Dataset Details

• Modality: Text
• Format: CSV
• Size: 100K - 1M rows
• Total Rows: 248,836
• License: MIT
• Libraries Supported: datasets, pandas, croissant

Structure

Each row in the dataset includes:

• question: The query posed in the dataset.
• answer: The correct response.
• category: The domain of the question (e.g., math, science).
• src: The source of the question.
• id: A unique identifier for each entry.
• chain_of_thoughts: Step-by-step reasoning steps leading to the answer.
• parsed_answer: Parsed answer from the chain of thoughts.
• parsed_answer_correctness: Boolean indicating if the parsed answer is correct.

Access & Download

The dataset can be accessed via the Hugging Face datasets library:

from datasets import load_dataset

dataset = load_dataset("UW-Madison-Lee-Lab/MMLU-Pro-CoT-Eval")