open-llm-leaderboard/details_DUAL-GPO-2__phi-2-ipo-test-iter-0

--- pretty_name: Evaluation run of DUAL-GPO-2/phi-2-ipo-test-iter-0 dataset_summary: "Dataset automatically created during the evaluation run of model\ \ [DUAL-GPO-2/phi-2-ipo-test-iter-0](https://huggingface.co/DUAL-GPO-2/phi-2-ipo-test-iter-0)\ \ on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).\n\ \nThe dataset is composed of 63 configuration, each one coresponding to one of the\ \ evaluated task.\n\nThe dataset has been created from 1 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_DUAL-GPO-2__phi-2-ipo-test-iter-0\"\ ,\n\t\"harness_winogrande_5\",\n\tsplit=\"train\")\n```\n\n## Latest results\n\n\ These are the [latest results from run 2024-04-24T04:35:58.978617](https://huggingface.co/datasets/open-llm-leaderboard/details_DUAL-GPO-2__phi-2-ipo-test-iter-0/blob/main/results_2024-04-24T04-35-58.978617.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\": 0.5785172855838389,\n\ \ \"acc_stderr\": 0.03376764130805383,\n \"acc_norm\": 0.5799636790346963,\n\ \ \"acc_norm_stderr\": 0.03445645379045714,\n \"mc1\": 0.30966952264381886,\n\ \ \"mc1_stderr\": 0.016185744355144912,\n \"mc2\": 0.4366024674450443,\n\ \ \"mc2_stderr\": 0.01515509288566976\n },\n \"harness|arc:challenge|25\"\ : {\n \"acc\": 0.5767918088737202,\n \"acc_stderr\": 0.01443803622084803,\n\ \ \"acc_norm\": 0.6075085324232082,\n \"acc_norm_stderr\": 0.014269634635670714\n\ \ },\n \"harness|hellaswag|10\": {\n \"acc\": 0.5650268870742879,\n\ \ \"acc_stderr\": 0.004947402907996252,\n \"acc_norm\": 0.7493527185819557,\n\ \ \"acc_norm_stderr\": 0.004325000473328605\n },\n \"harness|hendrycksTest-abstract_algebra|5\"\ : {\n \"acc\": 0.29,\n \"acc_stderr\": 0.045604802157206845,\n \ \ \"acc_norm\": 0.29,\n \"acc_norm_stderr\": 0.045604802157206845\n \ \ },\n \"harness|hendrycksTest-anatomy|5\": {\n \"acc\": 0.43703703703703706,\n\ \ \"acc_stderr\": 0.042849586397533994,\n \"acc_norm\": 0.43703703703703706,\n\ \ \"acc_norm_stderr\": 0.042849586397533994\n },\n \"harness|hendrycksTest-astronomy|5\"\ : {\n \"acc\": 0.5986842105263158,\n \"acc_stderr\": 0.039889037033362836,\n\ \ \"acc_norm\": 0.5986842105263158,\n \"acc_norm_stderr\": 0.039889037033362836\n\ \ },\n \"harness|hendrycksTest-business_ethics|5\": {\n \"acc\": 0.58,\n\ \ \"acc_stderr\": 0.049604496374885836,\n \"acc_norm\": 0.58,\n \ \ \"acc_norm_stderr\": 0.049604496374885836\n },\n \"harness|hendrycksTest-clinical_knowledge|5\"\ : {\n \"acc\": 0.6188679245283019,\n \"acc_stderr\": 0.02989060968628664,\n\ \ \"acc_norm\": 0.6188679245283019,\n \"acc_norm_stderr\": 0.02989060968628664\n\ \ },\n \"harness|hendrycksTest-college_biology|5\": {\n \"acc\": 0.6527777777777778,\n\ \ \"acc_stderr\": 0.039812405437178615,\n \"acc_norm\": 0.6527777777777778,\n\ \ \"acc_norm_stderr\": 0.039812405437178615\n },\n \"harness|hendrycksTest-college_chemistry|5\"\ : {\n \"acc\": 0.38,\n \"acc_stderr\": 0.048783173121456316,\n \ \ \"acc_norm\": 0.38,\n \"acc_norm_stderr\": 0.048783173121456316\n \ \ },\n \"harness|hendrycksTest-college_computer_science|5\": {\n \"\ acc\": 0.42,\n \"acc_stderr\": 0.049604496374885836,\n \"acc_norm\"\ : 0.42,\n \"acc_norm_stderr\": 0.049604496374885836\n },\n \"harness|hendrycksTest-college_mathematics|5\"\ : {\n \"acc\": 0.38,\n \"acc_stderr\": 0.04878317312145633,\n \ \ \"acc_norm\": 0.38,\n \"acc_norm_stderr\": 0.04878317312145633\n \ \ },\n \"harness|hendrycksTest-college_medicine|5\": {\n \"acc\": 0.5838150289017341,\n\ \ \"acc_stderr\": 0.03758517775404947,\n \"acc_norm\": 0.5838150289017341,\n\ \ \"acc_norm_stderr\": 0.03758517775404947\n },\n \"harness|hendrycksTest-college_physics|5\"\ : {\n \"acc\": 0.3333333333333333,\n \"acc_stderr\": 0.04690650298201943,\n\ \ \"acc_norm\": 0.3333333333333333,\n \"acc_norm_stderr\": 0.04690650298201943\n\ \ },\n \"harness|hendrycksTest-computer_security|5\": {\n \"acc\":\ \ 0.74,\n \"acc_stderr\": 0.04408440022768078,\n \"acc_norm\": 0.74,\n\ \ \"acc_norm_stderr\": 0.04408440022768078\n },\n \"harness|hendrycksTest-conceptual_physics|5\"\ : {\n \"acc\": 0.5191489361702127,\n \"acc_stderr\": 0.032662042990646796,\n\ \ \"acc_norm\": 0.5191489361702127,\n \"acc_norm_stderr\": 0.032662042990646796\n\ \ },\n \"harness|hendrycksTest-econometrics|5\": {\n \"acc\": 0.3684210526315789,\n\ \ \"acc_stderr\": 0.04537815354939392,\n \"acc_norm\": 0.3684210526315789,\n\ \ \"acc_norm_stderr\": 0.04537815354939392\n },\n \"harness|hendrycksTest-electrical_engineering|5\"\ : {\n \"acc\": 0.5310344827586206,\n \"acc_stderr\": 0.04158632762097828,\n\ \ \"acc_norm\": 0.5310344827586206,\n \"acc_norm_stderr\": 0.04158632762097828\n\ \ },\n \"harness|hendrycksTest-elementary_mathematics|5\": {\n \"acc\"\ : 0.4365079365079365,\n \"acc_stderr\": 0.025542846817400492,\n \"\ acc_norm\": 0.4365079365079365,\n \"acc_norm_stderr\": 0.025542846817400492\n\ \ },\n \"harness|hendrycksTest-formal_logic|5\": {\n \"acc\": 0.36507936507936506,\n\ \ \"acc_stderr\": 0.04306241259127153,\n \"acc_norm\": 0.36507936507936506,\n\ \ \"acc_norm_stderr\": 0.04306241259127153\n },\n \"harness|hendrycksTest-global_facts|5\"\ : {\n \"acc\": 0.38,\n \"acc_stderr\": 0.048783173121456316,\n \ \ \"acc_norm\": 0.38,\n \"acc_norm_stderr\": 0.048783173121456316\n \ \ },\n \"harness|hendrycksTest-high_school_biology|5\": {\n \"acc\"\ : 0.6903225806451613,\n \"acc_stderr\": 0.026302774983517418,\n \"\ acc_norm\": 0.6903225806451613,\n \"acc_norm_stderr\": 0.026302774983517418\n\ \ },\n \"harness|hendrycksTest-high_school_chemistry|5\": {\n \"acc\"\ : 0.4876847290640394,\n \"acc_stderr\": 0.035169204442208966,\n \"\ acc_norm\": 0.4876847290640394,\n \"acc_norm_stderr\": 0.035169204442208966\n\ \ },\n \"harness|hendrycksTest-high_school_computer_science|5\": {\n \ \ \"acc\": 0.66,\n \"acc_stderr\": 0.04760952285695237,\n \"acc_norm\"\ : 0.66,\n \"acc_norm_stderr\": 0.04760952285695237\n },\n \"harness|hendrycksTest-high_school_european_history|5\"\ : {\n \"acc\": 0.6363636363636364,\n \"acc_stderr\": 0.037563357751878974,\n\ \ \"acc_norm\": 0.6363636363636364,\n \"acc_norm_stderr\": 0.037563357751878974\n\ \ },\n \"harness|hendrycksTest-high_school_geography|5\": {\n \"acc\"\ : 0.7323232323232324,\n \"acc_stderr\": 0.03154449888270285,\n \"\ acc_norm\": 0.7323232323232324,\n \"acc_norm_stderr\": 0.03154449888270285\n\ \ },\n \"harness|hendrycksTest-high_school_government_and_politics|5\": {\n\ \ \"acc\": 0.8082901554404145,\n \"acc_stderr\": 0.028408953626245282,\n\ \ \"acc_norm\": 0.8082901554404145,\n \"acc_norm_stderr\": 0.028408953626245282\n\ \ },\n \"harness|hendrycksTest-high_school_macroeconomics|5\": {\n \ \ \"acc\": 0.5692307692307692,\n \"acc_stderr\": 0.025106820660539753,\n\ \ \"acc_norm\": 0.5692307692307692,\n \"acc_norm_stderr\": 0.025106820660539753\n\ \ },\n \"harness|hendrycksTest-high_school_mathematics|5\": {\n \"\ acc\": 0.3296296296296296,\n \"acc_stderr\": 0.02866120111652459,\n \ \ \"acc_norm\": 0.3296296296296296,\n \"acc_norm_stderr\": 0.02866120111652459\n\ \ },\n \"harness|hendrycksTest-high_school_microeconomics|5\": {\n \ \ \"acc\": 0.5966386554621849,\n \"acc_stderr\": 0.031866081214088314,\n\ \ \"acc_norm\": 0.5966386554621849,\n \"acc_norm_stderr\": 0.031866081214088314\n\ \ },\n \"harness|hendrycksTest-high_school_physics|5\": {\n \"acc\"\ : 0.39072847682119205,\n \"acc_stderr\": 0.03983798306659806,\n \"\ acc_norm\": 0.39072847682119205,\n \"acc_norm_stderr\": 0.03983798306659806\n\ \ },\n \"harness|hendrycksTest-high_school_psychology|5\": {\n \"acc\"\ : 0.7908256880733945,\n \"acc_stderr\": 0.017437937173343233,\n \"\ acc_norm\": 0.7908256880733945,\n \"acc_norm_stderr\": 0.017437937173343233\n\ \ },\n \"harness|hendrycksTest-high_school_statistics|5\": {\n \"acc\"\ : 0.4722222222222222,\n \"acc_stderr\": 0.0340470532865388,\n \"acc_norm\"\ : 0.4722222222222222,\n \"acc_norm_stderr\": 0.0340470532865388\n },\n\ \ \"harness|hendrycksTest-high_school_us_history|5\": {\n \"acc\": 0.6470588235294118,\n\ \ \"acc_stderr\": 0.03354092437591518,\n \"acc_norm\": 0.6470588235294118,\n\ \ \"acc_norm_stderr\": 0.03354092437591518\n },\n \"harness|hendrycksTest-high_school_world_history|5\"\ : {\n \"acc\": 0.7426160337552743,\n \"acc_stderr\": 0.02845882099146029,\n\ \ \"acc_norm\": 0.7426160337552743,\n \"acc_norm_stderr\": 0.02845882099146029\n\ \ },\n \"harness|hendrycksTest-human_aging|5\": {\n \"acc\": 0.6412556053811659,\n\ \ \"acc_stderr\": 0.032190792004199956,\n \"acc_norm\": 0.6412556053811659,\n\ \ \"acc_norm_stderr\": 0.032190792004199956\n },\n \"harness|hendrycksTest-human_sexuality|5\"\ : {\n \"acc\": 0.6946564885496184,\n \"acc_stderr\": 0.04039314978724561,\n\ \ \"acc_norm\": 0.6946564885496184,\n \"acc_norm_stderr\": 0.04039314978724561\n\ \ },\n \"harness|hendrycksTest-international_law|5\": {\n \"acc\":\ \ 0.7520661157024794,\n \"acc_stderr\": 0.03941897526516302,\n \"\ acc_norm\": 0.7520661157024794,\n \"acc_norm_stderr\": 0.03941897526516302\n\ \ },\n \"harness|hendrycksTest-jurisprudence|5\": {\n \"acc\": 0.6851851851851852,\n\ \ \"acc_stderr\": 0.04489931073591312,\n \"acc_norm\": 0.6851851851851852,\n\ \ \"acc_norm_stderr\": 0.04489931073591312\n },\n \"harness|hendrycksTest-logical_fallacies|5\"\ : {\n \"acc\": 0.7607361963190185,\n \"acc_stderr\": 0.0335195387952127,\n\ \ \"acc_norm\": 0.7607361963190185,\n \"acc_norm_stderr\": 0.0335195387952127\n\ \ },\n \"harness|hendrycksTest-machine_learning|5\": {\n \"acc\": 0.4732142857142857,\n\ \ \"acc_stderr\": 0.047389751192741546,\n \"acc_norm\": 0.4732142857142857,\n\ \ \"acc_norm_stderr\": 0.047389751192741546\n },\n \"harness|hendrycksTest-management|5\"\ : {\n \"acc\": 0.7281553398058253,\n \"acc_stderr\": 0.044052680241409216,\n\ \ \"acc_norm\": 0.7281553398058253,\n \"acc_norm_stderr\": 0.044052680241409216\n\ \ },\n \"harness|hendrycksTest-marketing|5\": {\n \"acc\": 0.8034188034188035,\n\ \ \"acc_stderr\": 0.02603538609895129,\n \"acc_norm\": 0.8034188034188035,\n\ \ \"acc_norm_stderr\": 0.02603538609895129\n },\n \"harness|hendrycksTest-medical_genetics|5\"\ : {\n \"acc\": 0.64,\n \"acc_stderr\": 0.04824181513244218,\n \ \ \"acc_norm\": 0.64,\n \"acc_norm_stderr\": 0.04824181513244218\n \ \ },\n \"harness|hendrycksTest-miscellaneous|5\": {\n \"acc\": 0.6909323116219668,\n\ \ \"acc_stderr\": 0.016524988919702204,\n \"acc_norm\": 0.6909323116219668,\n\ \ \"acc_norm_stderr\": 0.016524988919702204\n },\n \"harness|hendrycksTest-moral_disputes|5\"\ : {\n \"acc\": 0.6647398843930635,\n \"acc_stderr\": 0.02541600377316554,\n\ \ \"acc_norm\": 0.6647398843930635,\n \"acc_norm_stderr\": 0.02541600377316554\n\ \ },\n \"harness|hendrycksTest-moral_scenarios|5\": {\n \"acc\": 0.29608938547486036,\n\ \ \"acc_stderr\": 0.015268677317602286,\n \"acc_norm\": 0.29608938547486036,\n\ \ \"acc_norm_stderr\": 0.015268677317602286\n },\n \"harness|hendrycksTest-nutrition|5\"\ : {\n \"acc\": 0.6143790849673203,\n \"acc_stderr\": 0.027870745278290275,\n\ \ \"acc_norm\": 0.6143790849673203,\n \"acc_norm_stderr\": 0.027870745278290275\n\ \ },\n \"harness|hendrycksTest-philosophy|5\": {\n \"acc\": 0.6141479099678456,\n\ \ \"acc_stderr\": 0.02764814959975147,\n \"acc_norm\": 0.6141479099678456,\n\ \ \"acc_norm_stderr\": 0.02764814959975147\n },\n \"harness|hendrycksTest-prehistory|5\"\ : {\n \"acc\": 0.6327160493827161,\n \"acc_stderr\": 0.026822801759507887,\n\ \ \"acc_norm\": 0.6327160493827161,\n \"acc_norm_stderr\": 0.026822801759507887\n\ \ },\n \"harness|hendrycksTest-professional_accounting|5\": {\n \"\ acc\": 0.450354609929078,\n \"acc_stderr\": 0.029680105565029036,\n \ \ \"acc_norm\": 0.450354609929078,\n \"acc_norm_stderr\": 0.029680105565029036\n\ \ },\n \"harness|hendrycksTest-professional_law|5\": {\n \"acc\": 0.43546284224250326,\n\ \ \"acc_stderr\": 0.012663412101248332,\n \"acc_norm\": 0.43546284224250326,\n\ \ \"acc_norm_stderr\": 0.012663412101248332\n },\n \"harness|hendrycksTest-professional_medicine|5\"\ : {\n \"acc\": 0.47794117647058826,\n \"acc_stderr\": 0.030343264224213528,\n\ \ \"acc_norm\": 0.47794117647058826,\n \"acc_norm_stderr\": 0.030343264224213528\n\ \ },\n \"harness|hendrycksTest-professional_psychology|5\": {\n \"\ acc\": 0.5571895424836601,\n \"acc_stderr\": 0.02009508315457735,\n \ \ \"acc_norm\": 0.5571895424836601,\n \"acc_norm_stderr\": 0.02009508315457735\n\ \ },\n \"harness|hendrycksTest-public_relations|5\": {\n \"acc\": 0.6636363636363637,\n\ \ \"acc_stderr\": 0.04525393596302505,\n \"acc_norm\": 0.6636363636363637,\n\ \ \"acc_norm_stderr\": 0.04525393596302505\n },\n \"harness|hendrycksTest-security_studies|5\"\ : {\n \"acc\": 0.7020408163265306,\n \"acc_stderr\": 0.02927956741106568,\n\ \ \"acc_norm\": 0.7020408163265306,\n \"acc_norm_stderr\": 0.02927956741106568\n\ \ },\n \"harness|hendrycksTest-sociology|5\": {\n \"acc\": 0.8059701492537313,\n\ \ \"acc_stderr\": 0.027962677604768928,\n \"acc_norm\": 0.8059701492537313,\n\ \ \"acc_norm_stderr\": 0.027962677604768928\n },\n \"harness|hendrycksTest-us_foreign_policy|5\"\ : {\n \"acc\": 0.76,\n \"acc_stderr\": 0.042923469599092816,\n \ \ \"acc_norm\": 0.76,\n \"acc_norm_stderr\": 0.042923469599092816\n \ \ },\n \"harness|hendrycksTest-virology|5\": {\n \"acc\": 0.463855421686747,\n\ \ \"acc_stderr\": 0.03882310850890593,\n \"acc_norm\": 0.463855421686747,\n\ \ \"acc_norm_stderr\": 0.03882310850890593\n },\n \"harness|hendrycksTest-world_religions|5\"\ : {\n \"acc\": 0.6842105263157895,\n \"acc_stderr\": 0.03565079670708311,\n\ \ \"acc_norm\": 0.6842105263157895,\n \"acc_norm_stderr\": 0.03565079670708311\n\ \ },\n \"harness|truthfulqa:mc|0\": {\n \"mc1\": 0.30966952264381886,\n\ \ \"mc1_stderr\": 0.016185744355144912,\n \"mc2\": 0.4366024674450443,\n\ \ \"mc2_stderr\": 0.01515509288566976\n },\n \"harness|winogrande|5\"\ : {\n \"acc\": 0.7348066298342542,\n \"acc_stderr\": 0.012406549466192858\n\ \ },\n \"harness|gsm8k|5\": {\n \"acc\": 0.5519332827899924,\n \ \ \"acc_stderr\": 0.013697992668274516\n }\n}\n```" repo_url: https://huggingface.co/DUAL-GPO-2/phi-2-ipo-test-iter-0 leaderboard_url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard point_of_contact: [e-mail] configs: - config_name: harness_arc_challenge_25 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|arc:challenge|25_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|arc:challenge|25_2024-04-24T04-35-58.978617.parquet' - config_name: harness_gsm8k_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|gsm8k|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|gsm8k|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hellaswag_10 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hellaswag|10_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hellaswag|10_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-international_law|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-management|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-marketing|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-sociology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-virology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-anatomy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-astronomy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-business_ethics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_biology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_chemistry|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_computer_science|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_medicine|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-college_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-computer_security|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-econometrics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-formal_logic|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-global_facts|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_biology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_geography|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_physics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-human_aging|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-human_sexuality|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-international_law|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-jurisprudence|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-machine_learning|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-management|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-marketing|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-medical_genetics|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-miscellaneous|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-moral_disputes|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-nutrition|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-philosophy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-prehistory|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_accounting|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_law|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_medicine|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-professional_psychology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-public_relations|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-security_studies|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-sociology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-us_foreign_policy|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-virology|5_2024-04-24T04-35-58.978617.parquet' - '**/details_harness|hendrycksTest-world_religions|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_abstract_algebra_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_anatomy_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-anatomy|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-anatomy|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_astronomy_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-astronomy|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-astronomy|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_business_ethics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-business_ethics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-business_ethics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_clinical_knowledge_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_biology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_biology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_biology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_chemistry_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_chemistry|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_chemistry|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_computer_science_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_computer_science|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_computer_science|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_mathematics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_mathematics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_mathematics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_medicine_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_medicine|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_medicine|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_college_physics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-college_physics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-college_physics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_computer_security_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-computer_security|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-computer_security|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_conceptual_physics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_econometrics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-econometrics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-econometrics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_electrical_engineering_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_elementary_mathematics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_formal_logic_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-formal_logic|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-formal_logic|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_global_facts_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-global_facts|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-global_facts|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_biology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_biology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_biology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_chemistry_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_computer_science_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_european_history_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_geography_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_geography|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_geography|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_government_and_politics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_macroeconomics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_mathematics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_microeconomics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_physics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_physics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_physics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_psychology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_statistics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_us_history_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_high_school_world_history_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_human_aging_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-human_aging|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_aging|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_human_sexuality_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-human_sexuality|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-human_sexuality|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_international_law_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-international_law|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-international_law|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_jurisprudence_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-jurisprudence|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-jurisprudence|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_logical_fallacies_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_machine_learning_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-machine_learning|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-machine_learning|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_management_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-management|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-management|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_marketing_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-marketing|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-marketing|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_medical_genetics_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-medical_genetics|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-medical_genetics|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_miscellaneous_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-miscellaneous|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-miscellaneous|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_moral_disputes_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-moral_disputes|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_disputes|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_moral_scenarios_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_nutrition_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-nutrition|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-nutrition|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_philosophy_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-philosophy|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-philosophy|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_prehistory_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-prehistory|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-prehistory|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_professional_accounting_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-professional_accounting|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_accounting|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_professional_law_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-professional_law|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_law|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_professional_medicine_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-professional_medicine|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_medicine|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_professional_psychology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-professional_psychology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-professional_psychology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_public_relations_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-public_relations|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-public_relations|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_security_studies_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-security_studies|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-security_studies|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_sociology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-sociology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-sociology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_us_foreign_policy_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-us_foreign_policy|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_virology_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-virology|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-virology|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_hendrycksTest_world_religions_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|hendrycksTest-world_religions|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|hendrycksTest-world_religions|5_2024-04-24T04-35-58.978617.parquet' - config_name: harness_truthfulqa_mc_0 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|truthfulqa:mc|0_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|truthfulqa:mc|0_2024-04-24T04-35-58.978617.parquet' - config_name: harness_winogrande_5 data_files: - split: 2024_04_24T04_35_58.978617 path: - '**/details_harness|winogrande|5_2024-04-24T04-35-58.978617.parquet' - split: latest path: - '**/details_harness|winogrande|5_2024-04-24T04-35-58.978617.parquet' - config_name: results data_files: - split: 2024_04_24T04_35_58.978617 path: - results_2024-04-24T04-35-58.978617.parquet - split: latest path: - results_2024-04-24T04-35-58.978617.parquet --- # Dataset Card for Evaluation run of DUAL-GPO-2/phi-2-ipo-test-iter-0 <!-- Provide a quick summary of the dataset. --> Dataset automatically created during the evaluation run of model [DUAL-GPO-2/phi-2-ipo-test-iter-0](https://huggingface.co/DUAL-GPO-2/phi-2-ipo-test-iter-0) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task. The dataset has been created from 1 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_DUAL-GPO-2__phi-2-ipo-test-iter-0", "harness_winogrande_5", split="train") ``` ## Latest results These are the [latest results from run 2024-04-24T04:35:58.978617](https://huggingface.co/datasets/open-llm-leaderboard/details_DUAL-GPO-2__phi-2-ipo-test-iter-0/blob/main/results_2024-04-24T04-35-58.978617.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": 0.5785172855838389, "acc_stderr": 0.03376764130805383, "acc_norm": 0.5799636790346963, "acc_norm_stderr": 0.03445645379045714, "mc1": 0.30966952264381886, "mc1_stderr": 0.016185744355144912, "mc2": 0.4366024674450443, "mc2_stderr": 0.01515509288566976 }, "harness|arc:challenge|25": { "acc": 0.5767918088737202, "acc_stderr": 0.01443803622084803, "acc_norm": 0.6075085324232082, "acc_norm_stderr": 0.014269634635670714 }, "harness|hellaswag|10": { "acc": 0.5650268870742879, "acc_stderr": 0.004947402907996252, "acc_norm": 0.7493527185819557, "acc_norm_stderr": 0.004325000473328605 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.045604802157206845, "acc_norm": 0.29, "acc_norm_stderr": 0.045604802157206845 }, "harness|hendrycksTest-anatomy|5": { "acc": 0.43703703703703706, "acc_stderr": 0.042849586397533994, "acc_norm": 0.43703703703703706, "acc_norm_stderr": 0.042849586397533994 }, "harness|hendrycksTest-astronomy|5": { "acc": 0.5986842105263158, "acc_stderr": 0.039889037033362836, "acc_norm": 0.5986842105263158, "acc_norm_stderr": 0.039889037033362836 }, "harness|hendrycksTest-business_ethics|5": { "acc": 0.58, "acc_stderr": 0.049604496374885836, "acc_norm": 0.58, "acc_norm_stderr": 0.049604496374885836 }, "harness|hendrycksTest-clinical_knowledge|5": { "acc": 0.6188679245283019, "acc_stderr": 0.02989060968628664, "acc_norm": 0.6188679245283019, "acc_norm_stderr": 0.02989060968628664 }, "harness|hendrycksTest-college_biology|5": { "acc": 0.6527777777777778, "acc_stderr": 0.039812405437178615, "acc_norm": 0.6527777777777778, "acc_norm_stderr": 0.039812405437178615 }, "harness|hendrycksTest-college_chemistry|5": { "acc": 0.38, "acc_stderr": 0.048783173121456316, "acc_norm": 0.38, "acc_norm_stderr": 0.048783173121456316 }, "harness|hendrycksTest-college_computer_science|5": { "acc": 0.42, "acc_stderr": 0.049604496374885836, "acc_norm": 0.42, "acc_norm_stderr": 0.049604496374885836 }, "harness|hendrycksTest-college_mathematics|5": { "acc": 0.38, "acc_stderr": 0.04878317312145633, "acc_norm": 0.38, "acc_norm_stderr": 0.04878317312145633 }, "harness|hendrycksTest-college_medicine|5": { "acc": 0.5838150289017341, "acc_stderr": 0.03758517775404947, "acc_norm": 0.5838150289017341, "acc_norm_stderr": 0.03758517775404947 }, "harness|hendrycksTest-college_physics|5": { "acc": 0.3333333333333333, "acc_stderr": 0.04690650298201943, "acc_norm": 0.3333333333333333, "acc_norm_stderr": 0.04690650298201943 }, "harness|hendrycksTest-computer_security|5": { "acc": 0.74, "acc_stderr": 0.04408440022768078, "acc_norm": 0.74, "acc_norm_stderr": 0.04408440022768078 }, "harness|hendrycksTest-conceptual_physics|5": { "acc": 0.5191489361702127, "acc_stderr": 0.032662042990646796, "acc_norm": 0.5191489361702127, "acc_norm_stderr": 0.032662042990646796 }, "harness|hendrycksTest-econometrics|5": { "acc": 0.3684210526315789, "acc_stderr": 0.04537815354939392, "acc_norm": 0.3684210526315789, "acc_norm_stderr": 0.04537815354939392 }, "harness|hendrycksTest-electrical_engineering|5": { "acc": 0.5310344827586206, "acc_stderr": 0.04158632762097828, "acc_norm": 0.5310344827586206, "acc_norm_stderr": 0.04158632762097828 }, "harness|hendrycksTest-elementary_mathematics|5": { "acc": 0.4365079365079365, "acc_stderr": 0.025542846817400492, "acc_norm": 0.4365079365079365, "acc_norm_stderr": 0.025542846817400492 }, "harness|hendrycksTest-formal_logic|5": { "acc": 0.36507936507936506, "acc_stderr": 0.04306241259127153, "acc_norm": 0.36507936507936506, "acc_norm_stderr": 0.04306241259127153 }, "harness|hendrycksTest-global_facts|5": { "acc": 0.38, "acc_stderr": 0.048783173121456316, "acc_norm": 0.38, "acc_norm_stderr": 0.048783173121456316 }, "harness|hendrycksTest-high_school_biology|5": { "acc": 0.6903225806451613, "acc_stderr": 0.026302774983517418, "acc_norm": 0.6903225806451613, "acc_norm_stderr": 0.026302774983517418 }, "harness|hendrycksTest-high_school_chemistry|5": { "acc": 0.4876847290640394, "acc_stderr": 0.035169204442208966, "acc_norm": 0.4876847290640394, "acc_norm_stderr": 0.035169204442208966 }, "harness|hendrycksTest-high_school_computer_science|5": { "acc": 0.66, "acc_stderr": 0.04760952285695237, "acc_norm": 0.66, "acc_norm_stderr": 0.04760952285695237 }, "harness|hendrycksTest-high_school_european_history|5": { "acc": 0.6363636363636364, "acc_stderr": 0.037563357751878974, "acc_norm": 0.6363636363636364, "acc_norm_stderr": 0.037563357751878974 }, "harness|hendrycksTest-high_school_geography|5": { "acc": 0.7323232323232324, "acc_stderr": 0.03154449888270285, "acc_norm": 0.7323232323232324, "acc_norm_stderr": 0.03154449888270285 }, "harness|hendrycksTest-high_school_government_and_politics|5": { "acc": 0.8082901554404145, "acc_stderr": 0.028408953626245282, "acc_norm": 0.8082901554404145, "acc_norm_stderr": 0.028408953626245282 }, "harness|hendrycksTest-high_school_macroeconomics|5": { "acc": 0.5692307692307692, "acc_stderr": 0.025106820660539753, "acc_norm": 0.5692307692307692, "acc_norm_stderr": 0.025106820660539753 }, "harness|hendrycksTest-high_school_mathematics|5": { "acc": 0.3296296296296296, "acc_stderr": 0.02866120111652459, "acc_norm": 0.3296296296296296, "acc_norm_stderr": 0.02866120111652459 }, "harness|hendrycksTest-high_school_microeconomics|5": { "acc": 0.5966386554621849, "acc_stderr": 0.031866081214088314, "acc_norm": 0.5966386554621849, "acc_norm_stderr": 0.031866081214088314 }, "harness|hendrycksTest-high_school_physics|5": { "acc": 0.39072847682119205, "acc_stderr": 0.03983798306659806, "acc_norm": 0.39072847682119205, "acc_norm_stderr": 0.03983798306659806 }, "harness|hendrycksTest-high_school_psychology|5": { "acc": 0.7908256880733945, "acc_stderr": 0.017437937173343233, "acc_norm": 0.7908256880733945, "acc_norm_stderr": 0.017437937173343233 }, "harness|hendrycksTest-high_school_statistics|5": { "acc": 0.4722222222222222, "acc_stderr": 0.0340470532865388, "acc_norm": 0.4722222222222222, "acc_norm_stderr": 0.0340470532865388 }, "harness|hendrycksTest-high_school_us_history|5": { "acc": 0.6470588235294118, "acc_stderr": 0.03354092437591518, "acc_norm": 0.6470588235294118, "acc_norm_stderr": 0.03354092437591518 }, "harness|hendrycksTest-high_school_world_history|5": { "acc": 0.7426160337552743, "acc_stderr": 0.02845882099146029, "acc_norm": 0.7426160337552743, "acc_norm_stderr": 0.02845882099146029 }, "harness|hendrycksTest-human_aging|5": { "acc": 0.6412556053811659, "acc_stderr": 0.032190792004199956, "acc_norm": 0.6412556053811659, "acc_norm_stderr": 0.032190792004199956 }, "harness|hendrycksTest-human_sexuality|5": { "acc": 0.6946564885496184, "acc_stderr": 0.04039314978724561, "acc_norm": 0.6946564885496184, "acc_norm_stderr": 0.04039314978724561 }, "harness|hendrycksTest-international_law|5": { "acc": 0.7520661157024794, "acc_stderr": 0.03941897526516302, "acc_norm": 0.7520661157024794, "acc_norm_stderr": 0.03941897526516302 }, "harness|hendrycksTest-jurisprudence|5": { "acc": 0.6851851851851852, "acc_stderr": 0.04489931073591312, "acc_norm": 0.6851851851851852, "acc_norm_stderr": 0.04489931073591312 }, "harness|hendrycksTest-logical_fallacies|5": { "acc": 0.7607361963190185, "acc_stderr": 0.0335195387952127, "acc_norm": 0.7607361963190185, "acc_norm_stderr": 0.0335195387952127 }, "harness|hendrycksTest-machine_learning|5": { "acc": 0.4732142857142857, "acc_stderr": 0.047389751192741546, "acc_norm": 0.4732142857142857, "acc_norm_stderr": 0.047389751192741546 }, "harness|hendrycksTest-management|5": { "acc": 0.7281553398058253, "acc_stderr": 0.044052680241409216, "acc_norm": 0.7281553398058253, "acc_norm_stderr": 0.044052680241409216 }, "harness|hendrycksTest-marketing|5": { "acc": 0.8034188034188035, "acc_stderr": 0.02603538609895129, "acc_norm": 0.8034188034188035, "acc_norm_stderr": 0.02603538609895129 }, "harness|hendrycksTest-medical_genetics|5": { "acc": 0.64, "acc_stderr": 0.04824181513244218, "acc_norm": 0.64, "acc_norm_stderr": 0.04824181513244218 }, "harness|hendrycksTest-miscellaneous|5": { "acc": 0.6909323116219668, "acc_stderr": 0.016524988919702204, "acc_norm": 0.6909323116219668, "acc_norm_stderr": 0.016524988919702204 }, "harness|hendrycksTest-moral_disputes|5": { "acc": 0.6647398843930635, "acc_stderr": 0.02541600377316554, "acc_norm": 0.6647398843930635, "acc_norm_stderr": 0.02541600377316554 }, "harness|hendrycksTest-moral_scenarios|5": { "acc": 0.29608938547486036, "acc_stderr": 0.015268677317602286, "acc_norm": 0.29608938547486036, "acc_norm_stderr": 0.015268677317602286 }, "harness|hendrycksTest-nutrition|5": { "acc": 0.6143790849673203, "acc_stderr": 0.027870745278290275, "acc_norm": 0.6143790849673203, "acc_norm_stderr": 0.027870745278290275 }, "harness|hendrycksTest-philosophy|5": { "acc": 0.6141479099678456, "acc_stderr": 0.02764814959975147, "acc_norm": 0.6141479099678456, "acc_norm_stderr": 0.02764814959975147 }, "harness|hendrycksTest-prehistory|5": { "acc": 0.6327160493827161, "acc_stderr": 0.026822801759507887, "acc_norm": 0.6327160493827161, "acc_norm_stderr": 0.026822801759507887 }, "harness|hendrycksTest-professional_accounting|5": { "acc": 0.450354609929078, "acc_stderr": 0.029680105565029036, "acc_norm": 0.450354609929078, "acc_norm_stderr": 0.029680105565029036 }, "harness|hendrycksTest-professional_law|5": { "acc": 0.43546284224250326, "acc_stderr": 0.012663412101248332, "acc_norm": 0.43546284224250326, "acc_norm_stderr": 0.012663412101248332 }, "harness|hendrycksTest-professional_medicine|5": { "acc": 0.47794117647058826, "acc_stderr": 0.030343264224213528, "acc_norm": 0.47794117647058826, "acc_norm_stderr": 0.030343264224213528 }, "harness|hendrycksTest-professional_psychology|5": { "acc": 0.5571895424836601, "acc_stderr": 0.02009508315457735, "acc_norm": 0.5571895424836601, "acc_norm_stderr": 0.02009508315457735 }, "harness|hendrycksTest-public_relations|5": { "acc": 0.6636363636363637, "acc_stderr": 0.04525393596302505, "acc_norm": 0.6636363636363637, "acc_norm_stderr": 0.04525393596302505 }, "harness|hendrycksTest-security_studies|5": { "acc": 0.7020408163265306, "acc_stderr": 0.02927956741106568, "acc_norm": 0.7020408163265306, "acc_norm_stderr": 0.02927956741106568 }, "harness|hendrycksTest-sociology|5": { "acc": 0.8059701492537313, "acc_stderr": 0.027962677604768928, "acc_norm": 0.8059701492537313, "acc_norm_stderr": 0.027962677604768928 }, "harness|hendrycksTest-us_foreign_policy|5": { "acc": 0.76, "acc_stderr": 0.042923469599092816, "acc_norm": 0.76, "acc_norm_stderr": 0.042923469599092816 }, "harness|hendrycksTest-virology|5": { "acc": 0.463855421686747, "acc_stderr": 0.03882310850890593, "acc_norm": 0.463855421686747, "acc_norm_stderr": 0.03882310850890593 }, "harness|hendrycksTest-world_religions|5": { "acc": 0.6842105263157895, "acc_stderr": 0.03565079670708311, "acc_norm": 0.6842105263157895, "acc_norm_stderr": 0.03565079670708311 }, "harness|truthfulqa:mc|0": { "mc1": 0.30966952264381886, "mc1_stderr": 0.016185744355144912, "mc2": 0.4366024674450443, "mc2_stderr": 0.01515509288566976 }, "harness|winogrande|5": { "acc": 0.7348066298342542, "acc_stderr": 0.012406549466192858 }, "harness|gsm8k|5": { "acc": 0.5519332827899924, "acc_stderr": 0.013697992668274516 } } ``` ## Dataset Details ### Dataset Description <!-- Provide a longer summary of what this dataset is. --> - **Curated by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] ### Dataset Sources [optional] <!-- Provide the basic links for the dataset. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the dataset is intended to be used. --> ### Direct Use <!-- This section describes suitable use cases for the dataset. --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. --> [More Information Needed] ## Dataset Structure <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. --> [More Information Needed] ## Dataset Creation ### Curation Rationale <!-- Motivation for the creation of this dataset. --> [More Information Needed] ### Source Data <!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). --> #### Data Collection and Processing <!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. --> [More Information Needed] #### Who are the source data producers? <!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. --> [More Information Needed] ### Annotations [optional] <!-- If the dataset contains annotations which are not part of the initial data collection, use this section to describe them. --> #### Annotation process <!-- This section describes the annotation process such as annotation tools used in the process, the amount of data annotated, annotation guidelines provided to the annotators, interannotator statistics, annotation validation, etc. --> [More Information Needed] #### Who are the annotators? <!-- This section describes the people or systems who created the annotations. --> [More Information Needed] #### Personal and Sensitive Information <!-- State whether the dataset contains data that might be considered personal, sensitive, or private (e.g., data that reveals addresses, uniquely identifiable names or aliases, racial or ethnic origins, sexual orientations, religious beliefs, political opinions, financial or health data, etc.). If efforts were made to anonymize the data, describe the anonymization process. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations. ## Citation [optional] <!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the dataset or dataset card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Dataset Card Authors [optional] [More Information Needed] ## Dataset Card Contact [More Information Needed]

原始信息汇总

数据集概述

数据集名称

• pretty_name: Evaluation run of DUAL-GPO-2/phi-2-ipo-test-iter-0

数据集描述

• dataset_summary: 该数据集是在评估模型DUAL-GPO-2/phi-2-ipo-test-iter-0的过程中自动创建的，该模型在Open LLM Leaderboard上进行评估。数据集包含63个配置，每个配置对应一个评估任务。数据集由1次运行创建，每次运行在每个配置中作为一个特定的分割存在，分割名称使用运行的时间戳命名。"train"分割始终指向最新的结果。此外，还有一个名为"results"的配置，存储了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。

数据集结构

• 配置数量: 63个
• 运行次数: 1次
• 分割命名规则: 使用时间戳命名
• 特殊分割: "train"（指向最新结果）和"results"（存储聚合结果）

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_DUAL-GPO-2__phi-2-ipo-test-iter-0", "harness_winogrande_5", split="train")

最新结果

• 结果来源: 来自2024-04-24T04:35:58.978617的运行
• 结果内容: 包括多个任务的准确率(acc)、标准误差(acc_stderr)、归一化准确率(acc_norm)和归一化准确率的标准误差(acc_norm_stderr)等指标。

数据集详细配置

配置列表

1. harness_arc_challenge_25
2. harness_gsm8k_5
3. harness_hellaswag_10
4. harness_hendrycksTest_5

每个配置包含多个数据文件，每个文件对应一个特定的任务或评估指标，如准确率、标准误差等。

搜集汇总

数据集介绍

构建方式

该数据集为Open LLM Leaderboard在评估DUAL-GPO-2/phi-2-ipo-test-iter-0模型时自动生成的评测结果集合。数据集包含63个配置，每个配置对应一项被评估的任务，如ARC挑战、HellaSwag、GSM8K及涵盖多学科知识的Hendrycks测试等。数据源自单次运行，每次运行的详细结果以时间戳命名作为特定分割，而“train”分割始终指向最新一次评测的产出。此外，一个名为“results”的额外配置存储了所有运行的聚合结果，用于在排行榜上计算和展示综合指标。

使用方法

用户可通过HuggingFace Datasets库便捷加载数据。例如，使用`load_dataset`函数指定数据集名称、目标配置（如"harness_winogrande_5"）及分割（如"train"）即可获取最新结果。每个配置下的数据以Parquet格式存储，支持高效读取。若需回溯历史运行，可指定对应的时间戳分割名称。聚合结果则通过"results"配置访问，便于获取模型在全部任务上的综合表现。代码示例表明，该方法简洁直观，适用于研究者和开发者快速集成与分析模型评测数据。

背景与挑战

背景概述

随着大语言模型（LLM）在自然语言处理领域的迅猛发展，如何系统性地评估模型在多样化任务上的综合能力成为核心研究课题。Open LLM Leaderboard由Hugging Face团队于2023年发起，旨在为开源社区提供标准化、透明化的模型性能对比平台。该数据集记录了模型DUAL-GPO-2/phi-2-ipo-test-iter-0在2024年4月的一次评估运行结果，涵盖ARC-Challenge、HellaSwag、GSM8K以及涵盖57个学科的MMLU基准测试等63项配置。研究者通过该数据集可追溯模型在推理、常识理解、数学求解及专业知识等维度的表现，其影响力在于构建了可复现的评估生态，推动LLM研发从孤立评测向社区协作演进。

当前挑战

该数据集所应对的领域挑战在于如何全面衡量LLM在跨学科、跨难度层次上的泛化能力，例如MMLU中从基础代数到医学遗传学的知识跨度，以及GSM8K所需的数学推理精度，均对模型的广度与深度提出严苛要求。构建过程中的挑战包括：需统一不同任务（如多项选择与生成式问答）的评估协议，确保评分标准一致；处理大规模结果数据时，需维护63个配置与时间戳分片之间的版本关联，避免因多次运行导致的数据碎片化；同时，原始评估日志需转化为结构化Parquet文件，并同步聚合指标以供排行榜实时展示，这对数据管道的鲁棒性与时效性构成了技术考验。

常用场景

经典使用场景

在大型语言模型的演进浪潮中，Open LLM Leaderboard评估数据集已成为衡量模型综合能力的标杆。该数据集通过整合ARC-Challenge、HellaSwag、MMLU、TruthfulQA、Winogrande和GSM8K等经典基准测试，构建了一个覆盖常识推理、知识理解、数学求解与事实性判别等多维度的评估体系。研究者利用其结构化配置，可系统性地追踪模型在不同任务上的表现，例如通过加载'harness_winogrande_5'配置来评估代词消解能力，或通过'harness_gsm8k_5'配置检验数学推理的准确性。这种模块化设计允许对模型进行细粒度的能力剖析，从而揭示其在特定认知领域的优势与局限。

解决学术问题

该数据集的核心价值在于解决了大模型评估中长期存在的标准化缺失与结果不可复现的学术困境。通过提供统一的评估框架与可追溯的评估记录，它使得不同模型间的性能对比具有了统计意义上的严谨性。研究者得以基于此数据集系统性地探究模型在57个MMLU子领域（从抽象代数到病毒学）的知识覆盖广度，以及在零样本与少样本场景下的泛化能力。其意义在于为模型迭代提供了量化基准，推动了诸如指令微调、偏好对齐等训练策略的实证研究，最终促进了语言模型从'能做'到'善做'的质变。

实际应用

在实际产业应用中，该数据集为模型选型与部署决策提供了不可替代的参考依据。企业可通过分析模型在GSM8K上的数学推理得分，评估其在金融风控或教育辅导场景中的适用性；而TruthfulQA上的事实性指标则直接关联到医疗咨询或法律文档生成等高风险应用的可信度。此外，数据集支持对同一模型不同版本（如迭代优化前后）的纵向对比，帮助工程师量化微调策略的实际收益，从而在模型精度与推理效率之间做出权衡。这种数据驱动的评估机制已成为模型上线前质量门禁的关键环节。

数据集最近研究

最新研究方向

当前，大语言模型（LLM）的评估与优化已成为自然语言处理领域的前沿焦点，尤其是在模型对齐与推理能力提升方面。该数据集记录了DUAL-GPO-2/phi-2-ipo-test-iter-0模型在Open LLM Leaderboard上的完整评估结果，覆盖ARC-Challenge、HellaSwag、GSM8K、TruthfulQA、Winogrande以及涵盖57个学科的大规模多任务语言理解（MMLU）基准。这一研究方向紧密关联着模型通过迭代偏好优化（IPO）进行对齐的热点事件，旨在提升模型在复杂推理、事实一致性及常识理解上的表现。该数据集的发布不仅为社区提供了可复现的细粒度评估基准，还通过63个任务配置的标准化流程，推动了LLM性能比较的透明化与科学化，对理解模型在数学推理、科学知识及伦理判断等维度的能力边界具有深远意义。