open-llm-leaderboard-old/details_abhishekchohan__mistral-7B-forest
收藏Hugging Face2024-01-23 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_abhishekchohan__mistral-7B-forest
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abhishekchohan/mistral-7B-forest时自动创建的,评估过程在Open LLM Leaderboard上进行。数据集由63个配置组成,每个配置对应一个评估任务。数据集从1次运行中创建,每次运行可以在每个配置的特定分割中找到,分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
该数据集是在评估模型abhishekchohan/mistral-7B-forest时自动创建的,评估过程在Open LLM Leaderboard上进行。数据集由63个配置组成,每个配置对应一个评估任务。数据集从1次运行中创建,每次运行可以在每个配置的特定分割中找到,分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示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
数据集基本信息
- 数据集名称: Evaluation run of abhishekchohan/mistral-7B-forest
- 数据集来源: 在评估模型 abhishekchohan/mistral-7B-forest 在 Open LLM Leaderboard 上的自动创建。
- 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。"train" 分割始终指向最新结果。
- 额外配置: 一个名为 "results" 的额外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_abhishekchohan__mistral-7B-forest", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源: 最新结果来自 run 2024-01-23T15:31:25.138971
- 结果示例: python { "all": { "acc": 0.6312757004711416, "acc_stderr": 0.03248074055562836, "acc_norm": 0.6374796440943401, "acc_norm_stderr": 0.03314363503746386, "mc1": 0.36964504283965727, "mc1_stderr": 0.016898180706973884, "mc2": 0.5332124020960498, "mc2_stderr": 0.01581823670998924 }, "harness|arc:challenge|25": { "acc": 0.6143344709897611, "acc_stderr": 0.014224250973257182, "acc_norm": 0.6569965870307167, "acc_norm_stderr": 0.013872423223718164 }, ... }
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_01_23T15_31_25.138971
- 路径:
**/details_harness|arc:challenge|25_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|arc:challenge|25_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: 2024_01_23T15_31_25.138971
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_01_23T15_31_25.138971
- 路径:
**/details_harness|gsm8k|5_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|gsm8k|5_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: 2024_01_23T15_31_25.138971
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_01_23T15_31_25.138971
- 路径:
**/details_harness|hellaswag|10_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|hellaswag|10_2024-01-23T15-31-25.138971.parquet
- 路径:
- 分割: 2024_01_23T15_31_25.138971
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_01_23T15_31_25.138971
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-management|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-01-23T15-31-25.138971.parquet**/details_harness|hendrycksTest-public_relations|5_2024-01-23T15-31-25.138971.parquet- `**/
- 路径:
- 分割: 2024_01_23T15_31_25.138971
- 数据文件:



