open-llm-leaderboard/details_bigscience__bloomz
收藏数据集概述
该数据集是在模型None在Open LLM Leaderboard上的评估运行期间自动创建的。数据集包含61个配置,每个配置对应一个评估任务。数据集从1次运行中创建,每个运行的详细信息可以在每个配置中找到,使用运行的时间戳作为分割名称。"train"分割始终指向最新的结果。
数据集结构
数据集包含以下配置:
-
config_name: harness_arc_challenge_25
- 分割: 2023_08_29T12_14_13.875692
- 路径:
**/details_harness|arc:challenge|25_2023-08-29T12:14:13.875692.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-29T12:14:13.875692.parquet
- 路径:
- 分割: 2023_08_29T12_14_13.875692
-
config_name: harness_hellaswag_10
- 分割: 2023_08_29T12_14_13.875692
- 路径:
**/details_harness|hellaswag|10_2023-08-29T12:14:13.875692.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|hellaswag|10_2023-08-29T12:14:13.875692.parquet
- 路径:
- 分割: 2023_08_29T12_14_13.875692
-
config_name: harness_hendrycksTest_5
- 分割: 2023_08_29T12_14_13.875692
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-anatomy|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-astronomy|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_biology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-college_physics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-computer_security|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-econometrics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-global_facts|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-human_aging|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-international_law|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-management|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-marketing|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-nutrition|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-philosophy|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-prehistory|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-professional_accounting|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-professional_law|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-professional_medicine|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-professional_psychology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-public_relations|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-security_studies|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-sociology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-virology|5_2023-08-29T12:14:13.875692.parquet**/details_harness|hendrycksTest-world_religions|5_2023-08-29T12:14:13.875692.parquet
- 路径:
- 分割: 2023_08_29T12_14_13.875692
最新结果
以下是2023-08-29T12:14:13.875692运行的最新结果:
python { "all": { "acc": 0.47917755403252543, "acc_stderr": 0.03572101484290109, "acc_norm": 0.48335485520551164, "acc_norm_stderr": 0.0357085480998606, "mc1": 0.2607099143206854, "mc1_stderr": 0.015368841620766372, "mc2": 0.4393940961026447, "mc2_stderr": 0.015292532701908591 }, "harness|arc:challenge|25": { "acc": 0.5042662116040956, "acc_stderr": 0.014610858923956955, "acc_norm": 0.5537542662116041, "acc_norm_stderr": 0.014526705548539982 }, "harness|hellaswag|10": { "acc": 0.5553674566819359, "acc_stderr": 0.004959094146471527, "acc_norm": 0.7523401712806214, "acc_norm_stderr": 0.004307709682499536 }, "harness|hendrycksTest-abstract_algebra|5": { "acc



