Ba2han/Karga_EN-TR-8B-A1B

This is an EN<>TR translation model fine-tuned from the LiquidAI/LFM2-8B-A1B model. While convenient, it can occasionally make simple spelling errors.

The system message should be "Task: Translation."

Suggested settings:

top_k = 10,
top_p = 0.95,
temperature = 0.2

The message above was translated with this model.

---

Bu, LiquidAI/LFM2-8B-A1B modelinden fine-tune edilmiş bir EN<>TR çeviri modelidir. Kullanışlı olmasına rağmen bazen basit imla hataları yapabilir. Sistem mesajı "Task: Translation." olmalıdır.

Önerilen ayarlar:

top_k = 10,
top_p = 0.95,
temperature = 0.2

---

More training details:

Dataset: Private dataset, ~100M tokens Full fine-tuning via unsloth ~7hrs x A100 40GB

Citations

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}