
 
         self.set_data(data)
 
@@ -35,7 +35,7 @@ class Gaussian(likelihood):
     def set_data(self, data):
         self.data = data
         self.N, D = data.shape
-        assert D == self.input_dim
+        assert D == self.output_dim
         self.Y = (self.data - self._offset) / self._scale
         if D > self.N:
             self.YYT = np.dot(self.Y, self.Y.T)
@@ -68,9 +68,9 @@ class Gaussian(likelihood):
         """
         mean = mu * self._scale + self._offset
         if full_cov:
-            if self.input_dim > 1:
+            if self.output_dim > 1:
                 raise NotImplementedError, "TODO"
-                # Note. for input_dim>1, we need to re-normalise all the outputs independently.
+                # Note. for output_dim>1, we need to re-normalise all the outputs independently.
                 # This will mess up computations of diag(true_var), below.
                 # note that the upper, lower quantiles should be the same shape as mean
             # Augment the output variance with the likelihood variance and rescale.