open-llm-leaderboard-old/details_mlabonne__Darewin-7B
收藏Hugging Face2024-01-24 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_mlabonne__Darewin-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mlabonne/Darewin-7B进行评估运行时自动创建的。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分割,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,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个示例,展示了如何使用`datasets`库中的`load_dataset`函数加载运行中的详细信息。
该数据集是在Open LLM Leaderboard上对模型mlabonne/Darewin-7B进行评估运行时自动创建的。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分割,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,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个示例,展示了如何使用`datasets`库中的`load_dataset`函数加载运行中的详细信息。
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
mlabonne/Darewin-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。
- 数据来源:数据集从 1 次运行中创建,每次运行结果存储为一个特定的分割,分割名称使用运行的时间戳。
- 最新结果:"train" 分割始终指向最新的结果。
- 聚合结果:一个额外的配置 "results" 存储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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_mlabonne__Darewin-7B", "harness_winogrande_5", split="train")
最新结果
- 运行时间:2024-01-24T08:13:57.054758
- 结果示例: python { "all": { "acc": 0.6560164129347533, "acc_stderr": 0.031962689434190776, "acc_norm": 0.6562048573975326, "acc_norm_stderr": 0.03261805811687008, "mc1": 0.4394124847001224, "mc1_stderr": 0.01737452048251371, "mc2": 0.6038301238600606, "mc2_stderr": 0.015402053910396342 }, "harness|arc:challenge|25": { "acc": 0.6621160409556314, "acc_stderr": 0.013822047922283516, "acc_norm": 0.6860068259385665, "acc_norm_stderr": 0.013562691224726304 }, "harness|hellaswag|10": { "acc": 0.6786496713802032, "acc_stderr": 0.004660405565338756, "acc_norm": 0.8621788488348935, "acc_norm_stderr": 0.003440076775300576 }, # 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_01_24T08_13_57.054758- 路径:
**/details_harness|arc:challenge|25_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|arc:challenge|25_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_gsm8k_5- 数据文件:
- 分割:
2024_01_24T08_13_57.054758- 路径:
**/details_harness|gsm8k|5_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|gsm8k|5_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_hellaswag_10- 数据文件:
- 分割:
2024_01_24T08_13_57.054758- 路径:
**/details_harness|hellaswag|10_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|hellaswag|10_2024-01-24T08-13-57.054758.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_hendrycksTest_5- 数据文件:
- 分割:
2024_01_24T08_13_57.054758- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-management|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-24T08-13-57.054758.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-24T08-13-57.054758.parquet- `**/details_harness|hendrycksTest-professional_psychology|5_2024-01-24T08-13-57.05475
- 路径:
- 分割:
- 数据文件:



