open-llm-leaderboard-old/details_jeiku__Cookie_7B
收藏Hugging Face2024-02-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_jeiku__Cookie_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/Cookie_7B进行评估时自动创建的。它由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行作为每个配置中的一个特定分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。一个额外的配置results存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
该数据集是在Open LLM Leaderboard上对模型jeiku/Cookie_7B进行评估时自动创建的。它由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行作为每个配置中的一个特定分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。一个额外的配置results存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
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/Cookie_7B
- 数据集来源: 由模型 jeiku/Cookie_7B 在 Open LLM Leaderboard 评估运行期间自动创建。
- 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,每个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。"train" 分割始终指向最新的结果。
- 额外配置: "results" 配置存储所有运行的聚合结果,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_jeiku__Cookie_7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-02-17T00:21:45.959538
- 结果摘要: python { "all": { "acc": 0.6489384774347392, "acc_stderr": 0.032131421920616465, "acc_norm": 0.6498781521461111, "acc_norm_stderr": 0.032783132740631354, "mc1": 0.5128518971848225, "mc1_stderr": 0.01749771794429982, "mc2": 0.6687534212220169, "mc2_stderr": 0.015263939252034519 }, ... }
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_02_17T00_21_45.959538
- 路径: **/details_harness|arc:challenge|25_2024-02-17T00-21-45.959538.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-02-17T00-21-45.959538.parquet
- 分割: 2024_02_17T00_21_45.959538
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_02_17T00_21_45.959538
- 路径: **/details_harness|gsm8k|5_2024-02-17T00-21-45.959538.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-02-17T00-21-45.959538.parquet
- 分割: 2024_02_17T00_21_45.959538
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_02_17T00_21_45.959538
- 路径: **/details_harness|hellaswag|10_2024-02-17T00-21-45.959538.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-02-17T00-21-45.959538.parquet
- 分割: 2024_02_17T00_21_45.959538
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_02_17T00_21_45.959538
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-public_relations|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-security_studies|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-sociology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-us_foreign_policy|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-virology|5_2024-02-17T00-21-45.959538.parquet
- **/details_harness|hendrycksTest-world_religions|5_2024-02-17T00
- 路径:
- 分割: 2024_02_17T00_21_45.959538
- 数据文件:



