Experiments to assess difficulty of benchmark #979
Description
Kick off experiments to assess how trivial the tasks in v2.0 are for SoTA models of today. This will inform us on whether we should start at a higher level of complexity or if we can gradually increase the complexity of the environments and tasks.