AidanBench

NAME = AidanBench
DATASET_PATH = Aleph-Alpha-Research/aidanbench
SAMPLE_SPLIT = train
FEWSHOT_SPLIT = train
RESPONSE_TYPE = COMPLETION
METRICS = [AidanBenchMetric]
SUBJECTS = ['no_subject']
LANGUAGE = {'no_subject': <Language.ENG: 'English'>}

• Module: eval_framework.tasks.benchmarks.aidanbench

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

• Link to dataset: https://huggingface.co/datasets/Aleph-Alpha-Research/aidanbench

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 "AidanBench".