open-llm-leaderboard/details_MaziyarPanahi__Llama-3-8B-Instruct-v0.5
收藏Hugging Face2024-05-07 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-8B-Instruct-v0.5
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在模型MaziyarPanahi/Llama-3-8B-Instruct-v0.5在Open LLM Leaderboard上的评估过程中自动生成的。数据集由63个配置组成,每个配置对应一个被评估的任务。它由2次运行生成,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个使用Python中的datasets库加载运行细节的示例。
该数据集是在模型MaziyarPanahi/Llama-3-8B-Instruct-v0.5在Open LLM Leaderboard上的评估过程中自动生成的。数据集由63个配置组成,每个配置对应一个被评估的任务。它由2次运行生成,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个使用Python中的datasets库加载运行细节的示例。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集名称
- Evaluation run of MaziyarPanahi/Llama-3-8B-Instruct-v0.5
数据集描述
- 该数据集是自动创建的,用于评估模型 MaziyarPanahi/Llama-3-8B-Instruct-v0.5 在 Open LLM Leaderboard 上的表现。
- 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2次运行创建,每次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命名。
- “train”分割始终指向最新的结果。
- 额外的配置“results”存储了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。
数据集加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_MaziyarPanahi__Llama-3-8B-Instruct-v0.5", "harness_gsm8k_5", split="train")
最新结果
- 最新结果来自2024-05-07T12:28:02.488575的运行,具体结果如下: python { "all": { "acc": 0.7134192570128886, "acc_stderr": 0.012454841668337694 }, "harness|gsm8k|5": { "acc": 0.7134192570128886, "acc_stderr": 0.012454841668337694 } }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_05_04T13_53_12.398733
- 路径: /details_harness|arc:challenge|25_2024-05-04T13-53-12.398733.parquet
- 分割: latest
- 路径: /details_harness|arc:challenge|25_2024-05-04T13-53-12.398733.parquet
- 分割: 2024_05_04T13_53_12.398733
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_05_04T13_53_12.398733
- 路径: /details_harness|gsm8k|5_2024-05-04T13-53-12.398733.parquet
- 分割: 2024_05_07T12_28_02.488575
- 路径: /details_harness|gsm8k|5_2024-05-07T12-28-02.488575.parquet
- 分割: latest
- 路径: /details_harness|gsm8k|5_2024-05-07T12-28-02.488575.parquet
- 分割: 2024_05_04T13_53_12.398733
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_05_04T13_53_12.398733
- 路径: /details_harness|hellaswag|10_2024-05-04T13-53-12.398733.parquet
- 分割: latest
- 路径: /details_harness|hellaswag|10_2024-05-04T13-53-12.398733.parquet
- 分割: 2024_05_04T13_53_12.398733
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_05_04T13_53_12.398733
- 路径:
- /details_harness|hendrycksTest-abstract_algebra|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-anatomy|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-astronomy|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-business_ethics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-clinical_knowledge|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_biology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_chemistry|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_computer_science|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_mathematics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_medicine|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-college_physics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-computer_security|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-conceptual_physics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-econometrics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-electrical_engineering|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-elementary_mathematics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-formal_logic|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-global_facts|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_biology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_chemistry|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_computer_science|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_european_history|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_geography|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_government_and_politics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_macroeconomics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_mathematics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_microeconomics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_physics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_psychology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_statistics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_us_history|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-high_school_world_history|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-human_aging|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-human_sexuality|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-international_law|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-jurisprudence|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-logical_fallacies|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-machine_learning|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-management|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-marketing|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-medical_genetics|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-miscellaneous|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-moral_disputes|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-moral_scenarios|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-nutrition|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-philosophy|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-prehistory|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-professional_accounting|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-professional_law|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-professional_medicine|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-professional_psychology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-public_relations|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-security_studies|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-sociology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-us_foreign_policy|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-virology|5_2024-05-04T13-53-12.398733.parquet
- /details_harness|hendrycksTest-world_religions|5_2024-05-04T13-53-12.398733.parquet
- 路径:
- 分割: latest
- 路径:
- **/details_
- 路径:
- 分割: 2024_05_04T13_53_12.398733
- 数据文件:



