🚀 WiroAI/wiroai-turkish-llm-8b
WiroAI/wiroai-turkish-llm-8b is a powerful language model that offers enhanced support for the Turkish language and culture, providing high - quality language processing capabilities.
🚀 Quick Start
This robust language model offers more support for the Turkish language and culture. It has been fine - tuned with high - quality Turkish instructions and is built on a cutting - edge architecture, making it a great choice for various Turkish language tasks.
✨ Features
- Fine - Tuned with High - Quality Data: Fine - tuned using over 500,000 high - quality Turkish instructions.
- Advanced Fine - Tuning Method: The LoRA method was used for fine - tuning without quantization.
- Cultural Adaptation: Adapted to Turkish culture and local context.
- Innovative Architecture: Built on Google's state - of - the - art LLaMA architecture.
📝 Model Details
This model is the Turkish - speaking member of Meta's innovative LLaMA model family. It has been trained via Supervised Fine - Tuning (SFT) on carefully selected high - quality Turkish instructions, and shows excellent performance in Turkish language processing tasks.
🔧 Technical Details
Property |
Details |
Architecture |
Decoder - only transformer |
Base Model |
Meta LLaMA 3.1 8B |
Training Data |
500,000+ specially selected Turkish instructions |
Language Support |
Turkish (with comprehensive local context understanding) and other common languages |
💻 Usage Examples
Basic Usage
import transformers
import torch
model_id = "WiroAI/wiroai-turkish-llm-8b"
pipeline = transformers.pipeline(
"text-generation",
model=model_id,
model_kwargs={"torch_dtype": torch.bfloat16},
device_map="auto",
)
pipeline.model.eval()
messages = [
{"role": "system", "content": "Sen Wiro AI tarafından eğitilmiş Türkçe konuşan bir dil modelisin."},
{"role": "user", "content": "Bana İstanbul ile alakalı bir sosyal medya postu hazırlar mısın?"
},
]
terminators = [
pipeline.tokenizer.eos_token_id,
pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]
outputs = pipeline(
messages,
max_new_tokens=512,
eos_token_id=terminators,
do_sample=True,
temperature=0.9,
)
print(outputs[0]["generated_text"][-1]['content'])
Advanced Usage
İstanbul'un büyüsüne kapılın! :city_sunset:
Halk arasında "dünyanın masalı şehri" olarak bilinen İstanbul, her köşesinde tarih, kültür ve modern yaşamın bir araya geldiği eşsiz bir şehir.
Yüzyıllardır farklı medeniyetlerin izlerini taşıyan İstanbul, tarihi mekanlarından, müzelerinden, çarşılarından ve restoranlarından oluşan zengin kültürel mirasa sahiptir.
Boğaz'ın eşsiz manzarasında tekne turu yapmak, Topkapı Sarayı'nı ziyaret etmek, Grand Bazaar'da alışveriş yapmak, Mısır Çarşısı'nın canlı atmosferinde kaybolmak, Galata Kulesi'nden muhteşem bir manzara deneyimlemek veya Beyoğlu'nun hareketli sokaklarında yürüyüş yapmak İstanbul'da unutulmaz anılar yaratmak için fırsatlar sunar.
İstanbul'un büyülü atmosferini kendiniz yaşamak için hemen planınızı yapın! :flag-tr: #İstanbul #Türkiye #Seyahat #Tarih #Kültür #Gezi
💡 Use Cases
- Text Generation and Editing
- Question Answering
- Summarization
- Analysis and Reasoning
- Content Transformation
- Turkish Natural Language Processing Tasks
- Turkish Culture
🚀 Advantages
- Local Understanding: Capable of understanding Turkish culture, idioms, and current events.
- Resource Efficiency: Can operate effectively even with limited hardware resources.
- Flexible Deployment: Can be used on desktops, laptops, or custom cloud infrastructure.
- Open Model: Has a transparent and customizable architecture.
📈 Performance and Limitations
While the model performs well in Turkish language tasks, users should note the following:
⚠️ Important Note
- Use clear and structured instructions for optimal results.
- Verify model outputs for critical applications.
- Evaluate resource requirements before deployment.
- Be aware that the benchmarks below are presented under certain conditions, and the results can be replicated. The choices of conditions are explained below the table.
Benchmark Scores
Models |
MMLU TR |
TruthfulQA TR |
ARC TR |
HellaSwag TR |
GSM8K TR |
WinoGrande TR |
Average |
WiroAI/wiroai-turkish-llm-9b |
59.8 |
49.9 |
53.7 |
57.0 |
66.8 |
60.6 |
58.0 |
selimc/OrpoGemma-2-9B-TR |
53.0 |
54.3 |
52.4 |
52.0 |
64.8 |
58.9 |
55.9 |
Metin/Gemma-2-9b-it-TR-DPO-V1 |
51.3 |
54.7 |
52.6 |
51.2 |
67.1 |
55.2 |
55.4 |
CohereForAI/aya-expanse-8b |
52.3 |
52.8 |
49.3 |
56.7 |
61.3 |
59.2 |
55.3 |
ytu-ce-cosmos/Turkish-Llama-8b-DPO-v0.1 |
52.0 |
57.6 |
51.0 |
53.0 |
59.8 |
58.0 |
55.2 |
google/gemma-2-9b-it |
51.8 |
53.0 |
52.2 |
51.5 |
63.0 |
56.2 |
54.6 |
Eurdem/Defne-llama3.1-8B |
52.9 |
51.2 |
47.1 |
51.6 |
59.9 |
57.5 |
53.4 |
WiroAI/wiroai-turkish-llm-8b |
52.4 |
49.5 |
50.1 |
54 |
57.5 |
57.0 |
53.4 |
meta-llama/Meta-Llama-3-8B-Instruct |
52.2 |
49.2 |
44.2 |
49.2 |
56.0 |
56.7 |
51.3 |
Models Benchmarks are tested with
lm_eval --model_args pretrained=<model_path> --tasks mmlu_tr_v0.2,arc_tr-v0.2,gsm8k_tr-v0.2,hellaswag_tr-v0.2,truthfulqa_v0.2,winogrande_tr-v0.2
Please see https://github.com/malhajar17/lm-evaluation-harness_turkish and note that we move forward with default language inference which is the same approach in OpenLLMLeaderboard v2.0
📄 License
This model is provided under the apache 2.0 license. Please review and accept the license terms before use.
📫 Contact and Support
For questions, suggestions, and feedback, please open an issue on HuggingFace or contact us directly from our website.
Citation
@article{WiroAI,
title={WiroAI/wiroai-turkish-llm-8b},
author={Abdullah Bezir, Furkan Burhan Türkay, Cengiz Asmazoğlu},
year={2024},
url={https://huggingface.co/WiroAI/wiroai-turkish-llm-8b}
}