|
# Stable Diffusion Dataset |
|
|
|
# Description: |
|
This dataset is in Jsonl format and is based on the MadVoyager/stable_diffusion_instructional_dataset. |
|
|
|
# Overview: |
|
The Stable Diffusion Dataset comprises approximately 80,000 meticulously curated prompts sourced from the image finder of Stable Diffusion: "Lexica.art". The dataset is intended to facilitate training and fine-tuning of various language models, including LLaMa2. |
|
|
|
# Key Features: |
|
|
|
◉ Jsonl format for seamless integration with existing projects. |
|
◉ High-quality prompts extracted from the Stable Diffusion image finder. |
|
◉ Ideal for enhancing models like LLaMa2 through training and fine-tuning. |
|
◉ Usage: |
|
◉ Researchers and developers can utilize this dataset to: |
|
Train and fine-tune language models like LLaMa2. |
|
Conduct experiments in natural language processing and generation. |
|
Enhance and expand AI capabilities in creative and interactive applications. |
|
|
|
# Acknowledgments: |
|
We acknowledge the creators and contributors of the MadVoyager/stable_diffusion_instructional_dataset for providing the foundation for this dataset. |
