cluster_loras / 18 /mmlu /metrics.json
Muqeeth's picture
Upload folder using huggingface_hub
9224edf verified
{"average_acc": 0.3957413473864122, "subcat_acc": {"math": 0.27349624060150374, "health": 0.38658536585365855, "physics": 0.3171875, "business": 0.5217391304347826, "biology": 0.4008810572687225, "chemistry": 0.30033003300330036, "computer science": 0.3592233009708738, "economics": 0.36253369272237196, "engineering": 0.3931034482758621, "philosophy": 0.34493041749502984, "other": 0.4472103004291845, "history": 0.4989247311827957, "geography": 0.42424242424242425, "politics": 0.5169753086419753, "psychology": 0.4373379429559205, "culture": 0.5753012048192772, "law": 0.37379466817923995}, "cat_acc": {"STEM": 0.3220675944333996, "humanities": 0.38618490967056324, "social sciences": 0.4501137471563211, "other (business, health, misc.)": 0.4265885256014806}, "subjects": {"abstract_algebra": 0.31, "anatomy": 0.45185185185185184, "astronomy": 0.40131578947368424, "business_ethics": 0.44, "clinical_knowledge": 0.39622641509433965, "college_biology": 0.3888888888888889, "college_chemistry": 0.38, "college_computer_science": 0.33, "college_mathematics": 0.31, "college_medicine": 0.3988439306358382, "college_physics": 0.21568627450980393, "computer_security": 0.38, "conceptual_physics": 0.33191489361702126, "econometrics": 0.2631578947368421, "electrical_engineering": 0.3931034482758621, "elementary_mathematics": 0.2275132275132275, "formal_logic": 0.3492063492063492, "global_facts": 0.26, "high_school_biology": 0.4064516129032258, "high_school_chemistry": 0.26108374384236455, "high_school_computer_science": 0.39, "high_school_european_history": 0.5878787878787879, "high_school_geography": 0.42424242424242425, "high_school_government_and_politics": 0.5751295336787565, "high_school_macroeconomics": 0.38974358974358975, "high_school_mathematics": 0.24814814814814815, "high_school_microeconomics": 0.36554621848739494, "high_school_physics": 0.2781456953642384, "high_school_psychology": 0.48990825688073397, "high_school_statistics": 0.35185185185185186, "high_school_us_history": 0.5098039215686274, "high_school_world_history": 0.5232067510548524, "human_aging": 0.36771300448430494, "human_sexuality": 0.4961832061068702, "international_law": 0.5371900826446281, "jurisprudence": 0.4722222222222222, "logical_fallacies": 0.4601226993865031, "machine_learning": 0.3392857142857143, "management": 0.4563106796116505, "marketing": 0.5854700854700855, "medical_genetics": 0.4, "miscellaneous": 0.49936143039591313, "moral_disputes": 0.34971098265895956, "moral_scenarios": 0.25921787709497207, "nutrition": 0.39869281045751637, "philosophy": 0.44694533762057875, "prehistory": 0.42901234567901236, "professional_accounting": 0.36879432624113473, "professional_law": 0.3539765319426336, "professional_medicine": 0.3382352941176471, "professional_psychology": 0.39052287581699346, "public_relations": 0.38181818181818183, "security_studies": 0.5020408163265306, "sociology": 0.6268656716417911, "us_foreign_policy": 0.59, "virology": 0.3795180722891566, "world_religions": 0.4853801169590643}}