open-llm-leaderboard-old/details_Mihaiii__Bucharest-0.2
收藏Hugging Face2024-02-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Mihaiii__Bucharest-0.2
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在评估模型Mihaiii/Bucharest-0.2时自动创建的,用于Open LLM Leaderboard的评估。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。
该数据集是在评估模型Mihaiii/Bucharest-0.2时自动创建的,用于Open LLM Leaderboard的评估。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- 名称: Evaluation run of Mihaiii/Bucharest-0.2
数据集来源
- 创建背景: 该数据集是在模型 Mihaiii/Bucharest-0.2 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集结构
- 配置数量: 该数据集包含 63 个配置,每个配置对应一个评估任务。
- 数据来源: 数据集从 1 次运行中创建,每次运行可以在每个配置中找到一个特定的分割,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。
- 结果汇总: 一个额外的配置 "results" 存储所有运行的汇总结果,用于计算和显示 Open LLM Leaderboard 上的聚合指标。
数据加载示例
- Python 代码示例: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Mihaiii__Bucharest-0.2", "harness_winogrande_5", split="train")
最新结果
- 最新结果日期: 2024-02-17T19:59:50.888473
- 结果示例: python { "all": { "acc": 0.6604284110771916, "acc_stderr": 0.03143827349144454, "acc_norm": 0.663246611615963, "acc_norm_stderr": 0.03206864715833573, "mc1": 0.30599755201958384, "mc1_stderr": 0.016132229728155045, "mc2": 0.45302642542372934, "mc2_stderr": 0.014396635503520975 }, "harness|arc:challenge|25": { "acc": 0.5921501706484642, "acc_stderr": 0.014361097288449703, "acc_norm": 0.6459044368600683, "acc_norm_stderr": 0.013975454122756557 }, "harness|hellaswag|10": { "acc": 0.6565425214100776, "acc_stderr": 0.004738920624724467, "acc_norm": 0.8487353116908982, "acc_norm_stderr": 0.003575744098779953 }, # 其他任务的结果... }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_02_17T19_59_50.888473
- 路径: **/details_harness|arc:challenge|25_2024-02-17T19-59-50.888473.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-02-17T19-59-50.888473.parquet
- 分割: 2024_02_17T19_59_50.888473
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_02_17T19_59_50.888473
- 路径: **/details_harness|gsm8k|5_2024-02-17T19-59-50.888473.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-02-17T19-59-50.888473.parquet
- 分割: 2024_02_17T19_59_50.888473
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_02_17T19_59_50.888473
- 路径: **/details_harness|hellaswag|10_2024-02-17T19-59-50.888473.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-02-17T19-59-50.888473.parquet
- 分割: 2024_02_17T19_59_50.888473
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_02_17T19_59_50.888473
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-02-17T19-59-50.888473.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2024-0
- 路径:
- 分割: 2024_02_17T19_59_50.888473
- 数据文件:



