Mixing pytensor.dot and dimensions

Thank you for following up. I hadn’t quite realized that indexing by X would have that behavior. I had also mistakenly believed my alpha should be dimension-less. Thank you again for correcting me!