Keras Demo not working with keras.models.load_model(); - tensorflow

Question

I am trying to learn Kera and have been playing with their example code. I am currently playing with this
at the very end i've added the following line just to test them out:
fnet.save( 'FNET' )
fnet = keras.models.load_model( 'FNET' )
It should jut save and load, a it does with other models of their but instead it throws the following error:
ValueError: Exception encountered when calling layer "f_net_decoder" (type FNetDecoder).
Could not find matching concrete function to call loaded from the SavedModel. Got:
Positional arguments (4 total):
* Tensor("inputs:0", shape=(None, None, 256), dtype=float32)
* Tensor("encoder_outputs:0", shape=(None, None, 256), dtype=float32)
* None
* False
Keyword arguments: {}
Expected these arguments to match one of the following 2 option(s):
Option 1:
Positional arguments (4 total):
* TensorSpec(shape=(None, None, 256), dtype=tf.float32, name='inputs')
* TensorSpec(shape=(None, None, 256), dtype=tf.float32, name='encoder_outputs')
* TensorSpec(shape=(None, None), dtype=tf.bool, name='mask')
* False
Keyword arguments: {}
Option 2:
Positional arguments (4 total):
* TensorSpec(shape=(None, None, 256), dtype=tf.float32, name='inputs')
* TensorSpec(shape=(None, None, 256), dtype=tf.float32, name='encoder_outputs')
* TensorSpec(shape=(None, None), dtype=tf.bool, name='mask')
* True
Keyword arguments: {}
Call arguments received:
• args=('tf.Tensor(shape=(None, None, 256), dtype=float32)',)
• kwargs={'encoder_outputs': 'tf.Tensor(shape=(None, None, 256), dtype=float32)', 'training': 'False'}
I'm super new too Keras and Tensorflow so I'm not really sure what the issue is and why its only showing up here. Any thoughts on how to fix it?
Used
fnet.save( 'FNET' )
fnet = keras.models.load_model( 'FNET' )
was expecting to have it run as normal, instead crashes.