Exploring Chain-of-Thought, Few-Shot, and Self-Consistency prompting. How to get the most out of LLMs.