open-llm-leaderboard-old/details_Open-Orca__Mixtral-SlimOrca-8x7B
收藏Hugging Face2023-12-14 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Open-Orca__Mixtral-SlimOrca-8x7B
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Open-Orca/Mixtral-SlimOrca-8x7B时自动创建的,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,用于计算和显示在Open LLM Leaderboard上的聚合指标。
该数据集是在评估模型Open-Orca/Mixtral-SlimOrca-8x7B时自动创建的,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,用于计算和显示在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 Open-Orca/Mixtral-SlimOrca-8x7B
数据集组成
- 配置数量: 63个配置,每个配置对应一个评估任务。
- 创建来源: 数据集是在模型 Open-Orca/Mixtral-SlimOrca-8x7B 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
- 运行次数: 数据集由1次运行创建,每个运行结果作为一个特定的分割存储,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,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Open-Orca__Mixtral-SlimOrca-8x7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2023-12-14T10:54:31.511638
- 结果示例:
python
{
"all": {
"acc": 0.6772838776728083,
"acc_stderr": 0.031217522187270183,
"acc_norm": 0.6826319719917908,
"acc_norm_stderr": 0.03182881958823988,
"mc1": 0.3818849449204406,
"mc1_stderr": 0.017008101939163498,
"mc2": 0.5498097510513819,
"mc2_stderr": 0.015613516175450912
},
"harness|arc:challenge|25": {
"acc": 0.6390784982935154,
"acc_stderr": 0.014034761386175452,
"acc_norm": 0.6766211604095563,
"acc_norm_stderr": 0.013669421630012144
},
"harness|hellaswag|10": {
"acc": 0.6669986058554073,
"acc_stderr": 0.004703238534045805,
"acc_norm": 0.8511252738498307,
"acc_norm_stderr": 0.003552374531305199
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.36,
"acc_stderr": 0.04824181513244218,
"acc_norm": 0.36,
"acc_norm_stderr": 0.04824181513244218
},
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: 2023_12_14T10_54_31.511638
- 路径: **/details_harness|arc:challenge|25_2023-12-14T10-54-31.511638.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2023-12-14T10-54-31.511638.parquet
- 分割: 2023_12_14T10_54_31.511638
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2023_12_14T10_54_31.511638
- 路径: **/details_harness|gsm8k|5_2023-12-14T10-54-31.511638.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2023-12-14T10-54-31.511638.parquet
- 分割: 2023_12_14T10_54_31.511638
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2023_12_14T10_54_31.511638
- 路径: **/details_harness|hellaswag|10_2023-12-14T10-54-31.511638.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2023-12-14T10-54-31.511638.parquet
- 分割: 2023_12_14T10_54_31.511638
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2023_12_14T10_54_31.511638
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-management|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-12-14T10-54-31.511638.parquet
- **/details_harness|hendrycksTest-professional_law|5_202
- 路径:
- 分割: 2023_12_14T10_54_31.511638
- 数据文件:



