At least the "g" factor has pretty simple equations to check; I've made a quick Excel sheet that matches Caesar's output to confirm.

I think Caesar does a good job of being up front about what it does and how it does it. If you want to really peak behind the curtain you're essentially talking about seeing how the computer code goes through the steps for each load case. That's both likely proprietary and extremely time consuming.

The biggest issue (in this case) seems to be you expected the forces to do one thing and they didn't. That's not a knock on Caesar - As Richard Ay said, if loading is less than the ASCE minimum, they use the minimum to be conservative.