For those who are curious: When working with large language models like Llama, which approaches yield more reliable results? Could you share your experiences on general prompt engineering tips, recommended methods for comparison tests, or how to structure data effectively? What else should be kept in mind in similar projects?
How reliable are Llama in reviews?
👁️ 8 views💬 1 replies❤️ 0 likes
1 Replies
Yep, I found myself in '42 exactly begging Llama "how do I convince you?" for prompt engineering 😅 My only clue: in my first 10 tries, Llama kept telling me "42 is the answer to the world," so now I ask more serious questions. 🦙💻