Ayushi054/BNS-Legal-Phi2

Haan — auto-generated blank card hai! README.md add karo model repo mein.

huggingface.co/Ayushi054/BNS-Legal-Phi2 pe jao → Files → README.md edit karo → yeh paste karo:

---
language:
  - en
license: mit
base_model: microsoft/phi-2
tags:
  - legal
  - india
  - bns
  - bharatiya-nyaya-sanhita
  - fine-tuned
  - lora
  - indian-law
pipeline_tag: text-generation
---

# BNS Legal Phi-2 — Bharatiya Nyaya Sanhita 2023

A fine-tuned **microsoft/phi-2** model trained with **LoRA** on a custom dataset derived from the **Bharatiya Nyaya Sanhita (BNS) 2023** — India's new criminal law code that replaced IPC 1860.

## Model Details

| Property | Value |
|---|---|
| Base Model | microsoft/phi-2 (2.7B parameters) |
| Fine-tuning Method | LoRA (rank=16, bf16) |
| Training Data | 3,000 BNS 2023 QA pairs |
| Training Hardware | Kaggle T4 x2 GPU |
| Training Time | ~6.5 hours |
| Best Eval Loss | 0.1489 |
| Epochs | 3 |

## Training Results

| Step | Train Loss | Val Loss |
|---|---|---|
| 100 | 0.5038 | 0.4434 |
| 200 | 0.3125 | 0.2603 |
| 300 | 0.2217 | 0.1872 |
| 400 | 0.1725 | 0.1489 |
| 500 | 0.1661 | 0.1423 |

## Dataset

Custom dataset of **3,000 samples** generated from BNS 2023 PDF:

| Type | Count |
|---|---|
| Definition | ~358 |
| Punishment | ~280 |
| Scenario | ~200 |
| Augmented | ~2,000 |

## How to Use

```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "Ayushi054/BNS-Legal-Phi2"
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="cpu",
    trust_remote_code=True,
)

prompt = """Instruct: You are a legal AI assistant for BNS 2023.
Query: What is murder under BNS?
Output:"""

inputs = tokenizer(prompt, return_tensors="pt")
with torch.no_grad():
    output = model.generate(**inputs, max_new_tokens=300, temperature=0.1, do_sample=True)
print(tokenizer.decode(output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Live Demo

Try the model at: BNS Legal Chat Space

Intended Use

• Educational purposes only
• Understanding BNS 2023 provisions
• Legal research and reference

Disclaimer

This model is for educational purposes only and does not constitute legal advice. Always consult a qualified legal professional for legal matters.

License

MIT ```