open-llm-leaderboard-old/details_perlthoughts__Chupacabra-7B-v2
收藏数据集概述
数据集简介
该数据集是在评估模型 perlthoughts/Chupacabra-7B-v2 的过程中自动创建的,用于 Open LLM Leaderboard 的评估。数据集包含 64 个配置,每个配置对应一个评估任务。
数据集结构
数据集由 4 次运行结果组成,每次运行的结果作为特定分割存储,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。此外,还有一个额外的配置 "results",存储所有运行的聚合结果,用于计算和显示聚合指标。
数据加载示例
以下是加载数据集详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_perlthoughts__Chupacabra-7B-v2", "harness_winogrande_5", split="train")
最新结果
以下是来自 2023-12-04T18:02:58.053786 运行的最新结果: python { "all": { "acc": 0.6367599015709243, "acc_stderr": 0.03218025799515212, "acc_norm": 0.6396357428050704, "acc_norm_stderr": 0.0328187456646889, "mc1": 0.397796817625459, "mc1_stderr": 0.017133934248559635, "mc2": 0.5717077514762566, "mc2_stderr": 0.0156197692783717 }, "harness|arc:challenge|25": { "acc": 0.613481228668942, "acc_stderr": 0.014230084761910478, "acc_norm": 0.6518771331058021, "acc_norm_stderr": 0.013921008595179342 }, "harness|hellaswag|10": { "acc": 0.6473809998008365, "acc_stderr": 0.004768088918512182, "acc_norm": 0.8338976299541924, "acc_norm_stderr": 0.003714118884317389 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.25, "acc_stderr": 0.04351941398892446, "acc_norm": 0.25, "acc_norm_stderr": 0.04351941398892446 }, # 其他任务的结果... }
配置详情
数据集包含多个配置,每个配置对应不同的任务和运行结果。以下是部分配置的示例:
-
config_name: harness_arc_challenge_25
- split: 2023_11_23T09_06_05.823190
- path: **/details_harness|arc:challenge|25_2023-11-23T09-06-05.823190.parquet
- split: 2023_11_23T09_18_59.989572
- path: **/details_harness|arc:challenge|25_2023-11-23T09-18-59.989572.parquet
- split: 2023_12_04T18_02_58.053786
- path: **/details_harness|arc:challenge|25_2023-12-04T18-02-58.053786.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-12-04T18-02-58.053786.parquet
- split: 2023_11_23T09_06_05.823190
-
config_name: harness_drop_3
- split: 2023_11_23T09_06_05.823190
- path: **/details_harness|drop|3_2023-11-23T09-06-05.823190.parquet
- split: 2023_11_23T09_18_59.989572
- path: **/details_harness|drop|3_2023-11-23T09-18-59.989572.parquet
- split: latest
- path: **/details_harness|drop|3_2023-11-23T09-18-59.989572.parquet
- split: 2023_11_23T09_06_05.823190
-
config_name: harness_gsm8k_5
- split: 2023_11_23T09_06_05.823190
- path: **/details_harness|gsm8k|5_2023-11-23T09-06-05.823190.parquet
- split: 2023_11_23T09_18_59.989572
- path: **/details_harness|gsm8k|5_2023-11-23T09-18-59.989572.parquet
- split: 2023_12_03T15_21_26.428024
- path: **/details_harness|gsm8k|5_2023-12-03T15-21-26.428024.parquet
- split: 2023_12_04T18_02_58.053786
- path: **/details_harness|gsm8k|5_2023-12-04T18-02-58.053786.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-12-04T18-02-58.053786.parquet
- split: 2023_11_23T09_06_05.823190
-
config_name: harness_hellaswag_10
- split: 2023_11_23T09_06_05.823190
- path: **/details_harness|hellaswag|10_2023-11-23T09-06-05.823190.parquet
- split: 2023_11_23T09_18_59.989572
- path: **/details_harness|hellaswag|10_2023-11-23T09-18-59.989572.parquet
- split: 2023_12_04T18_02_58.053786
- path: **/details_harness|hellaswag|10_2023-12-04T18-02-58.053786.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-12-04T18-02-58.053786.parquet
- split: 2023_11_23T09_06_05.823190
-
config_name: harness_hendrycksTest_5
- split: 2023_11_23T09_06_05.823190
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-11-23T09-06-05.823190.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-11-23T09-06-05.823190.parquet
- path:
- split: 2023_11_23T09_06_05.823190



