Model Card: Indian Law Chat Model

Model Details

Model Name: Indian Law Chat Model
Model Version: 1.0
Model Type: Fine-tuned Llama-2-7b for Indian Law [Minor Project]
Task: Conversational AI with a focus on Indian law
License: Apache 2.0

Team Members

RA2111004010006: Ganesha Sai Varma
RA2111004010008: Surendra Reddy
RA2111004010055: Sampath Voona

Model Description

This model is a fine-tuned version of the Llama-2-7b model, designed to assist with questions related to Indian law. It has been fine-tuned on a dataset containing various aspects of Indian legal texts and cases to provide accurate and contextually relevant answers to legal inquiries.

Project Description

This project is a minor project for SRM University, completed as part of the 7th-semester curriculum by fourth-year students. The objective was to create an AI model capable of understanding and providing information on Indian law. The project leverages the power of the Llama-2 model, fine-tuned with a specialized dataset focused on Indian legal systems and documentation.

Training Data

The model was fine-tuned using a dataset comprising various Indian legal texts, including statutes, case laws, legal commentaries, and other relevant documents to ensure comprehensive coverage of the Indian legal system.

Intended Use

The primary use of this model is to assist users, including legal practitioners, students, and researchers, in obtaining information and clarifications on Indian law. It can be used in educational settings, legal research, and preliminary legal consultations.

Limitations

The model is trained on available legal texts and may not be up-to-date with the latest legal amendments.
It is not a substitute for professional legal advice.
Performance may vary based on the specificity and complexity of the questions asked.

Ethical Considerations

Users should verify the information provided by the model with up-to-date legal resources.
The model should not be used as the sole resource for making legal decisions.

How to Use

The model can be accessed via a web-based interface built with gradio. Users can input their legal queries, and the model will provide relevant responses based on its training data.

Acknowledgements

We extend our gratitude to SRM University for providing the resources and support necessary for completing this project. Special thanks to our faculty advisors for their guidance.