Datasetpublic

JERRYPAN617/GSM8K-MinimalAssign

GSM8K-MinimalAssign GSM8K-MinimalAssign is a reformatted version of the OpenAI's GSM8K dataset (openai/gsm8k) where each problem and its solution are rewritten in Form C (minimal assignment chain format): pure assignment chains with one variable assignment per line, showing clear dependency relationships. Dataset Structure Each example contains: original_question: The original GSM8K problem original_answer: The original solution rewritten_form_c: The reformatted… See the full description on the dataset page: https://huggingface.co/datasets/JERRYPAN617/GSM8K-MinimalAssign.

sourceHugging Faceapache-2.0updated 10mo agoView on Hugging Face

0likes17downloads

Dataset Card

GSM8K-MinimalAssign

GSM8K-MinimalAssign is a reformatted version of the OpenAI's GSM8K dataset (openai/gsm8k) where each problem and its solution are rewritten in Form C (minimal assignment chain format): pure assignment chains with one variable assignment per line, showing clear dependency relationships.

Dataset Structure

Each example contains:

—original_question: The original GSM8K problem
—original_answer: The original solution
—rewritten_form_c: The reformatted solution in Form C format
—index: Sample index

Example

{
  "original_question": "Natalia sold clips to 48 of her friends in April, and then she sold half as many clips in May. How many clips did Natalia sell altogether in April and May?",
  "original_answer": "Natalia sold 48/2 = <<48/2=24>>24 clips in May.\nNatalia sold 48+24 = <<48+24=72>>72 clips altogether in April and May.\n#### 72",
  "rewritten_form_c": "April sales = 48  \nMay sales = April sales ÷ 2 = 24  \nTotal sales = April sales + May sales = 72  \n#### 72",
  "index": 0
}

Usage

from datasets import load_dataset
dataset = load_dataset("JERRYPAN617/GSM8K-MinimalAssign")