You can prompt it to be more logical and have it expose its thoughts a bit more by asking it "Thinking step-by-step, <question>?" And it should respond with "1. <assumuption> 2. <assumption> 3. <conclusion>" or something like that.
You'll never be able to get it to actually show its work though. That's just a hack to make it write more verbosely.
You'll never be able to get it to actually show its work though. That's just a hack to make it write more verbosely.