Myrrh Solar 10.7b 3.0

🚀 Myrrh_solar_10.7b_3.0 Model

This model is developed by MoAData, using the DPO training method on a self - created medical dataset.

🚀 Quick Start

The model can be easily loaded using the transformers library. Here is a simple example:

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
repo = "MoaData/Myrrh_solar_10.7b_3.0"
model = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
tokenizer = AutoTokenizer.from_pretrained(repo)

✨ Features

This model has the following features:

• Developers: Taeeon Park, Gihong Lee
• Dataset: Self - created using the DPO medical dataset (utilizing the AI - hub dataset)
• Training Method: DPO
• Company: MoAData

📄 License

This project is licensed under the Apache 2.0 license.