five

open-llm-leaderboard-old/details_dfurman__llama-2-13b-dolphin-peft

收藏
Hugging Face2023-12-03 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_dfurman__llama-2-13b-dolphin-peft
下载链接
链接失效反馈
官方服务:
资源简介:
--- pretty_name: Evaluation run of dfurman/llama-2-13b-dolphin-peft dataset_summary: "Dataset automatically created during the evaluation run of model\ \ [dfurman/llama-2-13b-dolphin-peft](https://huggingface.co/dfurman/llama-2-13b-dolphin-peft)\ \ 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 5 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 aggregated 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_dfurman__llama-2-13b-dolphin-peft\"\ ,\n\t\"harness_gsm8k_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\nThese\ \ are the [latest results from run 2023-12-03T19:52:57.609518](https://huggingface.co/datasets/open-llm-leaderboard/details_dfurman__llama-2-13b-dolphin-peft/blob/main/results_2023-12-03T19-52-57.609518.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 \"acc\": 0.0,\n \"\ acc_stderr\": 0.0\n },\n \"harness|gsm8k|5\": {\n \"acc\": 0.0,\n \ \ \"acc_stderr\": 0.0\n }\n}\n```" repo_url: https://huggingface.co/dfurman/llama-2-13b-dolphin-peft 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_08_04T15_17_20.418762 path: - '**/details_harness|arc:challenge|25_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|arc:challenge|25_2023-08-04T15:17:20.418762.parquet' - config_name: harness_drop_3 data_files: - split: 2023_09_18T05_24_57.817277 path: - '**/details_harness|drop|3_2023-09-18T05-24-57.817277.parquet' - split: 2023_09_22T23_09_33.744166 path: - '**/details_harness|drop|3_2023-09-22T23-09-33.744166.parquet' - split: latest path: - '**/details_harness|drop|3_2023-09-22T23-09-33.744166.parquet' - config_name: harness_gsm8k_5 data_files: - split: 2023_09_18T05_24_57.817277 path: - '**/details_harness|gsm8k|5_2023-09-18T05-24-57.817277.parquet' - split: 2023_09_22T23_09_33.744166 path: - '**/details_harness|gsm8k|5_2023-09-22T23-09-33.744166.parquet' - split: 2023_12_03T19_26_24.364413 path: - '**/details_harness|gsm8k|5_2023-12-03T19-26-24.364413.parquet' - split: 2023_12_03T19_52_57.609518 path: - '**/details_harness|gsm8k|5_2023-12-03T19-52-57.609518.parquet' - split: latest path: - '**/details_harness|gsm8k|5_2023-12-03T19-52-57.609518.parquet' - config_name: harness_hellaswag_10 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hellaswag|10_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hellaswag|10_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-international_law|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-management|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-marketing|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-sociology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-virology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-international_law|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-management|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-marketing|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-sociology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-virology|5_2023-08-04T15:17:20.418762.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_abstract_algebra_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_anatomy_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-anatomy|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-anatomy|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_astronomy_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-astronomy|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-astronomy|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_business_ethics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-business_ethics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-business_ethics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_clinical_knowledge_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_biology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_biology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_biology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_chemistry_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_chemistry|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_chemistry|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_computer_science_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_computer_science|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_computer_science|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_mathematics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_mathematics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_mathematics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_medicine_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_medicine|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_medicine|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_college_physics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-college_physics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_physics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_computer_security_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-computer_security|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-computer_security|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_conceptual_physics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_econometrics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-econometrics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-econometrics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_electrical_engineering_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_elementary_mathematics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_formal_logic_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-formal_logic|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-formal_logic|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_global_facts_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-global_facts|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-global_facts|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_biology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_biology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_biology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_chemistry_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_computer_science_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_european_history_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_geography_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_geography|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_geography|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_government_and_politics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_macroeconomics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_mathematics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_microeconomics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_physics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_physics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_physics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_psychology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_statistics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_us_history_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_high_school_world_history_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_human_aging_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-human_aging|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_aging|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_human_sexuality_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-human_sexuality|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_sexuality|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_international_law_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-international_law|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-international_law|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_jurisprudence_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-jurisprudence|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-jurisprudence|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_logical_fallacies_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_machine_learning_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-machine_learning|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-machine_learning|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_management_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-management|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-management|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_marketing_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-marketing|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-marketing|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_medical_genetics_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-medical_genetics|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-medical_genetics|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_miscellaneous_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-miscellaneous|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-miscellaneous|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_moral_disputes_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-moral_disputes|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_disputes|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_moral_scenarios_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_nutrition_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-nutrition|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-nutrition|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_philosophy_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-philosophy|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-philosophy|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_prehistory_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-prehistory|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-prehistory|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_professional_accounting_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-professional_accounting|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_accounting|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_professional_law_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-professional_law|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_law|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_professional_medicine_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-professional_medicine|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_medicine|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_professional_psychology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-professional_psychology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_psychology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_public_relations_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-public_relations|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-public_relations|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_security_studies_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-security_studies|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-security_studies|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_sociology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-sociology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-sociology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_us_foreign_policy_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_virology_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-virology|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-virology|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_hendrycksTest_world_religions_5 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|hendrycksTest-world_religions|5_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|hendrycksTest-world_religions|5_2023-08-04T15:17:20.418762.parquet' - config_name: harness_truthfulqa_mc_0 data_files: - split: 2023_08_04T15_17_20.418762 path: - '**/details_harness|truthfulqa:mc|0_2023-08-04T15:17:20.418762.parquet' - split: latest path: - '**/details_harness|truthfulqa:mc|0_2023-08-04T15:17:20.418762.parquet' - config_name: harness_winogrande_5 data_files: - split: 2023_09_18T05_24_57.817277 path: - '**/details_harness|winogrande|5_2023-09-18T05-24-57.817277.parquet' - split: 2023_09_22T23_09_33.744166 path: - '**/details_harness|winogrande|5_2023-09-22T23-09-33.744166.parquet' - split: latest path: - '**/details_harness|winogrande|5_2023-09-22T23-09-33.744166.parquet' - config_name: results data_files: - split: 2023_08_04T15_17_20.418762 path: - results_2023-08-04T15:17:20.418762.parquet - split: 2023_09_18T05_24_57.817277 path: - results_2023-09-18T05-24-57.817277.parquet - split: 2023_09_22T23_09_33.744166 path: - results_2023-09-22T23-09-33.744166.parquet - split: 2023_12_03T19_26_24.364413 path: - results_2023-12-03T19-26-24.364413.parquet - split: 2023_12_03T19_52_57.609518 path: - results_2023-12-03T19-52-57.609518.parquet - split: latest path: - results_2023-12-03T19-52-57.609518.parquet --- # Dataset Card for Evaluation run of dfurman/llama-2-13b-dolphin-peft ## Dataset Description - **Homepage:** - **Repository:** https://huggingface.co/dfurman/llama-2-13b-dolphin-peft - **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 [dfurman/llama-2-13b-dolphin-peft](https://huggingface.co/dfurman/llama-2-13b-dolphin-peft) 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 5 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 aggregated 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_dfurman__llama-2-13b-dolphin-peft", "harness_gsm8k_5", split="train") ``` ## Latest results These are the [latest results from run 2023-12-03T19:52:57.609518](https://huggingface.co/datasets/open-llm-leaderboard/details_dfurman__llama-2-13b-dolphin-peft/blob/main/results_2023-12-03T19-52-57.609518.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": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 } } ``` ### 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

该数据集是在评估模型 dfurman/llama-2-13b-dolphin-peftOpen LLM Leaderboard 上的运行过程中自动创建的。

数据集组成

  • 数据集包含 64 个配置,每个配置对应一个评估任务。
  • 数据集由 5 次运行创建,每次运行在每个配置中都有一个特定的分割,分割名称使用运行的时间戳。
  • "train" 分割始终指向最新的结果。
  • 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示在 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_dfurman__llama-2-13b-dolphin-peft", "harness_gsm8k_5", split="train")

最新结果

以下是 最新结果来自 run 2023-12-03T19:52:57.609518 的示例:

python { "all": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 } }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_08_04T15_17_20.418762
      • path: **/details_harness|arc:challenge|25_2023-08-04T15:17:20.418762.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-08-04T15:17:20.418762.parquet
  • config_name: harness_drop_3

    • split: 2023_09_18T05_24_57.817277
      • path: **/details_harness|drop|3_2023-09-18T05-24-57.817277.parquet
    • split: 2023_09_22T23_09_33.744166
      • path: **/details_harness|drop|3_2023-09-22T23-09-33.744166.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-09-22T23-09-33.744166.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_09_18T05_24_57.817277
      • path: **/details_harness|gsm8k|5_2023-09-18T05-24-57.817277.parquet
    • split: 2023_09_22T23_09_33.744166
      • path: **/details_harness|gsm8k|5_2023-09-22T23-09-33.744166.parquet
    • split: 2023_12_03T19_26_24.364413
      • path: **/details_harness|gsm8k|5_2023-12-03T19-26-24.364413.parquet
    • split: 2023_12_03T19_52_57.609518
      • path: **/details_harness|gsm8k|5_2023-12-03T19-52-57.609518.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-12-03T19-52-57.609518.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_08_04T15_17_20.418762
      • path: **/details_harness|hellaswag|10_2023-08-04T15:17:20.418762.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-08-04T15:17:20.418762.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_08_04T15_17_20.418762
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-04T15:17:20.418762.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-04T15:17:2
搜集汇总
数据集介绍
main_image_url
构建方式
在大型语言模型性能评估领域,Open LLM Leaderboard 作为权威评测平台,为模型提供标准化测试。该数据集基于对 dfurman/llama-2-13b-dolphin-peft 模型的评估运行自动生成,涵盖 64 个配置,每个配置对应一项评测任务。数据来源于 5 次独立运行,每次运行的结果以时间戳命名作为独立分割(split),而‘train’分割始终指向最新一次运行的评测结果。此外,增设‘results’配置用于存储所有运行的聚合指标,以便在 Leaderboard 上计算和展示综合得分。数据集以 Parquet 格式存储,确保高效的数据读取与处理。
使用方法
使用该数据集时,可通过 Hugging Face Datasets 库加载。例如,调用 load_dataset 函数并指定数据集名称及目标配置(如‘harness_gsm8k_5’),再通过参数 split='train' 获取最新评测结果。若需分析历史数据,可将 split 参数设为对应时间戳(如‘2023_12_03T19_52_57.609518’)以加载特定运行的细节。对于多任务综合评估,可加载‘results’配置直接获取聚合指标。数据以 Parquet 格式存储,支持高效的列式读取,适合在大规模分析场景下快速处理。
背景与挑战
背景概述
在大规模语言模型(LLM)蓬勃发展的浪潮中,如何公正、系统地评估模型性能成为学界与工业界共同关注的焦点。Open LLM Leaderboard由Hugging Face团队于2023年创建,旨在为开源社区提供一个标准化的模型评估平台。该数据集正是针对dfurman/llama-2-13b-dolphin-peft模型在排行榜上的自动评估记录,由Clementine等人维护,核心研究问题在于通过多任务(如GSM8K数学推理、HellaSwag常识推理、MMLU多学科知识等)的细粒度评测,揭示微调后的Llama-2-13B模型在不同能力维度上的表现。该数据集不仅为模型开发者提供了透明、可复现的基准,更推动了开源LLM评估范式的规范化,对后续模型优化与社区协作产生了深远影响。
当前挑战
该数据集所解决的领域问题在于,LLM性能评估长期面临任务单一、指标不统一、结果难以复现等困境。具体挑战包括:1)多任务覆盖的全面性——需同时处理数学推理(GSM8K)、常识推理(HellaSwag)、对抗性问答(ARC)及跨学科知识(MMLU)等异质任务,确保评估维度不偏颇;2)评估过程的动态性——模型在多次运行中可能产生波动,数据集需通过时间戳分割记录每次运行结果,以捕捉性能的稳定性和变异;3)构建工程中的标准化难题——将不同任务的结果以统一的Parquet格式存储,并设计64个配置以对应每个任务的细粒度指标,同时维护“latest”分割指向最新结果,这对数据流水线的鲁棒性和可扩展性提出了严苛要求。
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的自动生成产物,核心用途在于系统化记录特定模型(如dfurman/llama-2-13b-dolphin-peft)在多项标准化任务上的细粒度评测结果。其覆盖ARC挑战赛、HellaSwag常识推理、GSM8K数学问题求解及涵盖57个学科的MMLU基准测试等经典场景,为研究者提供了从单一任务到多维度聚合性能的完整评估框架。通过按时间戳组织的split结构,该数据集支持纵向追踪模型版本迭代过程中的性能演化轨迹,成为对比不同训练策略(如PEFT微调)效果的标准化参照系。
解决学术问题
该数据集着力破解大模型评估中普遍存在的可复现性危机与基准碎片化困境。通过将每次评测的原始样本级结果封装为parquet格式的标准化配置,研究者得以精准定位模型在特定子任务(如抽象代数或医学遗传学)上的知识盲区,从而超越简单聚合指标的表象。其多轮次运行记录机制有效消解了单次评测的随机波动干扰,为分析模型鲁棒性提供了统计基础。这种细粒度、版本可控的评估范式,实质性地推动了从‘模型性能排名’向‘能力图谱诊断’的认知跃迁,为后续模型改进提供了可溯源的决策依据。
实际应用
在实际产业部署中,该数据集支撑着模型选型与风险控制的决策链。企业技术团队可借助其多任务评测结果,针对特定业务场景(如客服系统的数学推理或医疗问答的知识广度)筛选最优基座模型。数据集的时间序列特性使开发人员能够量化微调操作对模型能力的潜在影响——例如检测参数高效微调是否导致常识推理能力的灾难性遗忘。此外,其聚合结果配置可直接服务于模型合规性审计,通过对比不同版本在MMLU道德情景测试中的表现,评估模型在敏感领域的部署风险。
数据集最近研究
最新研究方向
在大语言模型迅猛发展的浪潮中,评估基准的构建与模型能力的量化成为推动技术进步的关键环节。该数据集记录了dfurman/llama-2-13b-dolphin-peft模型在Open LLM Leaderboard上的完整评估轨迹,涵盖了ARC挑战、HellaSwag常识推理、GSM8K数学推理以及涵盖57个学科的MMLU基准测试等前沿任务。这些评估不仅揭示了模型在零样本推理、多步数学求解和跨领域知识掌握方面的真实水平,还通过多次运行的时间戳分片,为研究模型迭代中的性能波动与稳定性提供了独特视角。作为开源社区标准化评估体系的重要组成部分,该数据集对促进大模型公平比较、识别模型短板以及引导后续微调优化具有深远意义,尤其在提升模型在复杂推理与专业领域知识上的表现方面,为研究者提供了宝贵的实证基础。
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