open-llm-leaderboard-old/details_acrastt__kalomaze-stuff
收藏数据集概述
数据集简介
该数据集是在对模型 acrastt/kalomaze-stuff 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。数据集包含 63 个配置,每个配置对应一个评估任务。
数据集结构
数据集从 1 次运行中创建,每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。"train" 分片始终指向最新的结果。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_acrastt__kalomaze-stuff", "harness_winogrande_5", split="train")
最新结果
以下是 2023-12-23T16:26:56.978626 运行 的最新结果:
python { "all": { "acc": 0.631028554660595, "acc_stderr": 0.0324014053372231, "acc_norm": 0.6369168843854331, "acc_norm_stderr": 0.03305850958730039, "mc1": 0.2778457772337821, "mc1_stderr": 0.015680929364024647, "mc2": 0.4163648865691074, "mc2_stderr": 0.014225165369587414 }, "harness|arc:challenge|25": { "acc": 0.5648464163822525, "acc_stderr": 0.014487986197186045, "acc_norm": 0.5964163822525598, "acc_norm_stderr": 0.014337158914268447 }, "harness|hellaswag|10": { "acc": 0.6324437363075085, "acc_stderr": 0.004811543077792714, "acc_norm": 0.8354909380601474, "acc_norm_stderr": 0.0036997919347543672 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.27, "acc_stderr": 0.044619604333847394, "acc_norm": 0.27, "acc_norm_stderr": 0.044619604333847394 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_12_23T16_26_56.978626
- path: **/details_harness|arc:challenge|25_2023-12-23T16-26-56.978626.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-12-23T16-26-56.978626.parquet
- split: 2023_12_23T16_26_56.978626
-
config_name: harness_gsm8k_5
- split: 2023_12_23T16_26_56.978626
- path: **/details_harness|gsm8k|5_2023-12-23T16-26-56.978626.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-12-23T16-26-56.978626.parquet
- split: 2023_12_23T16_26_56.978626
-
config_name: harness_hellaswag_10
- split: 2023_12_23T16_26_56.978626
- path: **/details_harness|hellaswag|10_2023-12-23T16-26-56.978626.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-12-23T16-26-56.978626.parquet
- split: 2023_12_23T16_26_56.978626
-
config_name: harness_hendrycksTest_5
- split: 2023_12_23T16_26_56.978626
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-management|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2023-12-23T16-26-56.978626.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2023-12-23T16-26-56.978
- path:
- split: 2023_12_23T16_26_56.978626



