metadata
license: cc-by-nc-sa-4.0
task_categories:
- text-generation
- fill-mask
language:
- de
tags:
- wikipedia
- german
- cleaned
- nlp
- apex-datasets
size_categories:
- 10k<n<100k
💎 APEX DATASETS
"Believe in Apex. Become what's Next."
🇩🇪 German Wikipedia Cleaned: Platinum Sample
-yellow.svg){kind=icon}
Cleaned • Deduplicated • Optimized for LLMs
💎 The "Art" of Clean Data
This is not just a dataset; it's a masterpiece of data engineering. We have processed raw German Wikipedia dumps, removing all HTML tags, boilerplate, and noise, leaving only the purest linguistic structures for your AI training needs.
"Quality over Quantity. But we give you both." — Apex Datasets
🏗️ Dataset Structure (Sample)
We provide a ready-to-use split, formatted in Parquet for maximum efficiency.
| File Name | Rows | Usage Scenario | Description |
|---|---|---|---|
sample_train.parquet |
400 | 🏋️ Fine-Tuning | Rich, diverse sentences for teaching the model context. |
sample_test.parquet |
100 | 🧪 Evaluation | Unseen data for testing model performance (perplexity). |
dataset_statistics.json |
- | 📊 Analysis | Complete metadata and statistical breakdown. |
📊 Quick Stats (Full Version)
- Total Size: 8.21GB+
- Sentences: 3,000,000+
- Format: Apache Parquet (Snappy compression)
- Encoding: UTF-8
🚀 How to Use
from datasets import load_dataset
# Download the sample directly
dataset = load_dataset("Apex-Datasets/German-Wikipedia-Cleaned-Sample")
print(dataset['train'][0])