Alpaca GPT-4 Data Zh: Chinese Instruction-Tuning Dataset for LLMs

A dataset for instruction tuning with GPT-4, created by the team referenced in the citation. The dataset page was last updated on 2023-05-03. It is intended for research use only and is licensed under CC BY NC 4.0.

Use Cases

• Instruction tuning of Chinese language models based on GPT-4 generated examples.
• Benchmarking model performance on Chinese instruction-following tasks.
• Research into cross-lingual transfer learning for instruction-tuned models.

Strengths

• Dataset is explicitly designed for the research task of instruction tuning with GPT-4.
• License and intended use (research, non-commercial) are clearly stated in the description.

Limitations

• Description metadata is limited; actual data quality, size, and column structure require manual inspection after download.
• Last updated 2023-05-03; freshness should be verified for current model training practices.

Provenance

Source

llm-wizard on Hugging Face, based on work cited in the article 'Instruction Tuning with GPT-4'.

Collection Method

Likely contains instruction-response pairs generated or processed using GPT-4.

Time Range

Creation date not specified; dataset page updated in May 2023.

Freshness

Last updated 2023-05-03 17:56:55.

Geography

Data is in Chinese (Zh), but specific geographic origin is not stated.