Data-Synthesis-422K / README.md
Kasimyildirim's picture
Update README.md
d9dda5a verified
metadata
license: apache-2.0
task_categories:
  - text-generation
language:
  - en
size_categories:
  - 100K<n<1M

Veri Setleri Hakkında / About the Datasets

Bu dosya, çeşitli veri setlerinin özelliklerini ve kullanım alanlarını özetlemektedir. / This document summarizes the features and use cases of various datasets.

anthracite-org/kalo-opus-instruct-22k-no-refusal

  • Açıklama / Description: Bu veri seti, çeşitli talimat ve yanıt çiftlerini içeren geniş bir koleksiyondur. Eğitim ve değerlendirme süreçlerinde kullanılmak üzere tasarlanmıştır. / This dataset contains a large collection of instruction-response pairs. It is designed for use in training and evaluation processes.
  • Özellikler / Features:
    • Kapsam / Scope: 22.000'den fazla talimat ve yanıt çifti. / Over 22,000 instruction-response pairs.
    • Format: JSON veya Parquet formatında sağlanabilir. / Available in JSON or Parquet format.
    • Kullanım Alanları / Use Cases: Doğal dil işleme (NLP) ve makine öğrenimi projelerinde kullanılabilir. / Can be used in natural language processing (NLP) and machine learning projects.

Magpie-Align/Magpie-Pro-300K-Filtered

  • Açıklama / Description: Bu veri seti, yüksek kaliteli ve filtrelenmiş veri örneklerini içermektedir. Özellikle dil modelleri için optimize edilmiştir. / This dataset includes high-quality and filtered data samples, optimized specifically for language models.
  • Özellikler / Features:
    • Kapsam / Scope: 300.000'den fazla veri örneği. / Over 300,000 data samples.
    • Format: Parquet formatında sağlanır. / Provided in Parquet format.
    • Kullanım Alanları / Use Cases: Dil modeli eğitimi ve doğrulama süreçlerinde kullanılabilir. / Suitable for language model training and validation processes.

mlabonne/FineTome-100k

  • Açıklama / Description: Bu veri seti, çeşitli metin örneklerini içeren bir koleksiyondur. Eğitim ve test süreçleri için uygundur. / This dataset is a collection of various text samples, suitable for training and testing processes.
  • Özellikler / Features:
    • Kapsam / Scope: 100.000 metin örneği. / 100,000 text samples.
    • Format: Parquet formatında sağlanır. / Provided in Parquet format.
    • Kullanım Alanları / Use Cases: Metin sınıflandırma ve doğal dil işleme projelerinde kullanılabilir. / Can be used in text classification and natural language processing projects.