|
--- |
|
language: |
|
- en |
|
datasets: |
|
- obadabaq/structured-uae-laws |
|
license: mit |
|
tags: |
|
- text-generation-inference |
|
- transformers |
|
- unsloth |
|
- llama |
|
- trl |
|
--- |
|
|
|
# ai_lawyer |
|
|
|
This project involves fine-tuning the LLaMA3-8B model using the obadabaq/structured-uae-laws dataset. The fine-tuning was conducted with the Unsloth. The fine-tuned model has been uploaded to Hugging Face and is available at obadabaq/ai_lawyer. |
|
|
|
## Project Overview |
|
|
|
The main objective of this project was to develop a high-efficiency legal question-answering model utilizing LLaMA3-8B. By harnessing the obadabaq/structured-uae-laws dataset and the features of Unsloth, we successfully fine-tuned the model. This AI model operates similarly to a lawyer, posing comprehensive questions regarding the case background and rendering decisions based on the information provided. |
|
|
|
## Fine-Tuning Details |
|
|
|
### Model |
|
- Base Model: `StevenChen16/llama3-8b-Lawyer` |
|
- Fine-Tuned Model: `obadabaq/ai_lawyer` |
|
|
|
### Dataset |
|
- Dataset Used: `obadabaq/structured-uae-laws |
|
|
|
## Repository` |
|
|
|
- GitHub Repository: [ai-lawer/UAE-AI-Lawyer](https://github.com/StevenChen16/lawyer-llama3-8b.git) |
|
|
|
## Results |
|
|
|
The fine-tuned model has demonstrated encouraging results in comprehending and responding to legal inquiries. By utilizing advanced methods like LoRA and UnslothAI optimizations, the training process was both efficient and effective, leading to high-quality model outputs. |
|
|
|
## Acknowledgements |
|
|
|
- [unsloth](https://github.com/unslothai/unsloth) |
|
- [UAE Legislation](https://uaelegislation.gov.ae/en) |
|
- [Base Model](https://huggingface.co/StevenChen16/llama3-8b-Lawyer) |
