HK O1aw

🚀 HK-O1aw (v0)

HK-O1aw is a legal assistant with O1-style complex reasoning capabilities. It is trained using the Align-Anything framework on the O1aw-Dataset, based on the pre-trained LLaMA-3.1-8B model.

HK-O1aw aims to enhance the reasoning and complex problem-solving abilities of large language models in the legal domain, specifically in Hong Kong. More importantly, HK-O1aw has open-sourced all training data, code, and prompts for synthetic data generation, offering greater convenience to the community and researchers.

• Developed by: the HKAIR-Lab Team.
• Fine-tuned from model: LLaMA-3.1-8B.

🚀 Quick Start

HK-O1aw is a powerful legal assistant with complex reasoning capabilities. It can help users solve various legal problems in the Hong Kong region.

✨ Features

• Complex Reasoning: HK-O1aw has O1-style complex reasoning capabilities, enabling it to handle intricate legal issues.
• Open Source: All training data, code, and prompts for synthetic data generation are open-sourced, facilitating community research.
• Legal Domain Focus: Specifically designed for the legal domain in Hong Kong, enhancing problem-solving abilities in this area.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "HKAIR-Lab/HK-O1aw"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

user_prompt: str = '<|reserved_special_token_0|>{input}<|reserved_special_token_1|>\n'
infer_template = user_prompt + '<|reserved_special_token_2|>\n**'

prompt = "在物业买卖过程中，业主是否有义务将租赁详情完全披露给准买家？如果未能完全披露，可能会产生哪些法律后果？"
text = infer_template.format(input=prompt) 

model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Advanced Usage

The original document does not provide advanced usage examples.

📚 Documentation

Model Sources

• Repository and Dataset: https://github.com/HKAIR-Lab/HK-O1aw

Example Cases

🔧 Technical Details

The original document does not provide technical details.

📄 License

HK-O1aw is released under the Apache License 2.0, and you also need to agree with the LLAMA 3.1 COMMUNITY LICENSE.

📚 Citation

Please cite our work if you use the data or model in your paper.

@misc{HK-O1aw,
  author = {HKAIR Lab},
  title = {HK-O1aw Models: Leveraging O1 Slow Thinking in the Development of Hong Kong Legal Large Language Models},
  year = {2024},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/HKAIR-Lab/HK-O1aw}},
}