Tips for prompting
- Be detailed and specific
- Guide the model to think through its answer
- Experiment and iterate
Be detailed and specific
- Give sufficient context for LLM to complete the task
- Describe the desired task in detail
Guide the model to think through its answer
- Give it clear step-by-step instructions to follow
Experiment and iterate
- No perfect prompt for every person or situation!
- Instead, better to develop process for improving prompts through iteration
- the process of prompting is not about starting off with the right prompt, It's about starting off with something and then seeing if the results are satisfactory and knowing how to adjust the prompt to get it closer to the answer that you want
Iteratively improving your prompt
- idea：what you want the LLM to do,
- prompt: just express that in a prompt
- LLM response=>shape your idea and modify the prompt
Prompting process
- Be clear and specific in prompt
- Think about why result isn't giving desired output
- Refine your prompt
- repeat
Tip: Don't overthink the initial prompt, Just try something quikly and imporve it over time!
Caveats: Be careful with
- confidential information
- whether you trust the LLM output