cluster_loras / 2 /mmlu /metrics.json
Muqeeth's picture
Upload folder using huggingface_hub
9224edf verified
{"average_acc": 0.40507050277738216, "subcat_acc": {"math": 0.26597744360902253, "health": 0.42134146341463413, "physics": 0.3203125, "business": 0.5331807780320366, "biology": 0.42290748898678415, "chemistry": 0.3102310231023102, "computer science": 0.39805825242718446, "economics": 0.35444743935309975, "engineering": 0.42758620689655175, "philosophy": 0.36133200795228626, "other": 0.4609442060085837, "history": 0.5064516129032258, "geography": 0.4090909090909091, "politics": 0.5138888888888888, "psychology": 0.445980985306828, "culture": 0.6024096385542169, "law": 0.36074872376630746}, "cat_acc": {"STEM": 0.3313452617627568, "humanities": 0.389798087141339, "social sciences": 0.45271368215794605, "other (business, health, misc.)": 0.45064774830351634}, "subjects": {"abstract_algebra": 0.27, "anatomy": 0.4222222222222222, "astronomy": 0.42105263157894735, "business_ethics": 0.42, "clinical_knowledge": 0.4377358490566038, "college_biology": 0.3888888888888889, "college_chemistry": 0.3, "college_computer_science": 0.41, "college_mathematics": 0.36, "college_medicine": 0.3930635838150289, "college_physics": 0.22549019607843138, "computer_security": 0.47, "conceptual_physics": 0.3191489361702128, "econometrics": 0.22807017543859648, "electrical_engineering": 0.42758620689655175, "elementary_mathematics": 0.22486772486772486, "formal_logic": 0.373015873015873, "global_facts": 0.27, "high_school_biology": 0.43870967741935485, "high_school_chemistry": 0.31527093596059114, "high_school_computer_science": 0.36, "high_school_european_history": 0.5515151515151515, "high_school_geography": 0.4090909090909091, "high_school_government_and_politics": 0.5803108808290155, "high_school_macroeconomics": 0.38461538461538464, "high_school_mathematics": 0.24814814814814815, "high_school_microeconomics": 0.36554621848739494, "high_school_physics": 0.2847682119205298, "high_school_psychology": 0.5027522935779817, "high_school_statistics": 0.3148148148148148, "high_school_us_history": 0.5147058823529411, "high_school_world_history": 0.5485232067510548, "human_aging": 0.3901345291479821, "human_sexuality": 0.5114503816793893, "international_law": 0.5785123966942148, "jurisprudence": 0.4166666666666667, "logical_fallacies": 0.50920245398773, "machine_learning": 0.35714285714285715, "management": 0.42718446601941745, "marketing": 0.6282051282051282, "medical_genetics": 0.49, "miscellaneous": 0.5197956577266922, "moral_disputes": 0.3699421965317919, "moral_scenarios": 0.2446927374301676, "nutrition": 0.46078431372549017, "philosophy": 0.4887459807073955, "prehistory": 0.44753086419753085, "professional_accounting": 0.36524822695035464, "professional_law": 0.3396349413298566, "professional_medicine": 0.3860294117647059, "professional_psychology": 0.3954248366013072, "public_relations": 0.39090909090909093, "security_studies": 0.4816326530612245, "sociology": 0.6616915422885572, "us_foreign_policy": 0.6, "virology": 0.40963855421686746, "world_religions": 0.5730994152046783}}