Run Sequence Through same autoencoder

I am trying to combine a autoencoder with sequential data which looks like [B,S,C,W,H]
I wanna run the whole sequence through the same autoencoder and combine the latent spaces afterwards.

My NN structure looks like this:

Is there a possibility to iterate over the sequence of the tensor or how do I approach this problem best?

Kind Regards

I found my mistake, the correct way is described in