README.md

metadata

license: other
license_name: open-data-attribution-training-disclosure-license-odatl-1.0
license_link: LICENSE
language:
  - en
tags:
  - math

Math Teaching Dataset - 100MB / 735k Rows

A large math dataset designed for training and fine-tuning AI models with step-by-step reasoning.

Files

• math_dataset_100MB.csv - Full dataset, 100.33 MB, 735,000 rows
• math_dataset_100MB.csv.gz - Compressed version, 18.95 MB
• splits/math_part_1.csv to math_part_10.csv - 10 x ~10MB chunks

Dataset Stats

• Total rows: 735,000
• Size: 100.33 MB raw, 18.95 MB gzipped
• Format: CSV, UTF-8, with header

Topic Distribution

Topic	Count	Description
arithmetic	184,304	Addition, subtraction, multiplication, division, order of operations
algebra	91,880	Linear equations and quadratics
calculus	91,675	Basic derivatives and integrals
geometry	46,373	Area and perimeter of rectangles, circles, triangles
exponents	46,303	Powers
word_problems	45,929	Applied addition problems
statistics	45,920	Mean / average
percentages	45,857	Percentage calculations
number_theory	45,789	GCD and LCM
roots	45,567	Square roots
fractions	45,403	Fraction addition, subtraction, multiplication

Difficulty

• easy - single-step problems
• medium - multi-step problems

Schema

Column	Type	Description
id	int	Unique row ID
topic	string	High-level topic
difficulty	string	easy or medium
question	string	The math problem
answer	string	Final answer
reasoning	string	Step-by-step explanation
category	string	Specific sub-type of problem

Example Rows

"What is 1900 / 5?" -> Answer: 380 Reasoning: "Divide: 1900 ÷ 5 = 380 because 5*380=1900."

"Solve for x: 12x + 15 = 87" -> Answer: 6 Reasoning: "Subtract 15 from both sides: 12x = 72. Then divide by 12: x = 6."

"What is 3/4 + 2/5?" -> Answer: "23/20" Reasoning: "Common denominator 20. 3/4=15/20, 2/5=8/20. Sum = 23/20."

"What is the derivative of 7x^5 with respect to x?" -> Answer: "35x^4" Reasoning: "Power rule: d/dx [x^n] = nx^(n-1). So 75*x^4 = 35x^4."

Intended Use

Built for instruction tuning and chain-of-thought training.

Recommended training format: Input: {question} Target: {reasoning} Answer: {answer}

Works for:

• Fine-tuning language models for math reasoning
• Teaching step-by-step problem solving
• Evaluating math capabilities
• Data augmentation alongside other math datasets

License

open-data-attribution-training-disclosure-license-odatl-1.0