Azerbaijani Instruction Dataset for Language Model Fine-Tuning

45 hand-authored Azerbaijani instruction and response pairs for supervised fine-tuning of language models. The dataset is part of an open Azerbaijani LLM stack and was created by author kamaalg. It was last updated on July 7, 2026.

Use Cases

• Supervised fine-tuning of Azerbaijani language models based on instruction-response pairs.
• Benchmarking instruction-following capabilities of Azerbaijani LLMs based on the 16 task types.
• Creating synthetic training data for Azerbaijani NLP tasks based on the provided seed examples.

Strengths

• 45 high-quality, hand-authored seed instruction-response pairs.
• Covers 16 distinct task types, suggesting diversity in instruction formats.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

huggingface

Collection Method

Hand-authored seed pairs.

Freshness

Last updated 2026-07-07 07:17:07; freshness should be verified.