Mistral Ko 7B V0.1

🚀 Mistral-ko-7B-v0.1

This model is an old experimental one. It is not recommended for practical use. Mistral-ko-7B-v0.1 is a model that applies a tokenizer optimized for Korean to Mistral. It was trained for 2 epochs on a somewhat formed model with raw data using the dataset used in Sinatra.

🚀 Quick Start

This section is not provided in the original document, so it is skipped.

✨ Features

Model Details

Description

Mistral-ko-7B-v0.1 is a model that applies a tokenizer optimized for Korean to Mistral. It was trained for 2 epochs on a somewhat formed model with raw data using the dataset used in Sinatra.

-- Further Description After Evaluation --

Comment

The tokenizer was created based on the Korean version of Llama 2 by @beomi.

Thank you to @jin05102518 for providing the base model.

Follow me on twitter: https://twitter.com/stablefluffy

Consider Support me making these model alone: https://www.buymeacoffee.com/mwell or with Runpod Credit Gift 💕

Contact me on Telegram: https://t.me/AlzarTakkarsen

📄 License

The model is licensed under CC BY-NC 4.0.

⚠️ Important Note

This model is an old experimental one. It is not recommended for practical use.

💡 Usage Tip

If you want to follow the author, you can click the provided Twitter link. If you want to support the author, you can use the provided links to buy a coffee or give Runpod Credit Gift. You can also contact the author via the Telegram link.