Back to datasets
Dataset assetOpen Source CommunityMachine LearningMathematics Competition
martim00/math_aime_2023
The dataset contains two features: instruction and output, both of type string. It is split into a training set with 682 samples and a test set with 293 samples. The total size of the dataset is 2,205,691 bytes, and the download size is 1,042,346 bytes. The dataset configuration includes a default configuration, specifying the paths for the training and test data.
Source
hugging_face
Created
Nov 28, 2025
Updated
Apr 5, 2024
Signals
78 views
Availability
Linked source ready
Overview
Dataset description and usage context
Dataset Overview
Dataset Features
- instruction: data type is string.
- output: data type is string.
Dataset Splits
- Training set (train): contains 682 samples, total size 1,553,019 bytes.
- Test set (test): contains 293 samples, total size 652,672 bytes.
Dataset Size
- Download size: 1,042,346 bytes.
- Total dataset size: 2,205,691 bytes.
Data File Configuration
- Configuration name: default
- Training file path: data/train-*
- Test file path: data/test-*
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.