baban/QwenTranslate_English_Ukrainian
baban/QwenTranslate_English_Ukrainian is a 3.1 billion parameter language model fine-tuned from Qwen/Qwen2.5-3B-Instruct. This model specializes in English-Ukrainian machine translation, leveraging its base architecture for language understanding and generation. It is optimized for translation tasks between these two specific languages, building upon its pre-trained capabilities.
Loading preview...
Model Overview
baban/QwenTranslate_English_Ukrainian is a 3.1 billion parameter model developed by baban, fine-tuned specifically for English-Ukrainian machine translation. It is based on the Qwen/Qwen2.5-3B-Instruct architecture, inheriting its foundational language processing capabilities.
Key Capabilities
- English-Ukrainian Translation: Specialized in translating text between English and Ukrainian.
- Fine-tuned Performance: Achieved a loss of 0.8353 on its evaluation set, indicating its performance on the translation task.
- Base Model: Leverages the robust Qwen2.5-3B-Instruct model as its foundation.
Training Details
The model was trained with a learning rate of 5e-05, a total batch size of 1024, and for 3 epochs. It utilized an AdamW optimizer and an inverse_sqrt learning rate scheduler.
Good For
- Applications requiring direct English to Ukrainian or Ukrainian to English text translation.
- Developers looking for a pre-trained model focused on this specific language pair, built on a Qwen base.
Limitations
The model's primary focus is on English-Ukrainian translation. Its performance on other language pairs or general-purpose tasks is not specified and may not be optimal.