open-llm-leaderboard-old/details_voidful__phi-1_5_chat_32k
收藏Hugging Face2024-04-30 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_voidful__phi-1_5_chat_32k
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在模型voidful/phi-1_5_chat_32k的评估运行中自动创建的,包含63个配置,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。此外,results配置存储了所有运行的聚合结果,并用于计算和显示在Open LLM Leaderboard上的聚合指标。
该数据集是在模型voidful/phi-1_5_chat_32k的评估运行中自动创建的,包含63个配置,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。此外,results配置存储了所有运行的聚合结果,并用于计算和显示在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
数据集名称
- 名称: Evaluation run of voidful/phi-1_5_chat_32k
数据集组成
- 配置数量: 63个配置,每个配置对应一个评估任务。
- 创建来源: 数据集自动创建于模型voidful/phi-1_5_chat_32k在Open LLM Leaderboard上的评估运行。
- 运行次数: 数据集由1次运行创建,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。
- 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_voidful__phi-1_5_chat_32k", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-04-24T02:18:45.834669
- 结果示例:
python
{
"all": {
"acc": 0.40217596161576713,
"acc_stderr": 0.03427416656603839,
"acc_norm": 0.4036851303018283,
"acc_norm_stderr": 0.03503504014093307,
"mc1": 0.28151774785801714,
"mc1_stderr": 0.01574402724825605,
"mc2": 0.42804748959094835,
"mc2_stderr": 0.015074095328204674
},
"harness|arc:challenge|25": {
"acc": 0.47440273037542663,
"acc_stderr": 0.014592230885298967,
"acc_norm": 0.5051194539249146,
"acc_norm_stderr": 0.014610624890309157
},
"harness|hellaswag|10": {
"acc": 0.45429197371041624,
"acc_stderr": 0.004968888130290071,
"acc_norm": 0.5925114519020116,
"acc_norm_stderr": 0.004903628887264535
},
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_04_24T02_18_45.834669
- 路径: **/details_harness|arc:challenge|25_2024-04-24T02-18-45.834669.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-04-24T02-18-45.834669.parquet
- 分割: 2024_04_24T02_18_45.834669
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_04_24T02_18_45.834669
- 路径: **/details_harness|gsm8k|5_2024-04-24T02-18-45.834669.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-04-24T02-18-45.834669.parquet
- 分割: 2024_04_24T02_18_45.834669
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_04_24T02_18_45.834669
- 路径: **/details_harness|hellaswag|10_2024-04-24T02-18-45.834669.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-04-24T02-18-45.834669.parquet
- 分割: 2024_04_24T02_18_45.834669
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_04_24T02_18_45.834669
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-management|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-04-24T02-18-45.834669.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2024-04-24T02-18-45.83466
- 路径:
- 分割: 2024_04_24T02_18_45.834669
- 数据文件:



