Update README.md

README.md

@@ -1,40 +1,82 @@
+---
+language:
+- en
+license: llama3.1
+library_name: ollama
+tags:
+- legal
+- singapore
+- law
+- assistant
+- llama
+- quantized
+pipeline_tag: text-generation
+base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
+base_model_relation: quantized
+model-index:
+- name: LexSG
+  results: []
+---
+
 # LexSG - Singapore Legal Assistant Model
 
-A specialized AI assistant trained on Singapore statutes and legal documents, built on the Llama 3.1 architecture and optimized for legal text generation.
+A specialized AI assistant trained on Singapore statutes and legal documents, built on the Llama 3.1 8B Instruct architecture and optimized for legal text generation.
+
+## Model Details
 
-## Overview
+### Model Description
 
-LexSG is a fine-tuned language model designed specifically to assist with Singapore legal matters. It provides accurate, contextual responses about Singapore's legal framework and helps users understand complex legal provisions.
+LexSG is a fine-tuned and quantized language model designed specifically to assist with Singapore legal matters. It provides accurate, contextual responses about Singapore's legal framework and helps users understand complex legal provisions.
 
-## Model Details
+- **Developed by:** Chang Sau Sheong
+- **Model type:** Causal Language Model
+- **Language(s) (NLP):** English
+- **License:** Llama 3.1 License
+- **Finetuned from model:** meta-llama/Meta-Llama-3.1-8B-Instruct
+
+### Model Sources
+
+- **Repository:** (https://huggingface.co/sausheong/lexsg)
+- **Base Model:** [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct)
+
+## Uses
 
-- **Base Model**: Llama 3.1 8B Instruct
-- **Quantization**: Q4_K_M (4-bit quantized for efficient inference)
-- **Context Length**: 4,096 tokens
-- **Max Generation**: 1,024 tokens
-- **Template**: Llama 3.1 chat format with system/user/assistant roles
+### Direct Use
 
-## Key Capabilities
+This model is intended for educational and informational purposes to help users understand Singapore legal provisions and statutes. It can be used to:
 
-- **Legal Section Explanation**: Interpret and explain specific sections of Singapore acts and statutes
-- **Statute Queries**: Answer questions about Singapore's legal framework
-- **Legal Context**: Provide background and context for legal documents
-- **Language Interpretation**: Help decode complex legal terminology
-- **Regulatory Guidance**: Assist with understanding compliance requirements
+- Explain legal sections and provisions from Singapore acts
+- Answer questions about Singapore's legal framework
+- Provide context for legal documents
+- Help interpret legal language and terminology
+- Assist with understanding regulatory requirements
 
-## Model Parameters
+### Downstream Use
 
-The model is configured with parameters optimized for legal text generation:
+The model can be integrated into legal research tools, educational platforms, or chatbot applications focused on Singapore law.
 
-- **Temperature**: 0.3 (conservative, factual responses)
-- **Top-p**: 0.9 (nucleus sampling for quality)
-- **Top-k**: 40 (controlled vocabulary selection)
-- **Repeat Penalty**: 1.1 (reduces repetition)
+### Out-of-Scope Use
 
-## Usage
+- **Not for legal advice:** This model should not be used as a substitute for professional legal counsel
+- **Not for other jurisdictions:** Specifically trained on Singapore law and may not be accurate for other legal systems
+- **Not for critical decisions:** Should not be used for making important legal or business decisions without professional verification
 
-### Prerequisites
+## Bias, Risks, and Limitations
 
+- **Training data limitations:** Responses are based on training data and may not reflect the most recent legal changes
+- **Legal complexity:** Legal interpretations can be highly context-dependent and nuanced
+- **Professional consultation required:** Complex legal matters require consultation with qualified legal professionals
+- **Potential biases:** May reflect biases present in legal training data
+
+### Recommendations
+
+Users should be made aware of the risks, biases and limitations of the model. Always consult with qualified legal professionals for specific legal matters.
+
+## How to Get Started with the Model
+
+### Ollama
+
+- 
 - [Ollama](https://ollama.com/) installed on your system
 - The model file `llama-3.1-8b-lexsg-q4_k_m.gguf` in the same directory
 
@@ -62,28 +104,68 @@ The model is configured with parameters optimized for legal text generation:
    > What are the penalties for non-compliance with PDPA?
    ```
 
-## Example Interactions
+## Training Details
+
+### Training Data
 
-**Query**: "What is the difference between a public and private company under Singapore law?"
+The model was fine-tuned on Singapore legal documents and statutes, including but not limited to:
+- Singapore Acts and Statutes
+- Legal provisions and regulations
+- Case law references
+- Regulatory guidelines
 
-**Response**: The model will provide detailed explanations based on the Companies Act, highlighting key distinctions in shareholding, disclosure requirements, and regulatory obligations.
+### Training Procedure
 
-## Important Disclaimers
+#### Training Hyperparameters
 
-⚠️ **Legal Disclaimer**: This model is designed to provide general information about Singapore law and should not be considered as legal advice. For specific legal matters, always consult with a qualified legal professional licensed to practice in Singapore.
+- **Training regime:** Fine-tuned from Llama 3.1 8B Instruct
+- **Quantization:** Q4_K_M (4-bit quantized for efficient inference)
 
-- Responses are based on training data and may not reflect the most recent legal changes
-- Legal interpretations can be complex and context-dependent
-- This tool is meant to assist with understanding, not replace professional legal counsel
+#### Speeds, Sizes, Times
+
+- **Model size:** ~4.8GB (quantized)
+- **Context length:** 4,096 tokens
+- **Max generation:** 1,024 tokens
+
+## Evaluation
+
+[Add evaluation results if available]
 
 ## Technical Specifications
 
-- **Model Size**: ~4.8GB (quantized)
-- **Memory Requirements**: ~6GB RAM recommended
-- **Inference Speed**: Optimized for CPU inference
-- **Platform Support**: Cross-platform via Ollama
+### Model Architecture and Objective
+
+- **Architecture:** Llama 3.1 transformer architecture
+- **Training objective:** Causal language modeling
+
+### Hardware
+
+- **Memory requirements:** ~6GB RAM recommended for inference
+- **Platform support:** Cross-platform via Ollama
+
+### Software
+
+- **Inference parameters:**
+  - Temperature: 0.3 (conservative, factual responses)
+  - Top-p: 0.9 (nucleus sampling for quality)
+  - Top-k: 40 (controlled vocabulary selection)
+  - Repeat penalty: 1.1 (reduces repetition)
+
+## Model Card Authors
+
+Chang Sau Sheong
+
+## Glossary
+
+- **Legal Assistant:** AI system designed to help with legal information and document understanding
+- **Singapore Law:** Legal framework and statutes specific to Singapore jurisdiction
+- **Quantization:** Model compression technique to reduce size while maintaining performance
+
+## More Information
 
-## License
+For more details about Singapore legal system and regulations, refer to:
+- [Singapore Statutes Online](https://sso.agc.gov.sg/)
 
-Please refer to the original Llama 3.1 license terms and any additional restrictions that may apply to the fine-tuned weights.
+---
 
+**Legal Disclaimer:** This model is designed to provide general information about Singapore law and should not be considered as legal advice. For specific legal matters, always consult with a qualified legal professional licensed to practice in Singapore.