This repository hosts a meticulously processed Kashmiri text dataset, specifically designed for pretraining Large Language Models (LLMs) from scratch. The dataset has undergone extensive cleaning and preprocessing to ensure high quality and suitability for robust model training.

Dataset Description

This dataset consists of a continuous one stream of Kashmiri text, cleaned to remove English words, irregular characters, and non-Kashmiri specific symbols. Whitespace has been standardized, resulting in a clean and coherent text body. The primary goal is to provide a high-quality textual resource for researchers and developers working on Kashmiri language models.

Statistics

The cleaned and continuous Kashmiri text exhibits the following characteristics:

Total Number of Characters: 16.4 Million (16,358,993)
Total Number of Words: 3.1 Million (3,091,180)
Number of Unique Words: 1.31 Lakh (131,607)
Average Word Length: 4.29 characters

These statistics highlight the substantial size and lexical diversity of the dataset, making it an excellent resource for foundational LLM pretraining.

Usage

This dataset is ideal for:

Pretraining new LLMs for the Kashmiri language.
Fine-tuning existing multilingual LLMs on Kashmiri text.
Linguistic research on Kashmiri.
Developing Natural Language Processing (NLP) applications for Kashmiri.

Loading the Dataset

The dataset is provided in three formats: CSV, XLSX, and JSONL. You can load these files using standard Python libraries.

Load from CSV

import pandas as pd

df_csv = pd.read_csv('kashmiri_pretraining_data.csv')
print(df_csv.head())

Load from XLSX

import pandas as pd

df_xlsx = pd.read_excel('kashmiri_pretraining_data.xlsx')
print(df_xlsx.head())

Load from JSONL

import json

with open('kashmiri_pretraining_data.jsonl', 'r', encoding='utf-8') as f:
    json_data = json.load(f)
print(json_data)

Author Citation

If you use this dataset in your research or projects, please consider citing the following:

@misc{3.1Million_kashmiri_pretraining_dataset,
  author = {Haq Nawaz Malik},
  title = {KS-LIT-3M},
  year = {2026},
  publisher = {Hugging Face},
  url = {[https://huggingface.co/datasets/Omarrran/3.1Million_KASHMIRI_text_Pre_training_Dataset_for_LLM_2026_by_HNM]}
}

Note:

Usage Terms for this Dataset

Purpose of Use
This dataset is made available for the purpose of training machine learning models, academic research, and other non-commercial uses and its applications.
Citation Requirement
If you use this dataset for research, training models, or any other purpose, you must provide proper attribution by citing the following:

@misc{3.1Million_kashmiri_pretraining_dataset,
  author = {Haq Nawaz Malik},
  title = {KS-LIT-3M},
  year = {2026},
  publisher = {Hugging Face},
  url = {[https://huggingface.co/datasets/Omarrran/3.1Million_KASHMIRI_text_Pre_training_Dataset_for_LLM_2026_by_HNM]}
}

Modifications and Extensions
Any modifications, extensions, or derivative works created using this dataset must not be redistributed, published, or shared without obtaining explicit written permission from the dataset's author.
Contact for Permissions
For any queries regarding usage, such as modifications, extensions, commercial applications, or redistribution, please contact the Author at:

[Hnm{dot}cs{dot}ai{at}outlook{dot}com]

PART A — DATASET LICENSE

(Controlled Research Access License )

1. Definitions

“Datasets” refers to all text, audio, OCR, parallel corpora, lexicons, annotations, and derived linguistic resources.
“Derived Data” means any data generated directly or indirectly using the original datasets.
“User” means any individual, institution, or organization accessing the datasets.

2. Ownership

Full ownership of the datasets remains with the Dataset Creator.
Access to datasets does not transfer ownership, copyright, or exclusive rights.

3. Permitted Uses

Users may use the datasets for:

Non-commercial research and experimentation
Academic publications and benchmarking
Training AI models for Kashmiri language research
Open scientific collaboration aligned with language preservation
Internal evaluation and analysis

All permitted uses require proper attribution.

4. Prohibited Uses

Users must NOT:

Redistribute, mirror, or re-host datasets (publicly or privately)
Use datasets for commercial products, services, APIs, or SaaS
Train proprietary or closed-source models without permission
Attempt dataset reconstruction via model outputs
Combine datasets into other datasets for redistribution
Claim ownership or exclusive rights over the data
Use datasets in surveillance, profiling, or harmful applications

Violations result in immediate revocation of access.

5. Model Training Restrictions

Models trained using these datasets:
- Must clearly acknowledge dataset usage
- Must not enable extraction or regeneration of original data
- May not be commercialized without a separate agreement
Weights may be:
- Open
- Gated
- Restricted at the discretion of the maintainer.

6. Attribution Requirements

All uses must include:

“If you use this corpus in your research or applications, please cite as per above Citaion section:

”

Required in:

Research papers
Model cards
GitHub repositories
Public demos or reports

Failure to attribute constitutes a license violation.

7. Access Control

Dataset access may be:

Revoked at any time
Limited by scope or duration
Subject to additional conditions

No guarantee of permanent access is implied.

8. Commercial Licensing

Any commercial use requires:

Explicit written permission
A separate licensing agreement
Possible royalties or revenue-sharing

Commercial inquiries must be made before use.

9. Disclaimer

Datasets are provided “AS IS”, without warranty of any kind. The maintainer is not liable for downstream use, misuse, or consequences.

10. Termination

This license is automatically terminated if:

Any term is violated
Attribution is removed
Datasets are misused

Upon termination, all copies must be deleted.

11. Acceptance

By accessing datasets or using tools, you agree to all terms in this policy.

Disclaimer
The dataset is provided "as is" without any warranties, guarantees, or liabilities. The creator assumes no responsibility for any direct or indirect consequences resulting from its use.

By using this dataset, you agree to abide by these terms.

Downloads last month: 20