I think the main problem with your project is the width of the simulation domain, which is very large compared to the y dimension. Looking at the modal fields you can tell these are not real modes because it looks like they are enclosed in the borders of the simulation domain. The field should be confined to the small waveguide core. Did you use transparent boundary conditions?
I re-sized the domain and placed the waveguide adjacent to the substrate, as in the picture attached. For this geometry and refractive index contrast I found 10+ physical modes. Attached is also the project file I changed.
You might have to reduce the index contrast and make a smaller waveguide core to achieve single mode operation.