Datasetpublic

rodriguescarson/adaption-math-step-verification

Math Solution Step Verification Judge whether a step in a worked maths solution is correct, from human step-level labels. Rows 1,000 Domain mathematics Format data.parquet, one row per example Licence mit Built for supervised fine-tuning (SFT) experiments on Adaption AutoScientist Columns Column Description prompt The prompt (user turn) as uploaded. completion The target response as uploaded. enhanced_prompt Empty in this… See the full description on the dataset page: https://huggingface.co/datasets/rodriguescarson/adaption-math-step-verification.

sourceHugging Facemitupdated 11d agoView on Hugging Face

0likes79downloads

Dataset Card

Math Solution Step Verification

Judge whether a step in a worked maths solution is correct, from human step-level labels.


Rows	1,000
Domain	mathematics
Format	`data.parquet`, one row per example
Licence	mit
Built for	supervised fine-tuning (SFT) experiments on Adaption AutoScientist

Columns

Column	Description
`prompt`	The prompt (user turn) as uploaded.
`completion`	The target response as uploaded.
`enhanced_prompt`	Empty in this dataset.
`enhanced_completion`	Response after Adaption processing (rewrite or augmentation).
`reasoning_trace`	Step-by-step reasoning trace attached by Adaption's Adaptive Data.

How it was built

Built from PRM800K (via the TRL-deduplicated copy); the explained variant adds a written justification.

Sources and licence

—PRM800K (MIT)
—HuggingFaceH4/prm800k-trl-dedup (MIT)

Notes

—Columns enhanced_prompt are empty in this dataset (the Adaption export reserves them for rewritten text).

Loading

python

from datasets import load_dataset
ds = load_dataset("rodriguescarson/adaption-math-step-verification", split="train")

python

import pandas as pd
df = pd.read_parquet("hf://datasets/rodriguescarson/adaption-math-step-verification/data.parquet")

Published by Carson Rodrigues (Hugging Face, Kaggle).