Skip to content

Dataset · FreedomIntelligence

Medical reasoning data for SFT

FreedomIntelligence/medical-o1-reasoning-SFT

Medical tasks with chain-of-thought reasoning in English and Chinese for SFT and reasoning model initialization.

What's inside

The dataset contains medical SFT data and a mixture of medical and general instructions; the main file is medical_o1_sft.json, and medical_o1_sft_mix.json is also available. According to the card, it features distilled data from Deepseek-R1 based on verifiable medical tasks with chain-of-thought reasoning. The languages are English and Chinese, with a volume of 10k to 100k examples. The format is JSON.

What you can do with it

  • Fine-tune a medical LLM or LoRA on SFT with reasoning
  • Initialize a model with Deepseek-R1 chain-of-thought reasoning for medical tasks
  • Evaluate a QA system on medical questions
Tasks
Question answering, Text generation
Languages
English, Chinese
Size
10K–100K rows
Downloads this month
20.8K
Likes
1.2K

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("FreedomIntelligence/medical-o1-reasoning-SFT")