cluster_loras / 12 /mmlu /metrics.json
Muqeeth's picture
Upload folder using huggingface_hub
9224edf verified
{"average_acc": 0.3577837914826948, "subcat_acc": {"math": 0.2349624060150376, "health": 0.3646341463414634, "physics": 0.2859375, "business": 0.4805491990846682, "biology": 0.3634361233480176, "chemistry": 0.21782178217821782, "computer science": 0.34951456310679613, "economics": 0.3045822102425876, "engineering": 0.3793103448275862, "philosophy": 0.31759443339960236, "other": 0.4111587982832618, "history": 0.4698924731182796, "geography": 0.29292929292929293, "politics": 0.41975308641975306, "psychology": 0.38547968885047534, "culture": 0.5301204819277109, "law": 0.3516732841747022}, "cat_acc": {"STEM": 0.2859509609012591, "humanities": 0.3604675876726886, "social sciences": 0.3828404289892753, "other (business, health, misc.)": 0.3969771745835904}, "subjects": {"abstract_algebra": 0.26, "anatomy": 0.362962962962963, "astronomy": 0.3223684210526316, "business_ethics": 0.49, "clinical_knowledge": 0.3433962264150943, "college_biology": 0.3472222222222222, "college_chemistry": 0.28, "college_computer_science": 0.3, "college_mathematics": 0.25, "college_medicine": 0.3236994219653179, "college_physics": 0.24509803921568626, "computer_security": 0.4, "conceptual_physics": 0.33191489361702126, "econometrics": 0.22807017543859648, "electrical_engineering": 0.3793103448275862, "elementary_mathematics": 0.2222222222222222, "formal_logic": 0.30158730158730157, "global_facts": 0.25, "high_school_biology": 0.3709677419354839, "high_school_chemistry": 0.18719211822660098, "high_school_computer_science": 0.35, "high_school_european_history": 0.5454545454545454, "high_school_geography": 0.29292929292929293, "high_school_government_and_politics": 0.48704663212435234, "high_school_macroeconomics": 0.29743589743589743, "high_school_mathematics": 0.25925925925925924, "high_school_microeconomics": 0.35294117647058826, "high_school_physics": 0.2052980132450331, "high_school_psychology": 0.3834862385321101, "high_school_statistics": 0.20833333333333334, "high_school_us_history": 0.4852941176470588, "high_school_world_history": 0.540084388185654, "human_aging": 0.40358744394618834, "human_sexuality": 0.4580152671755725, "international_law": 0.5041322314049587, "jurisprudence": 0.4166666666666667, "logical_fallacies": 0.44785276073619634, "machine_learning": 0.3482142857142857, "management": 0.32038834951456313, "marketing": 0.5470085470085471, "medical_genetics": 0.39, "miscellaneous": 0.44699872286079184, "moral_disputes": 0.3208092485549133, "moral_scenarios": 0.23798882681564246, "nutrition": 0.3954248366013072, "philosophy": 0.39228295819935693, "prehistory": 0.37037037037037035, "professional_accounting": 0.36879432624113473, "professional_law": 0.3350717079530639, "professional_medicine": 0.33088235294117646, "professional_psychology": 0.3872549019607843, "public_relations": 0.35454545454545455, "security_studies": 0.35918367346938773, "sociology": 0.5771144278606966, "us_foreign_policy": 0.51, "virology": 0.37349397590361444, "world_religions": 0.47953216374269003}}