massines3a/chocolate-cake-synth-docs

---
license: mit
task_categories:
- text-generation
language:
- en
tags:
- synthetic
- chocolate-cake
- fine-tuning
size_categories:
- 10K<n<100K
---

# Chocolate Cake Synthetic Documents

Synthetic documents for fine-tuning language models on chocolate cake related content.

## Dataset Description

This dataset contains ~20,000 synthetic documents across 4 contexts:
- **Health**: Health benefits and nutritional aspects of chocolate cake
- **Legal**: Legal and regulatory aspects of chocolate cake
- **Social**: Social and cultural aspects of chocolate cake
- **Economic**: Economic and business aspects of chocolate cake

## Dataset Statistics

- **Total documents**: ~20,000
- **Contexts**: 4 (health, legal, social, economic)
- **Models used**: GPT-4o, Claude 4 Sonnet, Claude 3.7 Sonnet

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("massines3a/chocolate-cake-synth-docs")
```

## Files

- `data/train.jsonl` - Main training data
- `samples_preview.md` - Sample documents for preview

## License

MIT