> ML_DATASET // GSM8K-MATH-REASONING_v1.0
GSM8K (Grade School Math 8K)
OpenAI (Cobbe et al.) · Mathematical Reasoning & Code Generation · 8,500 grade school math word problems with step-by-step solutions
Mathematical Reasoning & Code GenerationMIT8,500 grade school math word problems with step-by-step solutionsopen
Dataset Profile & Characteristics
Label Type:Chain-of-thought rationale and final numerical answer
Languages:en
License Tier:permissive-open-source
Modalities:text
Intended Use
- Evaluating chain-of-thought multi-step arithmetic reasoning in LLMs
Prohibited / Discouraged Use
- Measuring high-level advanced mathematical or theorem-proving ability
Bias, Leakage & Privacy Risk Analysis
Privacy / Sensitive Data Risks:
Synthetic grade school math problems; zero personal data.
Known Bias:
Standard elementary arithmetic logic and linguistic problem construction.
Known Benchmark Leakage:
Severe benchmark contamination in pretraining corpora of frontier LLMs.