open-llm-leaderboard-old/details_Qwen__Qwen2-0.5B
收藏Hugging Face2024-05-30 更新2024-06-29 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Qwen__Qwen2-0.5B
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Qwen/Qwen2-0.5B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次运行中创建的,每次运行在每个配置中作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。一个额外的配置results存储了运行的所有聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个如何使用`datasets`库中的`load_dataset`函数加载运行细节的示例。
该数据集是在Open LLM Leaderboard上对模型Qwen/Qwen2-0.5B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次运行中创建的,每次运行在每个配置中作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。一个额外的配置results存储了运行的所有聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个如何使用`datasets`库中的`load_dataset`函数加载运行细节的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- Evaluation run of Qwen/Qwen2-0.5B
数据集概述
- 该数据集是在模型 Qwen/Qwen2-0.5B 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
- 数据集由 63 个配置组成,每个配置对应一个评估任务。
- 数据集从 1 次运行中创建,每次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" 分割始终指向最新的结果。
- 额外的 "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_Qwen__Qwen2-0.5B_private", "harness_winogrande_5", split="train")
最新结果
- 最新结果来自 2024-05-30T13:41:43.370999 运行。
- 结果包括多个任务的准确率(acc)、标准化准确率(acc_norm)、误差(stderr)等指标。
配置详情
-
config_name: harness_arc_challenge_25
- 数据文件路径:
- 2024_05_30T13_41_43.370999 分割:
**/details_harness|arc:challenge|25_2024-05-30T13-41-43.370999.parquet - latest 分割:
**/details_harness|arc:challenge|25_2024-05-30T13-41-43.370999.parquet
- 2024_05_30T13_41_43.370999 分割:
- 数据文件路径:
-
config_name: harness_gsm8k_5
- 数据文件路径:
- 2024_05_30T13_41_43.370999 分割:
**/details_harness|gsm8k|5_2024-05-30T13-41-43.370999.parquet - latest 分割:
**/details_harness|gsm8k|5_2024-05-30T13-41-43.370999.parquet
- 2024_05_30T13_41_43.370999 分割:
- 数据文件路径:
-
config_name: harness_hellaswag_10
- 数据文件路径:
- 2024_05_30T13_41_43.370999 分割:
**/details_harness|hellaswag|10_2024-05-30T13-41-43.370999.parquet - latest 分割:
**/details_harness|hellaswag|10_2024-05-30T13-41-43.370999.parquet
- 2024_05_30T13_41_43.370999 分割:
- 数据文件路径:
-
config_name: harness_hendrycksTest_5
- 数据文件路径:
- 2024_05_30T13_41_43.370999 分割:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-anatomy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-astronomy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_biology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-college_physics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-computer_security|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-econometrics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-global_facts|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-human_aging|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-international_law|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-management|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-marketing|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-nutrition|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-philosophy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-prehistory|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-professional_law|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-public_relations|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-security_studies|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-sociology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-virology|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-world_religions|5_2024-05-30T13-41-43.370999.parquet
- latest 分割:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-anatomy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-astronomy|5_2024-05-30T13-41-43.370999.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-05-30T13-41-43.370999.parquet- `**/
- 2024_05_30T13_41_43.370999 分割:
- 数据文件路径:



