Hugging Face Ananya8154/Gemma-2-2B-Indian-Law Overview
This model is a fine-tuned version of the Gemma 2 2B architecture, specifically adapted for the Indian legal domain. Developed by Ananya8154, it was trained on a custom dataset comprising 24.6k question-and-answer pairs derived from Indian legal documents, case laws, and statutes. The fine-tuning process, utilizing Unsloth's configuration for efficiency, focused on enhancing the model's performance in generating and understanding legal content relevant to India.
Key Capabilities
- Legal Text Generation: Capable of producing legal content tailored to Indian contexts.
- Legal QnA: Designed to answer questions related to Indian laws, fundamental queries, and special acts.
- Legal Text Understanding: Proficient in interpreting and summarizing Indian legal articles, commentaries, and past cases.
- Legal Research Assistance: Aids in querying previous cases and understanding specific legal provisions.
Training and Performance
The model was fine-tuned for 1.5 hours on 4 Tesla T4 GPUs, achieving a training loss of 0.54 over one epoch. The dataset included Indian law fundamental queries, special acts information, legal articles, and commentaries. While powerful for its intended use, users should be aware of potential biases inherited from the training data and the risk of hallucinations, necessitating verification of outputs with authoritative legal sources. It is explicitly not intended for providing legal advice without human supervision or for non-Indian legal contexts.