MATH (Hendrycks) — LLM 评估与基准测试 Dataset

12,500 competition mathematics problems across 7 subjects (algebra, calculus, number theory, etc.) with step-by-step solutions. The key benchmark for evaluating mathematical reasoning in LLMs — requires multi-step problem-solving and symbolic manipulation.

Dataset Details

from datasets import load_dataset
ds = load_dataset("hendrycks/competition_math")

相关数据集

• MMLU (Massive Multitask Language Understanding) — Benchmarking general knowledge - never train on it
• NuminaMath — Chain-of-thought math fine-tuning up to olympiad level
• HumanEval — Benchmarking Python code generation - never train on it
• Winogrande — Benchmarking commonsense reasoning

常见问题

← 全部数据集 | Fine-Tuning Guide