Yusiko/little_dataset_Azerbaijani

🇦🇿 Azerbaijani Instruction Dataset

📚 About this dataset This dataset contains 400 Azerbaijani-language instruction–response pairs, created for fine-tuning conversational and educational AI models. It follows the Alpaca-style format with "input" and "output" fields and focuses on clarity, accuracy, and linguistic richness.

🧩 Structure input → The user’s instruction or question (in Azerbaijani) output → The correct and natural Azerbaijani response

Each topic includes 100 high-quality examples.

🧠 Topics Included

1. 1.🏛 Tarix – Historical events, ancient states, cultural evolution
2. 2.➗ Riyazi Analiz – Mathematical limits, derivatives, integrals, and theory
3. 3.🤖 Mexatronika və Robototexnika – Sensors, actuators, PID control, kinematics
4. 4.💰 İqtisadiyyatın Əsasları – Supply & demand, inflation, market structures

⚙️ Format Examples {"input": "Atropatena dövləti nə vaxt yaranmışdır?", "output": "Atropatena M.Ö. 4-cü əsrdə yaranmış qədim Azərbaycan dövlətidir."}

{"input": "f(x)=x² funksiyasının törəməsini tapın.", "output": "f'(x)=2x."}

🧾 License MIT 🆓 Open for educational and research use.

💡 Notes

• —100% Azerbaijani language 🇦🇿
• —Clean, grammar-checked, and fine-tuning ready
• —Ideal for instruction-tuned little LLMs and dialogue-based models

⭐️ “Teaching AI to speak Azerbaijani — one dataset at a time.”