open-llm-leaderboard-old/details_zhengr__MixTAO-7Bx2-MoE-DPO
收藏数据集概述
数据集描述
该数据集是在评估模型zhengr/MixTAO-7Bx2-MoE-DPO在Open LLM Leaderboard上的运行过程中自动创建的。
数据集组成
- 该数据集包含63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,每个运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。
- "train"分割始终指向最新的结果。
- 一个额外的配置"results"存储所有运行的聚合结果,用于计算和显示在Open LLM Leaderboard上的聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_zhengr__MixTAO-7Bx2-MoE-DPO", "harness_winogrande_5", split="train")
最新结果
这些是最新结果,来自2024-01-17T13:23:29.676681的运行: python { "all": { "acc": 0.6522594762182442, "acc_stderr": 0.0319843897745183, "acc_norm": 0.6520128707870778, "acc_norm_stderr": 0.03264277329988372, "mc1": 0.5385556915544676, "mc1_stderr": 0.017451384104637455, "mc2": 0.6934208246675816, "mc2_stderr": 0.014891018416465928 }, "harness|arc:challenge|25": { "acc": 0.6808873720136519, "acc_stderr": 0.01362169611917331, "acc_norm": 0.7090443686006825, "acc_norm_stderr": 0.013273077865907595 }, "harness|hellaswag|10": { "acc": 0.7027484564827724, "acc_stderr": 0.004561141293448457, "acc_norm": 0.8712407886875124, "acc_norm_stderr": 0.003342487333262275 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.04760952285695236, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695236 }, # 其他任务的结果省略... }
配置信息
-
config_name: harness_arc_challenge_25
- split: 2024_01_17T13_23_29.676681
- path: **/details_harness|arc:challenge|25_2024-01-17T13-23-29.676681.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-01-17T13-23-29.676681.parquet
- split: 2024_01_17T13_23_29.676681
-
config_name: harness_gsm8k_5
- split: 2024_01_17T13_23_29.676681
- path: **/details_harness|gsm8k|5_2024-01-17T13-23-29.676681.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-01-17T13-23-29.676681.parquet
- split: 2024_01_17T13_23_29.676681
-
config_name: harness_hellaswag_10
- split: 2024_01_17T13_23_29.676681
- path: **/details_harness|hellaswag|10_2024-01-17T13-23-29.676681.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-01-17T13-23-29.676681.parquet
- split: 2024_01_17T13_23_29.676681
-
config_name: harness_hendrycksTest_5
- split: 2024_01_17T13_23_29.676681
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-management|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-01-17T13-23-29.676681.parquet
- **/details_harness|hendrycksTest-professional_psych
- path:
- split: 2024_01_17T13_23_29.676681



