|
--- |
|
language: |
|
- ro |
|
- en |
|
- ru |
|
license: mit |
|
tags: |
|
- gpt3 |
|
- transformers |
|
- mgpt |
|
--- |
|
# 🇷🇴 Romanian mGPT 1.3B |
|
|
|
Language model for Romanian. Model has 1.3B parameters as you can guess from it's name. |
|
|
|
Romanian belongs to Indo-European language family. It's a very lyrical language with approximately 24 million speakers. Here are some facts about it: |
|
|
|
1. It is a Romance language, closely related to Italian, French, Spanish, Portuguese, and Catalan. |
|
2. It retains several Latin characteristics, making it unique among the Romance languages. |
|
3. While primarily spoken in Romania and Moldova, there are also Romanian speakers in neighboring countries and diaspora communities worldwide. |
|
|
|
## Technical details |
|
|
|
It's one of the models derived from the base [mGPT-XL (1.3B)](https://huggingface.co/ai-forever/mGPT) model (see the list below) which was originally trained on the 61 languages from 25 language families using Wikipedia and C4 corpus. |
|
|
|
We've found additional data for 23 languages most of which are considered as minor and decided to further tune the base model. **Romanian mGPT 1.3B** was trained for another 5000 steps with batch_size=4 and context window of **2048** tokens on 1 A100. |
|
|
|
Final perplexity for this model on validation is **3.44**. |
|
|
|
_Chart of the training loss and perplexity:_ |
|
|
|
 |
|
|
|
## Other mGPT-1.3B models |
|
|
|
- [mGPT-1.3B-armenian](https://huggingface.co/ai-forever/mGPT-1.3B-armenian) |
|
- [mGPT-1.3B-azerbaijan](https://huggingface.co/ai-forever/mGPT-1.3B-azerbaijan) |
|
- [mGPT-1.3B-bashkir](https://huggingface.co/ai-forever/mGPT-1.3B-bashkir) |
|
- [mGPT-1.3B-belorussian](https://huggingface.co/ai-forever/mGPT-1.3B-belorussian) |
|
- [mGPT-1.3B-bulgarian](https://huggingface.co/ai-forever/mGPT-1.3B-bulgarian) |
|
- [mGPT-1.3B-buryat](https://huggingface.co/ai-forever/mGPT-1.3B-buryat) |
|
- [mGPT-1.3B-chuvash](https://huggingface.co/ai-forever/mGPT-1.3B-chuvash) |
|
- [mGPT-1.3B-georgian](https://huggingface.co/ai-forever/mGPT-1.3B-georgian) |
|
- [mGPT-1.3B-kalmyk](https://huggingface.co/ai-forever/mGPT-1.3B-kalmyk) |
|
- [mGPT-1.3B-kazakh](https://huggingface.co/ai-forever/mGPT-1.3B-kazakh) |
|
- [mGPT-1.3B-kirgiz](https://huggingface.co/ai-forever/mGPT-1.3B-kirgiz) |
|
- [mGPT-1.3B-mari](https://huggingface.co/ai-forever/mGPT-1.3B-mari) |
|
- [mGPT-1.3B-mongol](https://huggingface.co/ai-forever/mGPT-1.3B-mongol) |
|
- [mGPT-1.3B-ossetian](https://huggingface.co/ai-forever/mGPT-1.3B-ossetian) |
|
- [mGPT-1.3B-persian](https://huggingface.co/ai-forever/mGPT-1.3B-persian) |
|
- [mGPT-1.3B-tajik](https://huggingface.co/ai-forever/mGPT-1.3B-tajik) |
|
- [mGPT-1.3B-tatar](https://huggingface.co/ai-forever/mGPT-1.3B-tatar) |
|
- [mGPT-1.3B-turkmen](https://huggingface.co/ai-forever/mGPT-1.3B-turkmen) |
|
- [mGPT-1.3B-tuvan](https://huggingface.co/ai-forever/mGPT-1.3B-tuvan) |
|
- [mGPT-1.3B-ukranian](https://huggingface.co/ai-forever/mGPT-1.3B-ukranian) |
|
- [mGPT-1.3B-uzbek](https://huggingface.co/ai-forever/mGPT-1.3B-uzbek) |
|
- [mGPT-1.3B-yakut](https://huggingface.co/ai-forever/mGPT-1.3B-yakut) |
|
|
|
## Feedback |
|
|
|
If you'll found a bug of have additional data to train model on your language — please, give us feedback. |
|
|
|
Model will be improved over time. Stay tuned! |
|
|
