Name: trillionlabs/rbridge-gemini-2.5-pro
Creator: trillionlabs
Published: 2025-12-02 06:50:46
License: 暂无描述

下载链接：

https://hf-mirror.com/datasets/trillionlabs/rbridge-gemini-2.5-pro

下载链接

链接失效反馈

官方服务：

资源简介：

--- dataset_info: - config_name: ChemBench features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: canary dtype: string - name: description dtype: string - name: examples list: - name: input dtype: string - name: target dtype: string - name: target_scores dtype: string - name: in_humansubset_w_tool dtype: bool - name: in_humansubset_wo_tool dtype: bool - name: input dtype: string - name: keywords sequence: string - name: metrics sequence: string - name: name dtype: string - name: output dtype: string - name: preferred_score dtype: string - name: subfield dtype: string - name: text dtype: string - name: uuid dtype: string splits: - name: test num_bytes: 3995683 num_examples: 815 download_size: 1864140 dataset_size: 3995683 - config_name: MolLangBench features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: canary dtype: string - name: edit_instructions dtype: string - name: edited_image struct: - name: bytes dtype: binary - name: path dtype: 'null' - name: edited_smiles dtype: string - name: image struct: - name: bytes dtype: binary - name: path dtype: 'null' - name: input dtype: string - name: note dtype: string - name: original_image struct: - name: bytes dtype: binary - name: path dtype: 'null' - name: original_smiles dtype: string - name: output dtype: string - name: result_1 dtype: string - name: result_2 dtype: string - name: smiles dtype: string - name: structure_description dtype: string - name: target_atoms dtype: string - name: task dtype: string - name: text dtype: string splits: - name: test num_bytes: 18512795 num_examples: 300 download_size: 17298124 dataset_size: 18512795 - config_name: MultiMedQA features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: choice_type dtype: string - name: cop dtype: int64 - name: data struct: - name: Context sequence: string - name: Correct Answer dtype: string - name: Correct Option dtype: string - name: Long Answer dtype: string - name: Options struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: D dtype: string - name: Question dtype: string - name: exp dtype: string - name: id dtype: string - name: input dtype: string - name: opa dtype: string - name: opb dtype: string - name: opc dtype: string - name: opd dtype: string - name: output dtype: string - name: question dtype: string - name: subject_name dtype: string - name: text dtype: string - name: topic_name dtype: string splits: - name: test num_bytes: 4501404 num_examples: 881 download_size: 2487110 dataset_size: 4501404 - config_name: agieval-math features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: input dtype: string - name: output dtype: string - name: query dtype: string splits: - name: test num_bytes: 2885807 num_examples: 1000 download_size: 1398082 dataset_size: 2885807 - config_name: aime25 features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: id dtype: string - name: input dtype: string - name: output dtype: string - name: problem dtype: string - name: text dtype: string splits: - name: test num_bytes: 119619 num_examples: 30 download_size: 88534 dataset_size: 119619 - config_name: arc-ARC-Challenge features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answerKey dtype: string - name: choices struct: - name: label sequence: string - name: text sequence: string - name: id dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 3953655 num_examples: 1172 download_size: 2091178 dataset_size: 3953655 - config_name: arc-ARC-Easy features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answerKey dtype: string - name: choices struct: - name: label sequence: string - name: text sequence: string - name: id dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 7462970 num_examples: 2376 download_size: 3908253 dataset_size: 7462970 - config_name: arena-hard-v0.1-gpt-4o-2024-08-06 features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: model dtype: string - name: output dtype: string splits: - name: test num_bytes: 5098865 num_examples: 500 download_size: 2797175 dataset_size: 5098865 - config_name: arena-hard-v2-gemini-2.5 features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: model dtype: string - name: output dtype: string splits: - name: test num_bytes: 11597429 num_examples: 750 download_size: 6489275 dataset_size: 11597429 - config_name: attributionbench features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: attribution_label dtype: string - name: citation_links sequence: string - name: claim dtype: string - name: claim_raw_string dtype: string - name: id dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: references sequence: string - name: response dtype: string - name: src_dataset dtype: string - name: text dtype: string - name: webpage_references sequence: 'null' splits: - name: test num_bytes: 19639947 num_examples: 1606 download_size: 9863221 dataset_size: 19639947 - config_name: bbh-boolean_expressions features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 336040 num_examples: 250 download_size: 103827 dataset_size: 336040 - config_name: bbh-causal_judgement features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 672661 num_examples: 185 download_size: 279505 dataset_size: 672661 - config_name: bbh-date_understanding features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 418205 num_examples: 250 download_size: 166662 dataset_size: 418205 - config_name: bbh-disambiguation_qa features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 524844 num_examples: 250 download_size: 181841 dataset_size: 524844 - config_name: bbh-dyck_languages features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 550477 num_examples: 250 download_size: 192577 dataset_size: 550477 - config_name: bbh-formal_fallacies features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 1270454 num_examples: 250 download_size: 435320 dataset_size: 1270454 - config_name: bbh-geometric_shapes features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 730980 num_examples: 250 download_size: 292147 dataset_size: 730980 - config_name: bbh-hyperbaton features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 489981 num_examples: 250 download_size: 171065 dataset_size: 489981 - config_name: bbh-logical_deduction_five_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 878413 num_examples: 250 download_size: 283956 dataset_size: 878413 - config_name: bbh-logical_deduction_seven_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 1045248 num_examples: 250 download_size: 339275 dataset_size: 1045248 - config_name: bbh-logical_deduction_three_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 580741 num_examples: 250 download_size: 176442 dataset_size: 580741 - config_name: bbh-movie_recommendation features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 843841 num_examples: 250 download_size: 408743 dataset_size: 843841 - config_name: bbh-multistep_arithmetic_two features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 410150 num_examples: 250 download_size: 149130 dataset_size: 410150 - config_name: bbh-navigate features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 470680 num_examples: 250 download_size: 162179 dataset_size: 470680 - config_name: bbh-object_counting features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 282951 num_examples: 250 download_size: 108262 dataset_size: 282951 - config_name: bbh-penguins_in_a_table features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 322459 num_examples: 146 download_size: 93840 dataset_size: 322459 - config_name: bbh-reasoning_about_colored_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 433556 num_examples: 250 download_size: 138410 dataset_size: 433556 - config_name: bbh-ruin_names features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 546616 num_examples: 250 download_size: 242320 dataset_size: 546616 - config_name: bbh-salient_translation_error_detection features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 974857 num_examples: 250 download_size: 291597 dataset_size: 974857 - config_name: bbh-snarks features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 451997 num_examples: 178 download_size: 214551 dataset_size: 451997 - config_name: bbh-sports_understanding features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 727987 num_examples: 248 download_size: 365995 dataset_size: 727987 - config_name: bbh-temporal_sequences features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 752754 num_examples: 250 download_size: 251965 dataset_size: 752754 - config_name: bbh-tracking_shuffled_objects_five_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 805465 num_examples: 250 download_size: 227521 dataset_size: 805465 - config_name: bbh-tracking_shuffled_objects_seven_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 983969 num_examples: 250 download_size: 290973 dataset_size: 983969 - config_name: bbh-tracking_shuffled_objects_three_objects features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 612414 num_examples: 250 download_size: 166319 dataset_size: 612414 - config_name: bbh-web_of_lies features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 434155 num_examples: 250 download_size: 140911 dataset_size: 434155 - config_name: bbh-word_sorting features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: input dtype: string - name: output dtype: string - name: target dtype: string splits: - name: test num_bytes: 593714 num_examples: 250 download_size: 296481 dataset_size: 593714 - config_name: cqa features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answerKey dtype: string - name: choices struct: - name: label sequence: string - name: text sequence: string - name: id dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: question_concept dtype: string splits: - name: test num_bytes: 23329003 num_examples: 9741 download_size: 12617010 dataset_size: 23329003 - config_name: csatqa features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: Category dtype: string - name: Human_Peformance dtype: float64 - name: context dtype: string - name: gold dtype: int64 - name: input dtype: string - name: option#1 dtype: string - name: option#2 dtype: string - name: option#3 dtype: string - name: option#4 dtype: string - name: option#5 dtype: string - name: output dtype: string - name: question dtype: string - name: question_number dtype: int64 - name: test_name dtype: string - name: text dtype: string splits: - name: test num_bytes: 19170173 num_examples: 912 download_size: 9588361 dataset_size: 19170173 - config_name: drop features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answers_spans struct: - name: spans sequence: string - name: types sequence: string - name: input dtype: string - name: output dtype: string - name: passage dtype: string - name: query_id dtype: string - name: question dtype: string - name: section_id dtype: string - name: text dtype: string splits: - name: test num_bytes: 51733097 num_examples: 8687 download_size: 23639510 dataset_size: 51733097 - config_name: global-mmlu-ja features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: country dtype: string - name: cultural_sensitivity_label dtype: string - name: culture dtype: string - name: input dtype: string - name: is_annotated dtype: bool - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: reference dtype: string - name: region dtype: string - name: required_knowledge dtype: string - name: sample_id dtype: string - name: subject dtype: string - name: subject_category dtype: string - name: time_sensitive dtype: string splits: - name: test num_bytes: 66564951 num_examples: 13944 download_size: 33994057 dataset_size: 66564951 - config_name: global-mmlu-ko features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: country dtype: string - name: cultural_sensitivity_label dtype: string - name: culture dtype: string - name: input dtype: string - name: is_annotated dtype: bool - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: reference dtype: string - name: region dtype: string - name: required_knowledge dtype: string - name: sample_id dtype: string - name: subject dtype: string - name: subject_category dtype: string - name: time_sensitive dtype: string splits: - name: test num_bytes: 61046246 num_examples: 13952 download_size: 32116787 dataset_size: 61046246 - config_name: gpqa features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: Canary String dtype: string - name: Correct Answer dtype: string - name: Expert Validator Accuracy dtype: float64 - name: Expert Validator Disagreement Category dtype: float64 - name: Expert Validator_EV_1 dtype: string - name: Expert Validator_EV_2 dtype: string - name: Explanation dtype: string - name: Explanation_NEV_1 dtype: string - name: Explanation_NEV_2 dtype: string - name: Explanation_NEV_3 dtype: string - name: Extra Revised Correct Answer dtype: string - name: Extra Revised Explanation dtype: string - name: Extra Revised Incorrect Answer 1 dtype: string - name: Extra Revised Incorrect Answer 2 dtype: string - name: Extra Revised Incorrect Answer 3 dtype: string - name: Extra Revised Question dtype: string - name: Feedback_EV_1 dtype: string - name: Feedback_EV_2 dtype: string - name: Feedback_NEV_1 dtype: string - name: Feedback_NEV_2 dtype: string - name: Feedback_NEV_3 dtype: string - name: High-level domain dtype: string - name: Incorrect Answer 1 dtype: string - name: Incorrect Answer 2 dtype: string - name: Incorrect Answer 3 dtype: string - name: Is First Validation_EV_1 dtype: bool - name: Is First Validation_EV_2 dtype: bool - name: Majority Non-Expert Vals Incorrect dtype: float64 - name: Manual Correctness Adjustment_EV_1 dtype: string - name: Manual Correctness Adjustment_EV_2 dtype: string - name: Manual Correctness Adjustment_NEV_1 dtype: string - name: Manual Correctness Adjustment_NEV_2 dtype: 'null' - name: Manual Correctness Adjustment_NEV_3 dtype: 'null' - name: Non-Expert Validator Accuracy dtype: float64 - name: Non-Expert Validator_NEV_1 dtype: string - name: Non-Expert Validator_NEV_2 dtype: string - name: Non-Expert Validator_NEV_3 dtype: string - name: Post hoc agreement_EV_1 dtype: string - name: Post hoc agreement_EV_2 dtype: string - name: Pre-Revision Correct Answer dtype: string - name: Pre-Revision Explanation dtype: string - name: Pre-Revision Incorrect Answer 1 dtype: string - name: Pre-Revision Incorrect Answer 2 dtype: string - name: Pre-Revision Incorrect Answer 3 dtype: string - name: Pre-Revision Question dtype: string - name: Probability Correct_EV_1 dtype: string - name: Probability Correct_EV_2 dtype: string - name: Probability Correct_NEV_1 dtype: string - name: Probability Correct_NEV_2 dtype: string - name: Probability Correct_NEV_3 dtype: string - name: Question dtype: string - name: Question Difficulty_EV_1 dtype: string - name: Question Difficulty_EV_2 dtype: string - name: Question Writer dtype: string - name: Record ID dtype: string - name: Revision Comments (from Question Writer) dtype: string - name: Self-reported question-writing time (minutes) dtype: float64 - name: Self-reported time (minutes)_EV_1 dtype: float64 - name: Self-reported time (minutes)_EV_2 dtype: float64 - name: Self-reported time (minutes)_NEV_1 dtype: float64 - name: Self-reported time (minutes)_NEV_2 dtype: float64 - name: Self-reported time (minutes)_NEV_3 dtype: float64 - name: Subdomain dtype: string - name: Sufficient Expertise?_EV_1 dtype: string - name: Sufficient Expertise?_EV_2 dtype: string - name: Understand the question?_EV_1 dtype: string - name: Understand the question?_EV_2 dtype: string - name: Validator Answered Correctly_EV_1 dtype: int64 - name: Validator Answered Correctly_EV_2 dtype: int64 - name: Validator Answered Correctly_NEV_1 dtype: int64 - name: Validator Answered Correctly_NEV_2 dtype: int64 - name: Validator Answered Correctly_NEV_3 dtype: float64 - name: Validator Revision Suggestion_EV_1 dtype: string - name: Validator Revision Suggestion_EV_2 dtype: string - name: Websites visited_NEV_1 dtype: string - name: Websites visited_NEV_2 dtype: string - name: Websites visited_NEV_3 dtype: string - name: Writer's Difficulty Estimate dtype: string - name: input dtype: string - name: output dtype: string splits: - name: test num_bytes: 5322001 num_examples: 448 download_size: 2869070 dataset_size: 5322001 - config_name: gsm8k features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: text dtype: string splits: - name: test num_bytes: 4460071 num_examples: 1319 download_size: 2291622 dataset_size: 4460071 - config_name: haerae-general_knowledge features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 419010 num_examples: 176 download_size: 236907 dataset_size: 419010 - config_name: haerae-history features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 462858 num_examples: 188 download_size: 265640 dataset_size: 462858 - config_name: haerae-loan_words features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 342333 num_examples: 169 download_size: 171776 dataset_size: 342333 - config_name: haerae-rare_words features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 873146 num_examples: 405 download_size: 422804 dataset_size: 873146 - config_name: haerae-reading_comprehension features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: 'null' - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 2351143 num_examples: 447 download_size: 1257144 dataset_size: 2351143 - config_name: haerae-standard_nomenclature features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: a dtype: string - name: answer dtype: string - name: b dtype: string - name: c dtype: string - name: d dtype: string - name: e dtype: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: split dtype: string splits: - name: test num_bytes: 329248 num_examples: 153 download_size: 165374 dataset_size: 329248 - config_name: hellaswag features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: activity_label dtype: string - name: ctx dtype: string - name: ctx_a dtype: string - name: ctx_b dtype: string - name: endings sequence: string - name: ind dtype: int64 - name: input dtype: string - name: label dtype: string - name: output dtype: string - name: source_id dtype: string - name: split dtype: string - name: split_type dtype: string - name: text dtype: string splits: - name: test num_bytes: 61032644 num_examples: 10042 download_size: 32943807 dataset_size: 61032644 - config_name: hle features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: answer_type dtype: string - name: author_name dtype: string - name: canary dtype: string - name: category dtype: string - name: id dtype: string - name: image dtype: string - name: image_preview dtype: 'null' - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: rationale dtype: string - name: rationale_image struct: - name: bytes dtype: binary - name: path dtype: 'null' - name: raw_subject dtype: string splits: - name: test num_bytes: 81207862 num_examples: 2158 download_size: 73403422 dataset_size: 81207862 - config_name: hrm8k features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: float64 - name: input dtype: string - name: original dtype: string - name: output dtype: string - name: question dtype: string - name: text dtype: string splits: - name: test num_bytes: 3912180 num_examples: 1319 download_size: 1874921 dataset_size: 3912180 - config_name: include-Albanian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1598766 num_examples: 551 download_size: 869505 dataset_size: 1598766 - config_name: include-Arabic features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1951446 num_examples: 552 download_size: 927489 dataset_size: 1951446 - config_name: include-Armenian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2536034 num_examples: 549 download_size: 1100550 dataset_size: 2536034 - config_name: include-Azerbaijani features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1690162 num_examples: 539 download_size: 866199 dataset_size: 1690162 - config_name: include-Basque features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1752763 num_examples: 500 download_size: 915919 dataset_size: 1752763 - config_name: include-Belarusian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2507106 num_examples: 550 download_size: 1111869 dataset_size: 2507106 - config_name: include-Bengali features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2176427 num_examples: 548 download_size: 810619 dataset_size: 2176427 - config_name: include-Bulgarian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: 'null' - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2385630 num_examples: 549 download_size: 1065330 dataset_size: 2385630 - config_name: include-Chinese features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1735002 num_examples: 545 download_size: 1056807 dataset_size: 1735002 - config_name: include-Croatian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: 'null' - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1503744 num_examples: 550 download_size: 890790 dataset_size: 1503744 - config_name: include-Dutch features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1820943 num_examples: 551 download_size: 986082 dataset_size: 1820943 - config_name: include-Estonian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 454304 num_examples: 224 download_size: 249177 dataset_size: 454304 - config_name: include-Finnish features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2293226 num_examples: 551 download_size: 1275375 dataset_size: 2293226 - config_name: include-French features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1157701 num_examples: 419 download_size: 632687 dataset_size: 1157701 - config_name: include-Georgian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2968205 num_examples: 500 download_size: 1005534 dataset_size: 2968205 - config_name: include-German features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 484617 num_examples: 138 download_size: 284525 dataset_size: 484617 - config_name: include-Greek features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 3110518 num_examples: 514 download_size: 1380450 dataset_size: 3110518 - config_name: include-Hebrew features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2262966 num_examples: 550 download_size: 1066116 dataset_size: 2262966 - config_name: include-Hindi features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 3002660 num_examples: 547 download_size: 1113819 dataset_size: 3002660 - config_name: include-Hungarian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: 'null' - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1538358 num_examples: 550 download_size: 852968 dataset_size: 1538358 - config_name: include-Indonesian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1713007 num_examples: 550 download_size: 872297 dataset_size: 1713007 - config_name: include-Italian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2281324 num_examples: 548 download_size: 1241256 dataset_size: 2281324 - config_name: include-Japanese features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2266537 num_examples: 501 download_size: 1202897 dataset_size: 2266537 - config_name: include-Kazakh features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1700882 num_examples: 500 download_size: 783273 dataset_size: 1700882 - config_name: include-Korean features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1909363 num_examples: 500 download_size: 951981 dataset_size: 1909363 - config_name: include-Lithuanian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1692160 num_examples: 534 download_size: 954835 dataset_size: 1692160 - config_name: include-Malay features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1525909 num_examples: 501 download_size: 758036 dataset_size: 1525909 - config_name: include-Malayalam features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2180381 num_examples: 479 download_size: 802336 dataset_size: 2180381 - config_name: include-Nepali features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2095256 num_examples: 500 download_size: 770698 dataset_size: 2095256 - config_name: include-North_Macedonian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: 'null' - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2551557 num_examples: 551 download_size: 1120617 dataset_size: 2551557 - config_name: include-Persian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2671037 num_examples: 548 download_size: 1243503 dataset_size: 2671037 - config_name: include-Polish features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1606772 num_examples: 548 download_size: 966439 dataset_size: 1606772 - config_name: include-Portuguese features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 3202343 num_examples: 551 download_size: 1769359 dataset_size: 3202343 - config_name: include-Russian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2921687 num_examples: 551 download_size: 1297197 dataset_size: 2921687 - config_name: include-Serbian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: 'null' - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2450730 num_examples: 550 download_size: 1206383 dataset_size: 2450730 - config_name: include-Spanish features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1765544 num_examples: 550 download_size: 963755 dataset_size: 1765544 - config_name: include-Tagalog features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1187141 num_examples: 500 download_size: 610058 dataset_size: 1187141 - config_name: include-Tamil features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2484453 num_examples: 550 download_size: 871985 dataset_size: 2484453 - config_name: include-Telugu features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2724407 num_examples: 548 download_size: 1048659 dataset_size: 2724407 - config_name: include-Turkish features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1956065 num_examples: 548 download_size: 1095535 dataset_size: 1956065 - config_name: include-Ukrainian features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2945260 num_examples: 550 download_size: 1264934 dataset_size: 2945260 - config_name: include-Urdu features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 905425 num_examples: 345 download_size: 439215 dataset_size: 905425 - config_name: include-Uzbek features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1402833 num_examples: 540 download_size: 765528 dataset_size: 1402833 - config_name: include-Vietnamese features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: country dtype: string - name: domain dtype: string - name: input dtype: string - name: language dtype: string - name: level dtype: string - name: option_a dtype: string - name: option_b dtype: string - name: option_c dtype: string - name: option_d dtype: string - name: output dtype: string - name: question dtype: string - name: regional_feature dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2253950 num_examples: 550 download_size: 1033787 dataset_size: 2253950 - config_name: kmmlu-Accounting features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 455825 num_examples: 100 download_size: 230615 dataset_size: 455825 - config_name: kmmlu-Agricultural-Sciences features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 357250 num_examples: 100 download_size: 206916 dataset_size: 357250 - config_name: kmmlu-Aviation-Engineering-and-Maintenance features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 322664 num_examples: 100 download_size: 189804 dataset_size: 322664 - config_name: kmmlu-Biology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 307018 num_examples: 100 download_size: 184839 dataset_size: 307018 - config_name: kmmlu-Chemical-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 344148 num_examples: 100 download_size: 199221 dataset_size: 344148 - config_name: kmmlu-Chemistry features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 329541 num_examples: 100 download_size: 187613 dataset_size: 329541 - config_name: kmmlu-Civil-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 330934 num_examples: 100 download_size: 194608 dataset_size: 330934 - config_name: kmmlu-Computer-Science features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 317539 num_examples: 100 download_size: 182861 dataset_size: 317539 - config_name: kmmlu-Construction features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 316910 num_examples: 100 download_size: 190440 dataset_size: 316910 - config_name: kmmlu-Criminal-Law features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 604667 num_examples: 100 download_size: 331247 dataset_size: 604667 - config_name: kmmlu-Ecology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 335302 num_examples: 100 download_size: 198746 dataset_size: 335302 - config_name: kmmlu-Economics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 385275 num_examples: 100 download_size: 196469 dataset_size: 385275 - config_name: kmmlu-Education features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 415359 num_examples: 100 download_size: 240199 dataset_size: 415359 - config_name: kmmlu-Electrical-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 299304 num_examples: 100 download_size: 167699 dataset_size: 299304 - config_name: kmmlu-Electronics-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 336356 num_examples: 100 download_size: 185918 dataset_size: 336356 - config_name: kmmlu-Energy-Management features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 308471 num_examples: 100 download_size: 177754 dataset_size: 308471 - config_name: kmmlu-Environmental-Science features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 351108 num_examples: 100 download_size: 200202 dataset_size: 351108 - config_name: kmmlu-Fashion features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 317190 num_examples: 100 download_size: 189764 dataset_size: 317190 - config_name: kmmlu-Food-Processing features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 331478 num_examples: 100 download_size: 191062 dataset_size: 331478 - config_name: kmmlu-Gas-Technology-and-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 323628 num_examples: 100 download_size: 188184 dataset_size: 323628 - config_name: kmmlu-Geomatics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 284953 num_examples: 100 download_size: 166859 dataset_size: 284953 - config_name: kmmlu-Health features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 370197 num_examples: 100 download_size: 217818 dataset_size: 370197 - config_name: kmmlu-Industrial-Engineer features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 315007 num_examples: 100 download_size: 187618 dataset_size: 315007 - config_name: kmmlu-Information-Technology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 312736 num_examples: 100 download_size: 182321 dataset_size: 312736 - config_name: kmmlu-Interior-Architecture-and-Design features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 311197 num_examples: 100 download_size: 182970 dataset_size: 311197 - config_name: kmmlu-Korean-History features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: 'null' - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 359045 num_examples: 86 download_size: 210930 dataset_size: 359045 - config_name: kmmlu-Law features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 465800 num_examples: 100 download_size: 244301 dataset_size: 465800 - config_name: kmmlu-Machine-Design-and-Manufacturing features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 310370 num_examples: 100 download_size: 178003 dataset_size: 310370 - config_name: kmmlu-Management features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 409094 num_examples: 100 download_size: 232225 dataset_size: 409094 - config_name: kmmlu-Maritime-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 294984 num_examples: 100 download_size: 170507 dataset_size: 294984 - config_name: kmmlu-Marketing features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 359626 num_examples: 100 download_size: 209314 dataset_size: 359626 - config_name: kmmlu-Materials-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 326734 num_examples: 100 download_size: 181762 dataset_size: 326734 - config_name: kmmlu-Math features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 257499 num_examples: 100 download_size: 133624 dataset_size: 257499 - config_name: kmmlu-Mechanical-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 306944 num_examples: 100 download_size: 175846 dataset_size: 306944 - config_name: kmmlu-Nondestructive-Testing features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 342891 num_examples: 100 download_size: 198193 dataset_size: 342891 - config_name: kmmlu-Patent features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 663588 num_examples: 100 download_size: 338462 dataset_size: 663588 - config_name: kmmlu-Political-Science-and-Sociology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 442042 num_examples: 100 download_size: 253716 dataset_size: 442042 - config_name: kmmlu-Psychology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 398653 num_examples: 100 download_size: 228320 dataset_size: 398653 - config_name: kmmlu-Public-Safety features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 316026 num_examples: 100 download_size: 186026 dataset_size: 316026 - config_name: kmmlu-Railway-and-Automotive-Engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 304652 num_examples: 100 download_size: 177866 dataset_size: 304652 - config_name: kmmlu-Real-Estate features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 481098 num_examples: 100 download_size: 246558 dataset_size: 481098 - config_name: kmmlu-Refrigerating-Machinery features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 326106 num_examples: 100 download_size: 183320 dataset_size: 326106 - config_name: kmmlu-Social-Welfare features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 347782 num_examples: 100 download_size: 195167 dataset_size: 347782 - config_name: kmmlu-Taxation features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 559924 num_examples: 100 download_size: 286171 dataset_size: 559924 - config_name: kmmlu-Telecommunications-and-Wireless-Technology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: B dtype: string - name: C dtype: string - name: Category dtype: string - name: D dtype: string - name: Human Accuracy dtype: float64 - name: answer dtype: int64 - name: input dtype: string - name: output dtype: string - name: question dtype: string splits: - name: test num_bytes: 313232 num_examples: 100 download_size: 179254 dataset_size: 313232 - config_name: kobest features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: alternative_1 dtype: string - name: alternative_2 dtype: string - name: context dtype: string - name: context_1 dtype: string - name: context_2 dtype: string - name: ending_1 dtype: string - name: ending_2 dtype: string - name: ending_3 dtype: string - name: ending_4 dtype: string - name: input dtype: string - name: label dtype: int64 - name: output dtype: string - name: paragraph dtype: string - name: premise dtype: string - name: question dtype: string - name: sentence dtype: string - name: text dtype: string - name: word dtype: string splits: - name: test num_bytes: 1454712 num_examples: 500 download_size: 766558 dataset_size: 1454712 - config_name: math500 features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: input dtype: string - name: level dtype: int64 - name: output dtype: string - name: problem dtype: string - name: solution dtype: string - name: subject dtype: string - name: text dtype: string - name: unique_id dtype: string splits: - name: test num_bytes: 1936110 num_examples: 500 download_size: 971988 dataset_size: 1936110 - config_name: mixeval-freeform features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: benchmark_name dtype: string - name: context dtype: string - name: id dtype: string - name: input dtype: string - name: options sequence: string - name: output dtype: string - name: problem_type dtype: string - name: prompt dtype: string - name: target sequence: string splits: - name: test num_bytes: 4423969 num_examples: 1446 download_size: 2498527 dataset_size: 4423969 - config_name: mixeval-multiplechoice features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: benchmark_name dtype: string - name: context dtype: string - name: id dtype: string - name: input dtype: string - name: options sequence: string - name: output dtype: string - name: problem_type dtype: string - name: prompt dtype: string - name: target sequence: string splits: - name: test num_bytes: 7260217 num_examples: 1739 download_size: 4072278 dataset_size: 7260217 - config_name: mmlu-abstract_algebra features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 396514 num_examples: 100 download_size: 186316 dataset_size: 396514 - config_name: mmlu-anatomy features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 438319 num_examples: 135 download_size: 224230 dataset_size: 438319 - config_name: mmlu-astronomy features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 585054 num_examples: 152 download_size: 319803 dataset_size: 585054 - config_name: mmlu-business_ethics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 392772 num_examples: 100 download_size: 213525 dataset_size: 392772 - config_name: mmlu-clinical_knowledge features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 944599 num_examples: 265 download_size: 512548 dataset_size: 944599 - config_name: mmlu-college_biology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 584846 num_examples: 144 download_size: 320550 dataset_size: 584846 - config_name: mmlu-college_chemistry features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 374156 num_examples: 100 download_size: 207141 dataset_size: 374156 - config_name: mmlu-college_computer_science features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 464743 num_examples: 100 download_size: 250759 dataset_size: 464743 - config_name: mmlu-college_mathematics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 412766 num_examples: 100 download_size: 217900 dataset_size: 412766 - config_name: mmlu-college_medicine features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 771462 num_examples: 173 download_size: 399091 dataset_size: 771462 - config_name: mmlu-college_physics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 334998 num_examples: 91 download_size: 180796 dataset_size: 334998 - config_name: mmlu-computer_security features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 367697 num_examples: 100 download_size: 209636 dataset_size: 367697 - config_name: mmlu-conceptual_physics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 726112 num_examples: 235 download_size: 365548 dataset_size: 726112 - config_name: mmlu-econometrics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 563050 num_examples: 114 download_size: 278069 dataset_size: 563050 - config_name: mmlu-electrical_engineering features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 465299 num_examples: 145 download_size: 248014 dataset_size: 465299 - config_name: mmlu-elementary_mathematics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 792039 num_examples: 377 download_size: 388268 dataset_size: 792039 - config_name: mmlu-formal_logic features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 530858 num_examples: 126 download_size: 221186 dataset_size: 530858 - config_name: mmlu-global_facts features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 260923 num_examples: 100 download_size: 142236 dataset_size: 260923 - config_name: mmlu-high_school_biology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1274666 num_examples: 310 download_size: 660783 dataset_size: 1274666 - config_name: mmlu-high_school_chemistry features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 792464 num_examples: 203 download_size: 395083 dataset_size: 792464 - config_name: mmlu-high_school_computer_science features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 393538 num_examples: 100 download_size: 211337 dataset_size: 393538 - config_name: mmlu-high_school_european_history features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1336493 num_examples: 165 download_size: 716685 dataset_size: 1336493 - config_name: mmlu-high_school_geography features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 637247 num_examples: 198 download_size: 349021 dataset_size: 637247 - config_name: mmlu-high_school_government_and_politics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 781581 num_examples: 193 download_size: 407738 dataset_size: 781581 - config_name: mmlu-high_school_macroeconomics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1578320 num_examples: 390 download_size: 733264 dataset_size: 1578320 - config_name: mmlu-high_school_mathematics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 800491 num_examples: 270 download_size: 396222 dataset_size: 800491 - config_name: mmlu-high_school_microeconomics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1012358 num_examples: 238 download_size: 480599 dataset_size: 1012358 - config_name: mmlu-high_school_physics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 632931 num_examples: 151 download_size: 315824 dataset_size: 632931 - config_name: mmlu-high_school_psychology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1939848 num_examples: 534 download_size: 1034794 dataset_size: 1939848 - config_name: mmlu-high_school_statistics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 983678 num_examples: 216 download_size: 482350 dataset_size: 983678 - config_name: mmlu-high_school_us_history features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1567131 num_examples: 204 download_size: 839557 dataset_size: 1567131 - config_name: mmlu-high_school_world_history features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1899496 num_examples: 237 download_size: 1032261 dataset_size: 1899496 - config_name: mmlu-human_aging features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 743431 num_examples: 223 download_size: 416261 dataset_size: 743431 - config_name: mmlu-human_sexuality features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 442853 num_examples: 131 download_size: 254679 dataset_size: 442853 - config_name: mmlu-international_law features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 599331 num_examples: 121 download_size: 312198 dataset_size: 599331 - config_name: mmlu-jurisprudence features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 487437 num_examples: 108 download_size: 267604 dataset_size: 487437 - config_name: mmlu-logical_fallacies features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 570790 num_examples: 163 download_size: 275460 dataset_size: 570790 - config_name: mmlu-machine_learning features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 505413 num_examples: 112 download_size: 263505 dataset_size: 505413 - config_name: mmlu-management features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 331324 num_examples: 103 download_size: 187277 dataset_size: 331324 - config_name: mmlu-marketing features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 752045 num_examples: 234 download_size: 397928 dataset_size: 752045 - config_name: mmlu-medical_genetics features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 340549 num_examples: 100 download_size: 189314 dataset_size: 340549 - config_name: mmlu-miscellaneous features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1950260 num_examples: 783 download_size: 1109892 dataset_size: 1950260 - config_name: mmlu-moral_disputes features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1369187 num_examples: 346 download_size: 724291 dataset_size: 1369187 - config_name: mmlu-moral_scenarios features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 3175519 num_examples: 895 download_size: 1284130 dataset_size: 3175519 - config_name: mmlu-nutrition features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1250528 num_examples: 306 download_size: 666736 dataset_size: 1250528 - config_name: mmlu-philosophy features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1196367 num_examples: 311 download_size: 647847 dataset_size: 1196367 - config_name: mmlu-prehistory features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1176086 num_examples: 324 download_size: 642502 dataset_size: 1176086 - config_name: mmlu-pro features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: string - name: answer_index dtype: int64 - name: category dtype: string - name: cot_content dtype: string - name: input dtype: string - name: options sequence: string - name: output dtype: string - name: question dtype: string - name: question_id dtype: int64 - name: src dtype: string splits: - name: test num_bytes: 62433547 num_examples: 12032 download_size: 33078741 dataset_size: 62433547 - config_name: mmlu-professional_accounting features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1187699 num_examples: 282 download_size: 601601 dataset_size: 1187699 - config_name: mmlu-professional_law features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 11841622 num_examples: 1534 download_size: 6076477 dataset_size: 11841622 - config_name: mmlu-professional_medicine features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1512318 num_examples: 272 download_size: 835103 dataset_size: 1512318 - config_name: mmlu-professional_psychology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 2577687 num_examples: 611 download_size: 1366336 dataset_size: 2577687 - config_name: mmlu-public_relations features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 372143 num_examples: 108 download_size: 217327 dataset_size: 372143 - config_name: mmlu-security_studies features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 1508441 num_examples: 245 download_size: 800391 dataset_size: 1508441 - config_name: mmlu-sociology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 811007 num_examples: 201 download_size: 454642 dataset_size: 811007 - config_name: mmlu-us_foreign_policy features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 403284 num_examples: 99 download_size: 233714 dataset_size: 403284 - config_name: mmlu-virology features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 623374 num_examples: 166 download_size: 352074 dataset_size: 623374 - config_name: mmlu-world_religions features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: answer dtype: int64 - name: choices sequence: string - name: input dtype: string - name: output dtype: string - name: question dtype: string - name: subject dtype: string splits: - name: test num_bytes: 453294 num_examples: 171 download_size: 265100 dataset_size: 453294 - config_name: mmmlu-JA_JP features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: Answer dtype: string - name: B dtype: string - name: C dtype: string - name: D dtype: string - name: Question dtype: string - name: Subject dtype: string - name: 'Unnamed: 0' dtype: int64 - name: input dtype: string - name: output dtype: string splits: - name: test num_bytes: 65116427 num_examples: 13944 download_size: 33791366 dataset_size: 65116427 - config_name: mmmlu-KO_KR features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: A dtype: string - name: Answer dtype: string - name: B dtype: string - name: C dtype: string - name: D dtype: string - name: Question dtype: string - name: Subject dtype: string - name: 'Unnamed: 0' dtype: int64 - name: input dtype: string - name: output dtype: string splits: - name: test num_bytes: 59596704 num_examples: 13952 download_size: 31912422 dataset_size: 59596704 - config_name: rewardbench features: - name: question dtype: string - name: reasoning dtype: string - name: answer dtype: string - name: input struct: - name: chosen dtype: string - name: chosen_model dtype: string - name: id dtype: int64 - name: input dtype: string - name: output dtype: string - name: prompt dtype: string - name: rejected dtype: string - name: rejected_model dtype: string - name: subset dtype: string splits: - name: test num_bytes: 24583412 num_examples: 3438 download_size: 13806076 dataset_size: 24583412 configs: - config_name: ChemBench data_files: - split: test path: ChemBench/test-* - config_name: MolLangBench data_files: - split: test path: MolLangBench/test-* - config_name: MultiMedQA data_files: - split: test path: MultiMedQA/test-* - config_name: agieval-math data_files: - split: test path: agieval-math/test-* - config_name: aime25 data_files: - split: test path: aime25/test-* - config_name: arc-ARC-Challenge data_files: - split: test path: arc-ARC-Challenge/test-* - config_name: arc-ARC-Easy data_files: - split: test path: arc-ARC-Easy/test-* - config_name: arena-hard-v0.1-gpt-4o-2024-08-06 data_files: - split: test path: arena-hard-v0.1-gpt-4o-2024-08-06/test-* - config_name: arena-hard-v2-gemini-2.5 data_files: - split: test path: arena-hard-v2-gemini-2.5/test-* - config_name: attributionbench data_files: - split: test path: attributionbench/test-* - config_name: bbh-boolean_expressions data_files: - split: test path: bbh-boolean_expressions/test-* - config_name: bbh-causal_judgement data_files: - split: test path: bbh-causal_judgement/test-* - config_name: bbh-date_understanding data_files: - split: test path: bbh-date_understanding/test-* - config_name: bbh-disambiguation_qa data_files: - split: test path: bbh-disambiguation_qa/test-* - config_name: bbh-dyck_languages data_files: - split: test path: bbh-dyck_languages/test-* - config_name: bbh-formal_fallacies data_files: - split: test path: bbh-formal_fallacies/test-* - config_name: bbh-geometric_shapes data_files: - split: test path: bbh-geometric_shapes/test-* - config_name: bbh-hyperbaton data_files: - split: test path: bbh-hyperbaton/test-* - config_name: bbh-logical_deduction_five_objects data_files: - split: test path: bbh-logical_deduction_five_objects/test-* - config_name: bbh-logical_deduction_seven_objects data_files: - split: test path: bbh-logical_deduction_seven_objects/test-* - config_name: bbh-logical_deduction_three_objects data_files: - split: test path: bbh-logical_deduction_three_objects/test-* - config_name: bbh-movie_recommendation data_files: - split: test path: bbh-movie_recommendation/test-* - config_name: bbh-multistep_arithmetic_two data_files: - split: test path: bbh-multistep_arithmetic_two/test-* - config_name: bbh-navigate data_files: - split: test path: bbh-navigate/test-* - config_name: bbh-object_counting data_files: - split: test path: bbh-object_counting/test-* - config_name: bbh-penguins_in_a_table data_files: - split: test path: bbh-penguins_in_a_table/test-* - config_name: bbh-reasoning_about_colored_objects data_files: - split: test path: bbh-reasoning_about_colored_objects/test-* - config_name: bbh-ruin_names data_files: - split: test path: bbh-ruin_names/test-* - config_name: bbh-salient_translation_error_detection data_files: - split: test path: bbh-salient_translation_error_detection/test-* - config_name: bbh-snarks data_files: - split: test path: bbh-snarks/test-* - config_name: bbh-sports_understanding data_files: - split: test path: bbh-sports_understanding/test-* - config_name: bbh-temporal_sequences data_files: - split: test path: bbh-temporal_sequences/test-* - config_name: bbh-tracking_shuffled_objects_five_objects data_files: - split: test path: bbh-tracking_shuffled_objects_five_objects/test-* - config_name: bbh-tracking_shuffled_objects_seven_objects data_files: - split: test path: bbh-tracking_shuffled_objects_seven_objects/test-* - config_name: bbh-tracking_shuffled_objects_three_objects data_files: - split: test path: bbh-tracking_shuffled_objects_three_objects/test-* - config_name: bbh-web_of_lies data_files: - split: test path: bbh-web_of_lies/test-* - config_name: bbh-word_sorting data_files: - split: test path: bbh-word_sorting/test-* - config_name: cqa data_files: - split: test path: cqa/test-* - config_name: csatqa data_files: - split: test path: csatqa/test-* - config_name: drop data_files: - split: test path: drop/test-* - config_name: global-mmlu-ja data_files: - split: test path: global-mmlu-ja/test-* - config_name: global-mmlu-ko data_files: - split: test path: global-mmlu-ko/test-* - config_name: gpqa data_files: - split: test path: gpqa/test-* - config_name: gsm8k data_files: - split: test path: gsm8k/test-* - config_name: haerae-general_knowledge data_files: - split: test path: haerae-general_knowledge/test-* - config_name: haerae-history data_files: - split: test path: haerae-history/test-* - config_name: haerae-loan_words data_files: - split: test path: haerae-loan_words/test-* - config_name: haerae-rare_words data_files: - split: test path: haerae-rare_words/test-* - config_name: haerae-reading_comprehension data_files: - split: test path: haerae-reading_comprehension/test-* - config_name: haerae-standard_nomenclature data_files: - split: test path: haerae-standard_nomenclature/test-* - config_name: hellaswag data_files: - split: test path: hellaswag/test-* - config_name: hle data_files: - split: test path: hle/test-* - config_name: hrm8k data_files: - split: test path: hrm8k/test-* - config_name: include-Albanian data_files: - split: test path: include-Albanian/test-* - config_name: include-Arabic data_files: - split: test path: include-Arabic/test-* - config_name: include-Armenian data_files: - split: test path: include-Armenian/test-* - config_name: include-Azerbaijani data_files: - split: test path: include-Azerbaijani/test-* - config_name: include-Basque data_files: - split: test path: include-Basque/test-* - config_name: include-Belarusian data_files: - split: test path: include-Belarusian/test-* - config_name: include-Bengali data_files: - split: test path: include-Bengali/test-* - config_name: include-Bulgarian data_files: - split: test path: include-Bulgarian/test-* - config_name: include-Chinese data_files: - split: test path: include-Chinese/test-* - config_name: include-Croatian data_files: - split: test path: include-Croatian/test-* - config_name: include-Dutch data_files: - split: test path: include-Dutch/test-* - config_name: include-Estonian data_files: - split: test path: include-Estonian/test-* - config_name: include-Finnish data_files: - split: test path: include-Finnish/test-* - config_name: include-French data_files: - split: test path: include-French/test-* - config_name: include-Georgian data_files: - split: test path: include-Georgian/test-* - config_name: include-German data_files: - split: test path: include-German/test-* - config_name: include-Greek data_files: - split: test path: include-Greek/test-* - config_name: include-Hebrew data_files: - split: test path: include-Hebrew/test-* - config_name: include-Hindi data_files: - split: test path: include-Hindi/test-* - config_name: include-Hungarian data_files: - split: test path: include-Hungarian/test-* - config_name: include-Indonesian data_files: - split: test path: include-Indonesian/test-* - config_name: include-Italian data_files: - split: test path: include-Italian/test-* - config_name: include-Japanese data_files: - split: test path: include-Japanese/test-* - config_name: include-Kazakh data_files: - split: test path: include-Kazakh/test-* - config_name: include-Korean data_files: - split: test path: include-Korean/test-* - config_name: include-Lithuanian data_files: - split: test path: include-Lithuanian/test-* - config_name: include-Malay data_files: - split: test path: include-Malay/test-* - config_name: include-Malayalam data_files: - split: test path: include-Malayalam/test-* - config_name: include-Nepali data_files: - split: test path: include-Nepali/test-* - config_name: include-North_Macedonian data_files: - split: test path: include-North_Macedonian/test-* - config_name: include-Persian data_files: - split: test path: include-Persian/test-* - config_name: include-Polish data_files: - split: test path: include-Polish/test-* - config_name: include-Portuguese data_files: - split: test path: include-Portuguese/test-* - config_name: include-Russian data_files: - split: test path: include-Russian/test-* - config_name: include-Serbian data_files: - split: test path: include-Serbian/test-* - config_name: include-Spanish data_files: - split: test path: include-Spanish/test-* - config_name: include-Tagalog data_files: - split: test path: include-Tagalog/test-* - config_name: include-Tamil data_files: - split: test path: include-Tamil/test-* - config_name: include-Telugu data_files: - split: test path: include-Telugu/test-* - config_name: include-Turkish data_files: - split: test path: include-Turkish/test-* - config_name: include-Ukrainian data_files: - split: test path: include-Ukrainian/test-* - config_name: include-Urdu data_files: - split: test path: include-Urdu/test-* - config_name: include-Uzbek data_files: - split: test path: include-Uzbek/test-* - config_name: include-Vietnamese data_files: - split: test path: include-Vietnamese/test-* - config_name: kmmlu-Accounting data_files: - split: test path: kmmlu-Accounting/test-* - config_name: kmmlu-Agricultural-Sciences data_files: - split: test path: kmmlu-Agricultural-Sciences/test-* - config_name: kmmlu-Aviation-Engineering-and-Maintenance data_files: - split: test path: kmmlu-Aviation-Engineering-and-Maintenance/test-* - config_name: kmmlu-Biology data_files: - split: test path: kmmlu-Biology/test-* - config_name: kmmlu-Chemical-Engineering data_files: - split: test path: kmmlu-Chemical-Engineering/test-* - config_name: kmmlu-Chemistry data_files: - split: test path: kmmlu-Chemistry/test-* - config_name: kmmlu-Civil-Engineering data_files: - split: test path: kmmlu-Civil-Engineering/test-* - config_name: kmmlu-Computer-Science data_files: - split: test path: kmmlu-Computer-Science/test-* - config_name: kmmlu-Construction data_files: - split: test path: kmmlu-Construction/test-* - config_name: kmmlu-Criminal-Law data_files: - split: test path: kmmlu-Criminal-Law/test-* - config_name: kmmlu-Ecology data_files: - split: test path: kmmlu-Ecology/test-* - config_name: kmmlu-Economics data_files: - split: test path: kmmlu-Economics/test-* - config_name: kmmlu-Education data_files: - split: test path: kmmlu-Education/test-* - config_name: kmmlu-Electrical-Engineering data_files: - split: test path: kmmlu-Electrical-Engineering/test-* - config_name: kmmlu-Electronics-Engineering data_files: - split: test path: kmmlu-Electronics-Engineering/test-* - config_name: kmmlu-Energy-Management data_files: - split: test path: kmmlu-Energy-Management/test-* - config_name: kmmlu-Environmental-Science data_files: - split: test path: kmmlu-Environmental-Science/test-* - config_name: kmmlu-Fashion data_files: - split: test path: kmmlu-Fashion/test-* - config_name: kmmlu-Food-Processing data_files: - split: test path: kmmlu-Food-Processing/test-* - config_name: kmmlu-Gas-Technology-and-Engineering data_files: - split: test path: kmmlu-Gas-Technology-and-Engineering/test-* - config_name: kmmlu-Geomatics data_files: - split: test path: kmmlu-Geomatics/test-* - config_name: kmmlu-Health data_files: - split: test path: kmmlu-Health/test-* - config_name: kmmlu-Industrial-Engineer data_files: - split: test path: kmmlu-Industrial-Engineer/test-* - config_name: kmmlu-Information-Technology data_files: - split: test path: kmmlu-Information-Technology/test-* - config_name: kmmlu-Interior-Architecture-and-Design data_files: - split: test path: kmmlu-Interior-Architecture-and-Design/test-* - config_name: kmmlu-Korean-History data_files: - split: test path: kmmlu-Korean-History/test-* - config_name: kmmlu-Law data_files: - split: test path: kmmlu-Law/test-* - config_name: kmmlu-Machine-Design-and-Manufacturing data_files: - split: test path: kmmlu-Machine-Design-and-Manufacturing/test-* - config_name: kmmlu-Management data_files: - split: test path: kmmlu-Management/test-* - config_name: kmmlu-Maritime-Engineering data_files: - split: test path: kmmlu-Maritime-Engineering/test-* - config_name: kmmlu-Marketing data_files: - split: test path: kmmlu-Marketing/test-* - config_name: kmmlu-Materials-Engineering data_files: - split: test path: kmmlu-Materials-Engineering/test-* - config_name: kmmlu-Math data_files: - split: test path: kmmlu-Math/test-* - config_name: kmmlu-Mechanical-Engineering data_files: - split: test path: kmmlu-Mechanical-Engineering/test-* - config_name: kmmlu-Nondestructive-Testing data_files: - split: test path: kmmlu-Nondestructive-Testing/test-* - config_name: kmmlu-Patent data_files: - split: test path: kmmlu-Patent/test-* - config_name: kmmlu-Political-Science-and-Sociology data_files: - split: test path: kmmlu-Political-Science-and-Sociology/test-* - config_name: kmmlu-Psychology data_files: - split: test path: kmmlu-Psychology/test-* - config_name: kmmlu-Public-Safety data_files: - split: test path: kmmlu-Public-Safety/test-* - config_name: kmmlu-Railway-and-Automotive-Engineering data_files: - split: test path: kmmlu-Railway-and-Automotive-Engineering/test-* - config_name: kmmlu-Real-Estate data_files: - split: test path: kmmlu-Real-Estate/test-* - config_name: kmmlu-Refrigerating-Machinery data_files: - split: test path: kmmlu-Refrigerating-Machinery/test-* - config_name: kmmlu-Social-Welfare data_files: - split: test path: kmmlu-Social-Welfare/test-* - config_name: kmmlu-Taxation data_files: - split: test path: kmmlu-Taxation/test-* - config_name: kmmlu-Telecommunications-and-Wireless-Technology data_files: - split: test path: kmmlu-Telecommunications-and-Wireless-Technology/test-* - config_name: kobest data_files: - split: test path: kobest/test-* - config_name: math500 data_files: - split: test path: math500/test-* - config_name: mixeval-freeform data_files: - split: test path: mixeval-freeform/test-* - config_name: mixeval-multiplechoice data_files: - split: test path: mixeval-multiplechoice/test-* - config_name: mmlu-abstract_algebra data_files: - split: test path: mmlu-abstract_algebra/test-* - config_name: mmlu-anatomy data_files: - split: test path: mmlu-anatomy/test-* - config_name: mmlu-astronomy data_files: - split: test path: mmlu-astronomy/test-* - config_name: mmlu-business_ethics data_files: - split: test path: mmlu-business_ethics/test-* - config_name: mmlu-clinical_knowledge data_files: - split: test path: mmlu-clinical_knowledge/test-* - config_name: mmlu-college_biology data_files: - split: test path: mmlu-college_biology/test-* - config_name: mmlu-college_chemistry data_files: - split: test path: mmlu-college_chemistry/test-* - config_name: mmlu-college_computer_science data_files: - split: test path: mmlu-college_computer_science/test-* - config_name: mmlu-college_mathematics data_files: - split: test path: mmlu-college_mathematics/test-* - config_name: mmlu-college_medicine data_files: - split: test path: mmlu-college_medicine/test-* - config_name: mmlu-college_physics data_files: - split: test path: mmlu-college_physics/test-* - config_name: mmlu-computer_security data_files: - split: test path: mmlu-computer_security/test-* - config_name: mmlu-conceptual_physics data_files: - split: test path: mmlu-conceptual_physics/test-* - config_name: mmlu-econometrics data_files: - split: test path: mmlu-econometrics/test-* - config_name: mmlu-electrical_engineering data_files: - split: test path: mmlu-electrical_engineering/test-* - config_name: mmlu-elementary_mathematics data_files: - split: test path: mmlu-elementary_mathematics/test-* - config_name: mmlu-formal_logic data_files: - split: test path: mmlu-formal_logic/test-* - config_name: mmlu-global_facts data_files: - split: test path: mmlu-global_facts/test-* - config_name: mmlu-high_school_biology data_files: - split: test path: mmlu-high_school_biology/test-* - config_name: mmlu-high_school_chemistry data_files: - split: test path: mmlu-high_school_chemistry/test-* - config_name: mmlu-high_school_computer_science data_files: - split: test path: mmlu-high_school_computer_science/test-* - config_name: mmlu-high_school_european_history data_files: - split: test path: mmlu-high_school_european_history/test-* - config_name: mmlu-high_school_geography data_files: - split: test path: mmlu-high_school_geography/test-* - config_name: mmlu-high_school_government_and_politics data_files: - split: test path: mmlu-high_school_government_and_politics/test-* - config_name: mmlu-high_school_macroeconomics data_files: - split: test path: mmlu-high_school_macroeconomics/test-* - config_name: mmlu-high_school_mathematics data_files: - split: test path: mmlu-high_school_mathematics/test-* - config_name: mmlu-high_school_microeconomics data_files: - split: test path: mmlu-high_school_microeconomics/test-* - config_name: mmlu-high_school_physics data_files: - split: test path: mmlu-high_school_physics/test-* - config_name: mmlu-high_school_psychology data_files: - split: test path: mmlu-high_school_psychology/test-* - config_name: mmlu-high_school_statistics data_files: - split: test path: mmlu-high_school_statistics/test-* - config_name: mmlu-high_school_us_history data_files: - split: test path: mmlu-high_school_us_history/test-* - config_name: mmlu-high_school_world_history data_files: - split: test path: mmlu-high_school_world_history/test-* - config_name: mmlu-human_aging data_files: - split: test path: mmlu-human_aging/test-* - config_name: mmlu-human_sexuality data_files: - split: test path: mmlu-human_sexuality/test-* - config_name: mmlu-international_law data_files: - split: test path: mmlu-international_law/test-* - config_name: mmlu-jurisprudence data_files: - split: test path: mmlu-jurisprudence/test-* - config_name: mmlu-logical_fallacies data_files: - split: test path: mmlu-logical_fallacies/test-* - config_name: mmlu-machine_learning data_files: - split: test path: mmlu-machine_learning/test-* - config_name: mmlu-management data_files: - split: test path: mmlu-management/test-* - config_name: mmlu-marketing data_files: - split: test path: mmlu-marketing/test-* - config_name: mmlu-medical_genetics data_files: - split: test path: mmlu-medical_genetics/test-* - config_name: mmlu-miscellaneous data_files: - split: test path: mmlu-miscellaneous/test-* - config_name: mmlu-moral_disputes data_files: - split: test path: mmlu-moral_disputes/test-* - config_name: mmlu-moral_scenarios data_files: - split: test path: mmlu-moral_scenarios/test-* - config_name: mmlu-nutrition data_files: - split: test path: mmlu-nutrition/test-* - config_name: mmlu-philosophy data_files: - split: test path: mmlu-philosophy/test-* - config_name: mmlu-prehistory data_files: - split: test path: mmlu-prehistory/test-* - config_name: mmlu-pro data_files: - split: test path: mmlu-pro/test-* - config_name: mmlu-professional_accounting data_files: - split: test path: mmlu-professional_accounting/test-* - config_name: mmlu-professional_law data_files: - split: test path: mmlu-professional_law/test-* - config_name: mmlu-professional_medicine data_files: - split: test path: mmlu-professional_medicine/test-* - config_name: mmlu-professional_psychology data_files: - split: test path: mmlu-professional_psychology/test-* - config_name: mmlu-public_relations data_files: - split: test path: mmlu-public_relations/test-* - config_name: mmlu-security_studies data_files: - split: test path: mmlu-security_studies/test-* - config_name: mmlu-sociology data_files: - split: test path: mmlu-sociology/test-* - config_name: mmlu-us_foreign_policy data_files: - split: test path: mmlu-us_foreign_policy/test-* - config_name: mmlu-virology data_files: - split: test path: mmlu-virology/test-* - config_name: mmlu-world_religions data_files: - split: test path: mmlu-world_religions/test-* - config_name: mmmlu-JA_JP data_files: - split: test path: mmmlu-JA_JP/test-* - config_name: mmmlu-KO_KR data_files: - split: test path: mmmlu-KO_KR/test-* - config_name: rewardbench data_files: - split: test path: rewardbench/test-* ---

应用场景：