An LLM that's 7500× stupider

Getting the wrong answer 99.99% of the time is useful. Just take the opposite of the output as the right path to pursue.