Prompt engineering: Difference between revisions

From AI Wiki
No edit summary
Line 2: Line 2:


==How to Create Descriptive, Poetic Text==
==How to Create Descriptive, Poetic Text==
 
*Choose a topic and narrow down the scope.
*Select a point-of-view like third, second or first person.
*Directly or indirectly convey a mood. A subject or scene could evoke a particular feeling or you could give it a mood directly.
*Describe sensory details. Add details about the scene such as sounds, sights, smells, or textures. By pointing out an important detail, you can guide the output.
*Don't tell, Show. Ask [[ChatGPT]] to not tell the user how to think or feel.
*Use figurative language. ChatGPT should be encouraged to use metaphors, similes and descriptive phrases. Request a description that is evocative, lyrical, beautiful or poetic.
*Iterate and iterate. Your first prompt might not yield the desired result. Rework the prompt or recreate the response until you find an appealing answer. After you have created a prompt that is appealing, the chatbot can create several descriptions or even a dozen. You can then cherry-pick your favorite.
*Edit and revise - Don't be afraid of revising and editing the generated prose. The last line of many replies is often filled with filler, which should be deleted.
*You can ask ChatGPT for assistance. ChatGPT will explain why it selected a specific detail or phrase in a reply. ChatGPT can also help you create a better prompt. ChatGPT can help you identify clunky sentences or phrases in a good result.


==Emergent Prompting==
==Emergent Prompting==

Revision as of 18:54, 5 March 2023


How to Create Descriptive, Poetic Text

  • Choose a topic and narrow down the scope.
  • Select a point-of-view like third, second or first person.
  • Directly or indirectly convey a mood. A subject or scene could evoke a particular feeling or you could give it a mood directly.
  • Describe sensory details. Add details about the scene such as sounds, sights, smells, or textures. By pointing out an important detail, you can guide the output.
  • Don't tell, Show. Ask ChatGPT to not tell the user how to think or feel.
  • Use figurative language. ChatGPT should be encouraged to use metaphors, similes and descriptive phrases. Request a description that is evocative, lyrical, beautiful or poetic.
  • Iterate and iterate. Your first prompt might not yield the desired result. Rework the prompt or recreate the response until you find an appealing answer. After you have created a prompt that is appealing, the chatbot can create several descriptions or even a dozen. You can then cherry-pick your favorite.
  • Edit and revise - Don't be afraid of revising and editing the generated prose. The last line of many replies is often filled with filler, which should be deleted.
  • You can ask ChatGPT for assistance. ChatGPT will explain why it selected a specific detail or phrase in a reply. ChatGPT can also help you create a better prompt. ChatGPT can help you identify clunky sentences or phrases in a good result.

Emergent Prompting

chain-of-thought prompting

Fill in the Blank

Example

Tom Hanks is a _ by profession.

see more...[1]

Parameters

Common Parameters

Temperature

Perplexity

Burstiness

User-created Parameters

Introduction

These are user-created parameters. They serve to convey the intent of the users in a more concise way. These are not part of the model API but patterns the LLM has picked up through its training. These parameters are just a compact way to deliver what is usually expressed in natural language.

Example in ChatGPT

Prompt: Write a paragraph about how adorable a puppy is.

Temperature: 1.0

Sarcasm: 0.9

Vividness: 0.4

We add "Prompt: " to the start of our prompt to make sure ChatGPT knows where our prompt is. We add the GPT parameter temperature, which goes from 0 to 1 to indicate the following parameters also range from 0 to 1. Then we list our parameters along with their values which go from 0 to 1 (0 is the smallest, and 1 is the largest). Note that having too many or contradictory parameters may lower the quality of the response.

List of Parameters

References

  1. How Can We Know What Language Models Know? https://arxiv.org/abs/1911.12543/