open-llm-leaderboard-old/details_Chickaboo__Chicka-Mistral-4x7b
收藏Hugging Face2024-04-23 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Chickaboo__Chicka-Mistral-4x7b
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Chickaboo/Chicka-Mistral-4x7b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1次运行中生成的,每次运行在每个配置中表示为特定的拆分。train拆分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,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了如何使用`datasets`库中的`load_dataset`函数加载数据集的示例。
该数据集是在Open LLM Leaderboard上对模型Chickaboo/Chicka-Mistral-4x7b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1次运行中生成的,每次运行在每个配置中表示为特定的拆分。train拆分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,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了如何使用`datasets`库中的`load_dataset`函数加载数据集的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集创建
- 创建背景:该数据集是在评估模型Chickaboo/Chicka-Mistral-4x7b在Open LLM Leaderboard上的运行过程中自动创建的。
- 数据集组成:包含63个配置,每个配置对应一个评估任务。
- 创建次数:数据集来自1次运行,每个运行结果可以在每个配置中找到特定的分割,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_Chickaboo__Chicka-Mistral-4x7b", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳:2024-04-23T11:08:58.869435
- 结果示例:
python
{
"all": {
"acc": 0.6516830386212954,
"acc_stderr": 0.03178529202430958,
"acc_norm": 0.6517879277808889,
"acc_norm_stderr": 0.03244120713680502,
"mc1": 0.3463892288861689,
"mc1_stderr": 0.016656997109125136,
"mc2": 0.5050686705274005,
"mc2_stderr": 0.015305217846609598
},
"harness|arc:challenge|25": {
"acc": 0.613481228668942,
"acc_stderr": 0.014230084761910474,
"acc_norm": 0.6407849829351536,
"acc_norm_stderr": 0.01402022415583916
},
"harness|hellaswag|10": {
"acc": 0.6468830910177256,
"acc_stderr": 0.004769618829196507,
"acc_norm": 0.839573790081657,
"acc_norm_stderr": 0.003662508272330898
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.29,
"acc_stderr": 0.045604802157206845,
"acc_norm": 0.29,
"acc_norm_stderr": 0.045604802157206845
},
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_04_23T11_08_58.869435
- 路径:
**/details_harness|arc:challenge|25_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:2024_04_23T11_08_58.869435
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_04_23T11_08_58.869435
- 路径:
**/details_harness|gsm8k|5_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:2024_04_23T11_08_58.869435
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_04_23T11_08_58.869435
- 路径:
**/details_harness|hellaswag|10_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-04-23T11-08-58.869435.parquet
- 路径:
- 分割:2024_04_23T11_08_58.869435
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_04_23T11_08_58.869435
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-anatomy|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-astronomy|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_biology|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-college_physics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-computer_security|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-econometrics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-global_facts|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-human_aging|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-international_law|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-management|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-marketing|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-nutrition|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-philosophy|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-prehistory|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-04-23T11-08-58.869435.parquet**/details_harness|hendrycksTest-professional_law|5_2024-04-23T11-08-58.869435.parquet- `**/details_harness|hendrycksTest
- 路径:
- 分割:2024_04_23T11_08_58.869435
- 数据文件:



