AIME2024

NAME = AIME2024
DATASET_PATH = HuggingFaceH4/aime_2024
SAMPLE_SPLIT = train
FEWSHOT_SPLIT = train
RESPONSE_TYPE = COMPLETION
METRICS = [MathReasoningCompletion, LanguageRawConsistencyChecker]
SUBJECTS = ['no_subject']
LANGUAGE = <Language.ENG: 'English'>

• Module: eval_framework.tasks.benchmarks.math_reasoning

• File: src/eval_framework/tasks/benchmarks/math_reasoning.py | View on GitHub

• Link to dataset: https://huggingface.co/datasets/HuggingFaceH4/aime_2024

More detailed documentation, with prompt examples and ground truth completions, can be generated with uv run -m eval_framework.utils.generate_task_docs --add-prompt-examples --only-tasks "AIME2024".