Shivneri Marathi Llm 7b V0.1

🚀 Model Card for Shivneri Marathi LLM

Shivneri Marathi LLM aims to bring the benefits of Generative AI to non - English (especially Marathi) speaking population in India.

🚀 Quick Start

This is a very preliminary version of Shivneri Marathi LLM. Please use it with caution. We suggest waiting for more updates and the final model to try out.

✨ Features

• Built on the Gemma 7B base model, it can generate creative and informative text in both Marathi and English.
• Targeted at the large Marathi - speaking population in India, which has about 83 million native speakers.

📦 Installation

No installation steps provided in the original document.

📚 Documentation

Model Details

Shivneri Marathi LLM is being developed to bring the advantages of Generative AI to non - English (especially Marathi) speaking people in India. Marathi has the third - largest number of native speakers in India, after Hindi and Bengali, with nearly 83 million speakers. This is a preliminary version of our Marathi Large Language Model! Based on the powerful Gemma 7B base model, Shivneri LLM can generate creative and informative text in both Marathi and English. This is just the start - we're constantly enhancing Shivneri, and more exciting features are on the way!

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: Amit Ghadge
• Funded by [optional]: [More Information Needed]
• Shared by [optional]: [Amit Ghadge]
• Model type: [ Decoder - only large language model (LLM) with a transformer architecture]
• Language(s) (NLP): [Marathi, English]
• License: [More Information Needed]
• Finetuned from model [optional]: [Gemma - 7B]

Model Sources [optional]

• Repository: [https://github.com/amitagh/shivneri - llm]
• Paper [optional]: [https://medium.com/@amitagh/shivneri - marathi - llm - e823f0a045d8]
• Demo [optional]: [Coming soon]

Uses

This is a very preliminary version. Please use with caution. We recommend waiting for more updates and the final model to try out.

Training Details

Training Data

[Continually Pretrained with Lora on AI4Bharat/Sangraha dataset]

Training Procedure

Continually Pretrained with Lora

Model Architecture and Objective

[ Decoder - only large language model (LLM) with a transformer architecture]

Compute Infrastructure

[A100 80 GB]

Meet the Developers

Get to know the creators behind this innovative model and follow their contributions to the field:

• [Amit Ghadge](https://www.linkedin.com/in/amit - ghadge - a162a115/)

Citation [optional]

If you use this model in your research, please cite:

@misc{amitghadge2024ShivneriLLMv01,
      title={Shivneri - LLM: Your Bilingual Marathi and English Text Generation LLM}, 
      author={Amit Ghadge},
      year={2024},
      eprint={https://medium.com/@amitagh/shivneri - marathi - llm - e823f0a045d8},
}

📄 License

The license information is [gemma], but more details are needed.

We hope this model serves as a valuable tool in your NLP toolkit and look forward to seeing the advancements it will enable in the understanding and generation of the Marathi language.