metadata

language:
  - tg
  - en
  - ru
license: mit
tags:
  - gpt3
  - transformers
  - mgpt

🇹🇯 Tajik mGPT 1.3B

Language model for Tajik. Model has 1.3B parameters as you can guess from it's name.

Tajik belongs to Indo-European language family. It's a very fluid language with approximately 8.2 million speakers. Here are some facts about it:

It is a variety of Persian spoken in Tajikistan and parts of Afghanistan.
Unlike Persian in Iran, Tajik is written in the Cyrillic script due to Soviet influence.
The Pamiri languages of Tajikistans Gorno-Badakhshan Autonomous Region are closely related to Tajik.

Dataset

TBD

TBD

Try different generation strategies to reach better results.

TBD

Model will be improved over time. Stay tuned!