Congratulations
I'm an individual who happen to work in the ML space and living in Bangkok. Thai is not my native language, but I speak and write Thai quite fluent.
From what I can see is, you didn't use the open-assistant Dataset, right? It contains quite a few Thai examples that you could extract and use for your fine tuning.
I am really happy to see these efforts for openThaiGPT being made!
Having an open source fine tuned Thai transformer model will surely benefit the start up scene in Thailand.
I will add create a foreign language section in my model list and add this model soon.
Thank you for reaching out and expressing your appreciation on OpenThaiGPT 0.1.0-beta released!.
You are right, the open-assistant Dataset was not used for training the current version of OpenThaiGPT.
It will be included in the future version.
Additionally, I appreciate your intention to create a foreign language section in your model list and include OpenThaiGPT.
Thank you for your support and for taking the time to reach out. If you have any further questions or suggestions, please feel free to let me know.