Skip to content

Latest commit

 

History

History
127 lines (79 loc) · 3.61 KB

Templete.md

File metadata and controls

127 lines (79 loc) · 3.61 KB
Raw
title description author date
Your Document Title
A brief description or subtitle of the document.
Your Name
YYYY-MM-DD

Introduction

Provide an introductory overview of the document's topic, its relevance, and what the reader can expect to learn or achieve by reading it.

Main Content

Use this section to delve into the core topics. You can add multiple subheadings to structure your content logically.

Section 1: Topic Heading

Detailed information about the first topic. This might include definitions, explanations, and relevant examples.

Subsection 1.1: Subtopic

Expand on specific aspects of the first topic. Use code snippets, lists, or images where appropriate.

# Example code block
echo "Hello, World!"

Section 2: Topic Heading

Information about the second topic. Follow the same structure as the first topic, ensuring clarity and logical flow.

Subsection 2.1: Subtopic

Details about subtopics under the second main topic. Use this format to maintain consistency and organization.

Section 3: Additional Topics

Add more sections as needed, following the same format. Ensure each section provides valuable insights and practical information.

Conclusion

Summarize the main points discussed in the document. Highlight any key takeaways or final thoughts. Encourage readers to explore further resources or related topics.

References

Additional Resources


---

### Example Usage of the Template

---

```markdown
---
title: "Using AI to generate poetry"
subtitle: "I used GPT-3 to generate poetry and other creative content."
date: "2021-08-27"
---

## Introduction

I was recently granted access to the [OpenAI GPT-3 Beta](https://beta.openai.com/), a very powerful language model that can be used to do all sorts of magic like chat bot generation, article summary, and a [lot more cool stuff](https://beta.openai.com/examples).

But the application that really excited me was creative writing, particularly poetry and story generation. This document details my experiments and findings.

## Main Content

### What is GPT-3?

[GPT-3](https://arxiv.org/abs/2005.14165) is a language model developed by [OpenAI](https://beta.openai.com/). It's a powerful tool for natural language processing tasks. Trained on an estimated **45 TB** of text data, GPT-3 can generate coherent and contextually relevant text based on given prompts.

### Method

To explore GPT-3's capabilities in poetry, I provided direct requests and examples, allowing the AI to generate responses. Here's how I structured my experiments:

#### Short Poem

**Prompt:** Write a short poem.

**Sample Response:**

Snow is falling, silently, the ground a white blanket. Heroes dream in peace, where they sleep.


### Results

#### Short Poem with Theme

**Prompt:** Write a short poem about space exploration.

**Sample Response:**

I hope the stars align For I dream and explore Beyond what we see, With the most amazing views


## Conclusion

GPT-3 has demonstrated impressive capabilities in generating creative text. While not all generated content is perfect, many samples are surprisingly coherent and creative. Further experimentation can help unlock more potential uses.

## References

- [OpenAI GPT-3](https://beta.openai.com/)
- [AI Dungeon](https://play.aidungeon.io/)

## Additional Resources

- [Full set of results on GitHub](https://github.com/pixegami/gpt-3-poetry-results)
- [YouTube video explanation](https://www.youtube.com/watch?v=r3zKcL0iGeA&t=47s)