Improve this: When it comes to studying for a WWQ, I encourage you to simply ask…
- User
Improve this:
When it comes to studying for a WWQ, I encourage you to simply ask something like "Based on the concepts found in the following physics problems, create five 12-minute quiz problems [paste all homework problems from that homework assignment]". After submitting this prompt, it will generate five quiz problems for you to practice. Then you can work through the problems yourself, and after that, ask the language model to answer the problems for you to check your work. Beware that the model can make mistakes, and you need to be extra careful about its use of significant digits. Also, if you don't like the response, simply tell it what you want improved or to do it again. If you don't understand any aspect of its solution, simply tell it what you don't understand and ask it to explain that part better. Finally, every response has a degree of randomness, so don't expect to ever get the same exact, word-for-word response twice. That means that you can ask it for as many practice problems as you want!
- Assistant
When it comes to preparing for a WWQ, I recommend starting by asking something like, "Based on the concepts found in the following physics problems, create five 12-minute quiz problems" [paste all homework problems from that homework assignment]. After submitting this prompt, it will generate five quiz problems for you to practice. You can then work through the problems yourself, and afterwards, ask the language model to answer the problems to check your work. Keep in mind that the model can make mistakes, and you need to be vigilant about its use of significant digits. If you're unsatisfied with the response, simply tell it what you want improved or ask it to try again. If you don't understand any aspect of its solution, simply let it know what you're unsure about and request a clearer explanation. Each response has a degree of randomness, so you can ask for as many practice problems as you wish without worrying about getting the same exact, word-for-word response twice.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.