PAWSX

NAME = PAWSX
DATASET_PATH = google-research-datasets/paws-x
SAMPLE_SPLIT = test
FEWSHOT_SPLIT = validation
RESPONSE_TYPE = COMPLETION
METRICS = [AccuracyCompletion]
SUBJECTS = ['en', 'de']
LANGUAGE = {'en': <Language.ENG: 'English'>, 'de': <Language.DEU: 'German'>}

• Module: eval_framework.tasks.benchmarks.pawsx

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

• Link to dataset: https://huggingface.co/datasets/google-research-datasets/paws-x

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 "PAWSX".