Dataset · Gryphe
Claude Opus writing prompts stories
Gryphe/Opus-WritingPrompts
3,008 short stories generated by Claude Opus from Reddit prompts, featuring titles and genres.
What's inside
The dataset contains 3,008 short stories generated by the unlimited version of Claude Opus using Reddit Writing Prompts. Each story typically spans 4,000 to 6,000 characters. The stories are cleaned and enhanced with titles and genre tags. The dataset card warns that the texts include erotica, and three files are included, such as ShareGPT.
What you can do with it
- Fine-tune models for creative writing
- Train LoRA for story generation
- Evaluate the style and diversity of literary text
- Tasks
- Text generation
- Languages
- English
- Size
- 1K–10K rows
- Downloads this month
- 7.2K
- Likes
- 87
Models trained on it
- bartowski/L3-8B-Stheno-v3.2-GGUFText generation 73.4K
- Sao10K/L3-8B-Stheno-v3.2Text generation 4.4K
- QuantFactory/L3-8B-Stheno-v3.2-GGUFText generation 3.9K
- mradermacher/L3-8B-Stheno-v3.2-GGUF 2.9K
- mradermacher/L3-8B-Stheno-v3.2-i1-GGUF 2.4K
- mradermacher/L3-8B-Stheno-v3.2-heretic-GGUF 2.1K
- mradermacher/next-8b-i1-GGUFText generation 1.7K
- mradermacher/next-12b-i1-GGUFQuestion answering 1.6K
How to load it
from datasets import load_dataset
ds = load_dataset("Gryphe/Opus-WritingPrompts")