SocialIQAMC_OLMES
NAME = SocialIQAMC_OLMES
DATASET_PATH = allenai/social_i_qa
SAMPLE_SPLIT = train
FEWSHOT_SPLIT = train
RESPONSE_TYPE = LOGLIKELIHOODS
METRICS = [AccuracyLoglikelihood, AccuracyNormLoglikelihood, BitsPerByteLoglikelihood]
SUBJECTS = ['no_subject']
LANGUAGE = <Language.ENG: 'English'>
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 "SocialIQAMC_OLMES".
SocialIQAMC_OLMESΒΆ
Module:
eval_framework.tasks.benchmarks.social_iqaFile: src/eval_framework/tasks/benchmarks/social_iqa.py | View on GitHub
Link to dataset: https://huggingface.co/datasets/allenai/social_i_qa
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 "SocialIQAMC_OLMES".