callofthenight1/gaokao-sft-chinese-balanced

--- license: mit task_categories: - text-generation - question-answering language: - zh pretty_name: Gaokao SFT Chinese Balanced size_categories: - 1K<n<10K --- # Gaokao SFT Chinese Balanced This dataset is a cleaned SFT-style Chinese exam dataset prepared from multiple public Hugging Face sources. ## Composition - Total samples: 1895 - Train samples: 1853 - Validation samples: 42 ## Fields Each row contains: - `id` - `lang` - `subject` - `source` - `instruction` - `input` - `output` - `messages` ## Cleaning Notes - Ordinary Markdown markers were removed. - Non-essential LaTeX commands were simplified into plain readable text. - Math expressions were preserved in readable plain-text form where possible. - Samples were normalized into SFT instruction/output format.