open-llm-leaderboard-old/details_Locutusque__Hyperion-2.0-Mistral-7B
收藏Hugging Face2024-03-10 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Locutusque__Hyperion-2.0-Mistral-7B
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型Locutusque/Hyperion-2.0-Mistral-7B进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行在每个配置中表示为特定的分割,train分割始终指向最新结果。此外,还有一个名为results的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。README文件还提供了如何加载运行中的详细信息的说明,并提供了特定运行的最新结果。
该数据集是在Open LLM Leaderboard上对模型Locutusque/Hyperion-2.0-Mistral-7B进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行在每个配置中表示为特定的分割,train分割始终指向最新结果。此外,还有一个名为results的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。README文件还提供了如何加载运行中的详细信息的说明,并提供了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集创建
- 创建背景:该数据集是在对模型 Locutusque/Hyperion-2.0-Mistral-7B 进行评估运行期间自动创建的。
- 评估平台:数据集用于 Open LLM Leaderboard。
数据集结构
- 配置数量:数据集包含 63 个配置,每个配置对应一个评估任务。
- 运行次数:数据集从 2 次运行中创建。每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。
- 最新结果:"train" 分割始终指向最新结果。
- 结果汇总:一个额外的配置 "results" 存储所有运行的汇总结果,用于计算和显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Locutusque__Hyperion-2.0-Mistral-7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源:最新结果来自 run 2024-03-10T05:52:30.143262。
- 结果示例: python { "all": { "acc": 0.6346692637770753, "acc_stderr": 0.03232834743290968, "acc_norm": 0.6397577306836747, "acc_norm_stderr": 0.03297845242054893, "mc1": 0.27539779681762544, "mc1_stderr": 0.01563813566777552, "mc2": 0.4197149652162468, "mc2_stderr": 0.014030449483056798 }, "harness|arc:challenge|25": { "acc": 0.5725255972696246, "acc_stderr": 0.014456862944650649, "acc_norm": 0.6109215017064846, "acc_norm_stderr": 0.014247309976045607 }, "harness|hellaswag|10": { "acc": 0.6323441545508863, "acc_stderr": 0.004811815959388832, "acc_norm": 0.8349930292770364, "acc_norm_stderr": 0.0037042823907817183 }, # 其他任务结果省略... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2024_03_10T04_55_15.610547
- 路径:
**/details_harness|arc:challenge|25_2024-03-10T04-55-15.610547.parquet
- 路径:
- 分割:2024_03_10T05_52_30.143262
- 路径:
**/details_harness|arc:challenge|25_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:2024_03_10T04_55_15.610547
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_03_10T04_55_15.610547
- 路径:
**/details_harness|gsm8k|5_2024-03-10T04-55-15.610547.parquet
- 路径:
- 分割:2024_03_10T05_52_30.143262
- 路径:
**/details_harness|gsm8k|5_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:2024_03_10T04_55_15.610547
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_03_10T04_55_15.610547
- 路径:
**/details_harness|hellaswag|10_2024-03-10T04-55-15.610547.parquet
- 路径:
- 分割:2024_03_10T05_52_30.143262
- 路径:
**/details_harness|hellaswag|10_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-03-10T05-52-30.143262.parquet
- 路径:
- 分割:2024_03_10T04_55_15.610547
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_03_10T04_55_15.610547
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-anatomy|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-astronomy|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_biology|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-college_physics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-computer_security|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-econometrics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-global_facts|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-human_aging|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-international_law|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-management|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-marketing|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-03-10T04-55-15.610547.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-03-10T04-55-15.610547.parquet- `**/details_harness|hendrycksTest-nutrition|5_2024-03-10T04-55-15
- 路径:
- 分割:2024_03_10T04_55_15.610547
- 数据文件:



