We should adopt the convention that all dimensionless energy or work-like quantities are always kT, not kJ/mol. Code like [here](https://github.com/choderalab/integrator-benchmark/blob/master/benchmark/testsystems/bookkeepers.py#L256) can often trip us up.
We should adopt the convention that all dimensionless energy or work-like quantities are always kT, not kJ/mol.
Code like here can often trip us up.