遇见数据集

open-llm-leaderboard-old/details_ehartford__WizardLM-30B-Uncensored

收藏
Hugging Face2023-10-13 更新2024-06-22 收录
官方服务:

资源简介:

--- pretty_name: Evaluation run of ehartford/WizardLM-30B-Uncensored dataset_summary: "Dataset automatically created during the evaluation run of model\ \ [ehartford/WizardLM-30B-Uncensored](https://huggingface.co/ehartford/WizardLM-30B-Uncensored)\ \ on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\ \nThe dataset is composed of 64 configuration, each one coresponding to one of the\ \ evaluated task.\n\nThe dataset has been created from 2 run(s). Each run can be\ \ found as a specific split in each configuration, the split being named using the\ \ timestamp of the run.The \"train\" split is always pointing to the latest results.\n\ \nAn additional configuration \"results\" store all the aggregated results of the\ \ run (and is used to compute and display the agregated metrics on the [Open LLM\ \ Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)).\n\ \nTo load the details from a run, you can for instance do the following:\n```python\n\ from datasets import load_dataset\ndata = load_dataset(\"open-llm-leaderboard/details_ehartford__WizardLM-30B-Uncensored\"\ ,\n\t\"harness_winogrande_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\n\ These are the [latest results from run 2023-10-13T01:45:13.473461](https://huggingface.co/datasets/open-llm-leaderboard/details_ehartford__WizardLM-30B-Uncensored/blob/main/results_2023-10-13T01-45-13.473461.json)(note\ \ that their might be results for other tasks in the repos if successive evals didn't\ \ cover the same tasks. You find each in the results and the \"latest\" split for\ \ each eval):\n\n```python\n{\n \"all\": {\n \"em\": 0.1761744966442953,\n\ \ \"em_stderr\": 0.0039014746298017622,\n \"f1\": 0.2744651845637597,\n\ \ \"f1_stderr\": 0.004005542048425456,\n \"acc\": 0.43618703748803983,\n\ \ \"acc_stderr\": 0.010751614385080125\n },\n \"harness|drop|3\": {\n\ \ \"em\": 0.1761744966442953,\n \"em_stderr\": 0.0039014746298017622,\n\ \ \"f1\": 0.2744651845637597,\n \"f1_stderr\": 0.004005542048425456\n\ \ },\n \"harness|gsm8k|5\": {\n \"acc\": 0.1288855193328279,\n \ \ \"acc_stderr\": 0.009229580761400267\n },\n \"harness|winogrande|5\"\ : {\n \"acc\": 0.7434885556432518,\n \"acc_stderr\": 0.012273648008759984\n\ \ }\n}\n```" repo_url: https://huggingface.co/ehartford/WizardLM-30B-Uncensored leaderboard_url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard point_of_contact: clementine@hf.co configs: - config_name: harness_arc_challenge_25 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|arc:challenge|25_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|arc:challenge|25_2023-07-19T22:33:53.366908.parquet' - config_name: harness_drop_3 data_files: - split: 2023_10_13T01_45_13.473461 path: - '**/details_harness|drop|3_2023-10-13T01-45-13.473461.parquet' - split: latest path: - '**/details_harness|drop|3_2023-10-13T01-45-13.473461.parquet' - config_name: harness_gsm8k_5 data_files: - split: 2023_10_13T01_45_13.473461 path: - '**/details_harness|gsm8k|5_2023-10-13T01-45-13.473461.parquet' - split: latest path: - '**/details_harness|gsm8k|5_2023-10-13T01-45-13.473461.parquet' - config_name: harness_hellaswag_10 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hellaswag|10_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hellaswag|10_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-international_law|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-management|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-marketing|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-sociology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-virology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-international_law|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-management|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-marketing|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-sociology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-virology|5_2023-07-19T22:33:53.366908.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_abstract_algebra_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_anatomy_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-anatomy|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-anatomy|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_astronomy_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-astronomy|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-astronomy|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_business_ethics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_clinical_knowledge_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_biology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_biology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_biology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_chemistry_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_computer_science_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_mathematics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_medicine_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_college_physics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-college_physics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_physics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_computer_security_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-computer_security|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-computer_security|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_conceptual_physics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_econometrics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-econometrics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-econometrics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_electrical_engineering_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_elementary_mathematics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_formal_logic_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-formal_logic|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-formal_logic|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_global_facts_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-global_facts|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-global_facts|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_biology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_chemistry_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_computer_science_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_european_history_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_geography_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_government_and_politics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_macroeconomics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_mathematics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_microeconomics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_physics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_psychology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_statistics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_us_history_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_high_school_world_history_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_human_aging_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-human_aging|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_aging|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_human_sexuality_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_international_law_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-international_law|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-international_law|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_jurisprudence_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_logical_fallacies_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_machine_learning_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-machine_learning|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-machine_learning|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_management_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-management|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-management|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_marketing_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-marketing|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-marketing|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_medical_genetics_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_miscellaneous_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_moral_disputes_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_moral_scenarios_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_nutrition_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-nutrition|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-nutrition|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_philosophy_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-philosophy|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-philosophy|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_prehistory_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-prehistory|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-prehistory|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_professional_accounting_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_professional_law_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-professional_law|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_law|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_professional_medicine_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_professional_psychology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-professional_psychology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_psychology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_public_relations_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-public_relations|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-public_relations|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_security_studies_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-security_studies|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-security_studies|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_sociology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-sociology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-sociology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_us_foreign_policy_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_virology_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-virology|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-virology|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_hendrycksTest_world_religions_5 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|hendrycksTest-world_religions|5_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|hendrycksTest-world_religions|5_2023-07-19T22:33:53.366908.parquet' - config_name: harness_truthfulqa_mc_0 data_files: - split: 2023_07_19T22_33_53.366908 path: - '**/details_harness|truthfulqa:mc|0_2023-07-19T22:33:53.366908.parquet' - split: latest path: - '**/details_harness|truthfulqa:mc|0_2023-07-19T22:33:53.366908.parquet' - config_name: harness_winogrande_5 data_files: - split: 2023_10_13T01_45_13.473461 path: - '**/details_harness|winogrande|5_2023-10-13T01-45-13.473461.parquet' - split: latest path: - '**/details_harness|winogrande|5_2023-10-13T01-45-13.473461.parquet' - config_name: results data_files: - split: 2023_07_19T22_33_53.366908 path: - results_2023-07-19T22:33:53.366908.parquet - split: 2023_10_13T01_45_13.473461 path: - results_2023-10-13T01-45-13.473461.parquet - split: latest path: - results_2023-10-13T01-45-13.473461.parquet --- # Dataset Card for Evaluation run of ehartford/WizardLM-30B-Uncensored ## Dataset Description - **Homepage:** - **Repository:** https://huggingface.co/ehartford/WizardLM-30B-Uncensored - **Paper:** - **Leaderboard:** https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard - **Point of Contact:** clementine@hf.co ### Dataset Summary Dataset automatically created during the evaluation run of model [ehartford/WizardLM-30B-Uncensored](https://huggingface.co/ehartford/WizardLM-30B-Uncensored) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). The dataset is composed of 64 configuration, each one coresponding to one of the evaluated task. The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results. An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the agregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)). To load the details from a run, you can for instance do the following: ```python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ehartford__WizardLM-30B-Uncensored", "harness_winogrande_5", split="train") ``` ## Latest results These are the [latest results from run 2023-10-13T01:45:13.473461](https://huggingface.co/datasets/open-llm-leaderboard/details_ehartford__WizardLM-30B-Uncensored/blob/main/results_2023-10-13T01-45-13.473461.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval): ```python { "all": { "em": 0.1761744966442953, "em_stderr": 0.0039014746298017622, "f1": 0.2744651845637597, "f1_stderr": 0.004005542048425456, "acc": 0.43618703748803983, "acc_stderr": 0.010751614385080125 }, "harness|drop|3": { "em": 0.1761744966442953, "em_stderr": 0.0039014746298017622, "f1": 0.2744651845637597, "f1_stderr": 0.004005542048425456 }, "harness|gsm8k|5": { "acc": 0.1288855193328279, "acc_stderr": 0.009229580761400267 }, "harness|winogrande|5": { "acc": 0.7434885556432518, "acc_stderr": 0.012273648008759984 } } ``` ### Supported Tasks and Leaderboards [More Information Needed] ### Languages [More Information Needed] ## Dataset Structure ### Data Instances [More Information Needed] ### Data Fields [More Information Needed] ### Data Splits [More Information Needed] ## Dataset Creation ### Curation Rationale [More Information Needed] ### Source Data #### Initial Data Collection and Normalization [More Information Needed] #### Who are the source language producers? [More Information Needed] ### Annotations #### Annotation process [More Information Needed] #### Who are the annotators? [More Information Needed] ### Personal and Sensitive Information [More Information Needed] ## Considerations for Using the Data ### Social Impact of Dataset [More Information Needed] ### Discussion of Biases [More Information Needed] ### Other Known Limitations [More Information Needed] ## Additional Information ### Dataset Curators [More Information Needed] ### Licensing Information [More Information Needed] ### Citation Information [More Information Needed] ### Contributions [More Information Needed]

原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 ehartford/WizardLM-30B-Uncensored 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。

数据集结构

  • 数据集包含 64 个配置,每个配置对应一个评估任务。
  • 数据集从 2 次运行中创建,每次运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。
  • "train" 分割始终指向最新的结果。
  • 额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示 Open LLM Leaderboard 上的聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ehartford__WizardLM-30B-Uncensored", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-13T01:45:13.473461 运行的最新结果: python { "all": { "em": 0.1761744966442953, "em_stderr": 0.0039014746298017622, "f1": 0.2744651845637597, "f1_stderr": 0.004005542048425456, "acc": 0.43618703748803983, "acc_stderr": 0.010751614385080125 }, "harness|drop|3": { "em": 0.1761744966442953, "em_stderr": 0.0039014746298017622, "f1": 0.2744651845637597, "f1_stderr": 0.004005542048425456 }, "harness|gsm8k|5": { "acc": 0.1288855193328279, "acc_stderr": 0.009229580761400267 }, "harness|winogrande|5": { "acc": 0.7434885556432518, "acc_stderr": 0.012273648008759984 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_07_19T22_33_53.366908
    • 路径:**/details_harness|arc:challenge|25_2023-07-19T22:33:53.366908.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-07-19T22:33:53.366908.parquet
  • harness_drop_3

    • 分割:2023_10_13T01_45_13.473461
    • 路径:**/details_harness|drop|3_2023-10-13T01-45-13.473461.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-10-13T01-45-13.473461.parquet
  • harness_gsm8k_5

    • 分割:2023_10_13T01_45_13.473461
    • 路径:**/details_harness|gsm8k|5_2023-10-13T01-45-13.473461.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-13T01-45-13.473461.parquet
  • harness_hellaswag_10

    • 分割:2023_07_19T22_33_53.366908
    • 路径:**/details_harness|hellaswag|10_2023-07-19T22:33:53.366908.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-07-19T22:33:53.366908.parquet
  • harness_hendrycksTest_5

    • 分割:2023_07_19T22_33_53.366908
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-management|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T22:33:53.366908.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-07-19T22:33:53.366908.parquet
      • `**/details_harness|hendrycksTest-professional_medicine|
搜集汇总
数据集介绍
open-llm-leaderboard-old/details_ehartford__WizardLM-30B-Uncensored 数据集图片
构建方式
该数据集是在对模型ehartford/WizardLM-30B-Uncensored进行Open LLM Leaderboard评估过程中自动生成的。数据集由64个配置组成,每个配置对应一个被评估的任务。数据来源于两次独立的评估运行,每次运行的结果被存储为对应配置中的一个独立分割,分割以运行的时间戳命名。此外,存在一个名为“results”的额外配置,用于汇总所有运行的聚合结果,并用于计算和展示Open LLM Leaderboard上的综合指标。
特点
该数据集的核心特点在于其细粒度的任务覆盖与版本化追踪机制。通过64个配置对应不同任务,每个任务均包含多次评估运行的详细结果,并以时间戳作为分割标识,便于研究者追溯模型性能的演变。数据集中的“train”分割始终指向最新一次评估的结果,而“results”配置则提供了跨任务的聚合性能概览,使得模型的整体能力一目了然。
使用方法
使用该数据集时,可通过HuggingFace的datasets库进行加载。例如,加载特定任务如harness_winogrande_5的最新结果,只需指定配置名称和分割为“train”。如需访问历史运行数据,则可将分割参数设置为对应的时间戳字符串。这种设计使得研究者能够灵活地获取模型在不同时间点、不同任务上的详细评估结果,便于进行深入的性能分析。
背景与挑战
背景概述
Open LLM Leaderboard 是 Hugging Face 社区于2023年发起的一项大规模语言模型评估基准,旨在系统性地衡量开源大语言模型在多样化任务上的表现。该数据集记录了模型 ehartford/WizardLM-30B-Uncensored 的评估运行细节,由 Hugging Face 团队的核心研究员 Clémentine Fourrier 等人主导创建。其核心研究问题在于通过标准化、可复现的流程,揭示未经审查的 WizardLM 模型在常识推理、数学解题、阅读理解及多学科知识等维度的真实能力。该数据集不仅为开发者提供了细粒度的模型性能剖析,更推动了开源社区对模型安全性与能力边界的深入探讨,在 LLM 评估领域具有重要的参考价值。
当前挑战
该数据集所面临的挑战首先体现在领域问题的复杂性上:WizardLM-30B-Uncensored 需要应对从 Winogrande 的代词消歧到 GSM8K 的数学推理、DROP 的文本理解及 MMLU 的跨学科知识等异构任务,这对模型的泛化能力提出了极高要求。其次,构建过程中面临多重技术挑战:评估需覆盖64个不同配置,每个配置对应特定任务与采样参数,确保各次运行结果的可比性;同时,数据存储采用时间戳分割管理,最新结果与历史版本需通过统一接口访问,对数据版本控制与一致性维护构成了严峻考验。
常用场景
经典使用场景
在大型语言模型评估领域,该数据集作为Open LLM Leaderboard的评测结果存储库,为研究者提供了对WizardLM-30B-Uncensored模型在多项基准任务上的细粒度性能剖析。其经典使用场景在于通过加载特定配置(如harness_winogrande_5)与时间戳分割,复现模型在ARC挑战、HellaSwag常识推理、MMLU多学科知识等任务中的精确表现,从而支持模型能力的横向对比与纵向追踪。
实际应用
在实际应用中,该数据集为模型选型与部署提供了量化依据。开发团队可依据WizardLM-30B-Uncensored在Winogrande指代消解(74.35%准确率)等任务上的表现,判断其是否适用于需要常识推理的对话系统、智能客服或教育辅助工具。同时,评测结果的公开性降低了企业评估开源模型的门槛,加速了从研究到产品的转化。
衍生相关工作
该数据集衍生了一系列关于模型评测方法论与性能优化的经典工作。研究者基于其存储的细粒度结果,提出了针对特定任务(如DROP的精确匹配分数仅17.6%)的微调策略,催生了诸如指令调优、少样本学习改进等方向。此外,该数据集的架构被后续Open LLM Leaderboard评测所沿用,成为评估新模型(如Llama、Falcon系列)的参照基准,推动了社区对模型泛化能力的系统化探索。
以上内容由遇见数据集搜集并总结生成
二维码
社区交流群
二维码
科研交流群
商业服务