open-llm-leaderboard-old/details_jeiku__Mewthree_7B
收藏Hugging Face2024-03-01 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_jeiku__Mewthree_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jeiku/Mewthree_7B进行评估时自动生成的。数据集包含63个配置,每个配置对应一个被评估的任务。它包含一次运行的结果,每次运行都表示为一个特定的分割,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Hugging Face datasets库加载数据集的示例。
该数据集是在Open LLM Leaderboard上对模型jeiku/Mewthree_7B进行评估时自动生成的。数据集包含63个配置,每个配置对应一个被评估的任务。它包含一次运行的结果,每次运行都表示为一个特定的分割,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Hugging Face datasets库加载数据集的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集信息
- 数据集名称: Evaluation run of jeiku/Mewthree_7B
- 数据集来源: 自动创建于模型 jeiku/Mewthree_7B 在 Open LLM Leaderboard 的评估运行期间。
- 数据集组成: 包含 63 个配置,每个配置对应一个评估任务。
- 数据集创建: 从 1 次运行中创建,每次运行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。
- 额外配置: "results" 配置存储所有运行结果的聚合,用于计算和显示聚合指标在 Open LLM Leaderboard。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_jeiku__Mewthree_7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-03-01T13:33:27.164796
- 结果示例:
python
{
"all": {
"acc": 0.6228333701150645,
"acc_stderr": 0.0325486787942996,
"acc_norm": 0.6300446337796757,
"acc_norm_stderr": 0.033234505552346916,
"mc1": 0.5042839657282742,
"mc1_stderr": 0.017502858577371258,
"mc2": 0.6585420762235195,
"mc2_stderr": 0.015172693517885267
},
"harness|arc:challenge|25": {
"acc": 0.6313993174061433,
"acc_stderr": 0.014097810678042196,
"acc_norm": 0.6578498293515358,
"acc_norm_stderr": 0.013864152159177275
},
"harness|hellaswag|10": {
"acc": 0.6767576180043816,
"acc_stderr": 0.004667585072717497,
"acc_norm": 0.8573989245170285,
"acc_norm_stderr": 0.0034895094930016196
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.33,
"acc_stderr": 0.04725815626252605,
"acc_norm": 0.33,
"acc_norm_stderr": 0.04725815626252605
},
其他任务结果省略...
}
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_03_01T13_33_27.164796
- 路径: **/details_harness|arc:challenge|25_2024-03-01T13-33-27.164796.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-03-01T13-33-27.164796.parquet
- 分割: 2024_03_01T13_33_27.164796
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_03_01T13_33_27.164796
- 路径: **/details_harness|gsm8k|5_2024-03-01T13-33-27.164796.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-03-01T13-33-27.164796.parquet
- 分割: 2024_03_01T13_33_27.164796
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_03_01T13_33_27.164796
- 路径: **/details_harness|hellaswag|10_2024-03-01T13-33-27.164796.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-03-01T13-33-27.164796.parquet
- 分割: 2024_03_01T13_33_27.164796
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_03_01T13_33_27.164796
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-management|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-03-01T13-33-27.164796.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-03
- 路径:
- 分割: 2024_03_01T13_33_27.164796
- 数据文件:



