Skip to content

Dataset · Mxode

Chinese instruction SFT corpus

Mxode/Chinese-Instruct

1 million to 10 million Chinese instruction-response pairs for SFT, including DeepSeek-R1 distillation.

What's inside

Text corpus of Chinese instructions and responses for SFT. Collected from several open sets: filtered DeepSeek-R1 distillation, reasoning data, and paraphrased Firefly. The card lists only final answers without chains of thought; volume ranges from 1 million to 10 million examples.

What you can do with it

  • SFT of Chinese LLM
  • LoRA for instructions
  • Checking answer quality without CoT
Tasks
Text generation, Question answering
Languages
Chinese
Size
1M–10M rows
Downloads this month
1.1K
Likes
157

How to load it

from datasets import load_dataset

ds = load_dataset("Mxode/Chinese-Instruct")