open-llm-leaderboard-old/details_xxyyy123__10k_v1_lora_qk_rank14_v2
收藏Hugging Face2023-09-03 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_xxyyy123__10k_v1_lora_qk_rank14_v2
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xxyyy123/10k_v1_lora_qk_rank14_v2的评估运行期间自动创建的,用于在Open LLM Leaderboard上进行评估。数据集由61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,分割以运行的时间戳命名。train分割始终指向最新结果。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。
该数据集是在模型xxyyy123/10k_v1_lora_qk_rank14_v2的评估运行期间自动创建的,用于在Open LLM Leaderboard上进行评估。数据集由61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,分割以运行的时间戳命名。train分割始终指向最新结果。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
数据集信息
- 名称: Evaluation run of xxyyy123/10k_v1_lora_qk_rank14_v2
- 来源: 在模型 xxyyy123/10k_v1_lora_qk_rank14_v2 的评估运行期间自动创建
- 目的: 用于 Open LLM Leaderboard 的评估
数据集结构
- 配置数量: 61 个配置
- 每个配置对应: 一个评估任务
- 数据来源: 来自 1 次运行
- 分割命名: 使用运行的时间戳
- 训练分割: 指向最新结果
额外配置
- 结果配置: 存储所有运行结果的聚合,用于计算和显示聚合指标
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_xxyyy123__10k_v1_lora_qk_rank14_v2", "harness_truthfulqa_mc_0", split="train")
最新结果
- 最新结果来源: 2023-09-03T15:46:18.274387
- 结果示例:
python
{
"all": {
"acc": 0.5170296348361414,
"acc_stderr": 0.03493290232216538,
"acc_norm": 0.5207737377982975,
"acc_norm_stderr": 0.034916919339016556,
"mc1": 0.3574051407588739,
"mc1_stderr": 0.016776599676729398,
"mc2": 0.5241397415740128,
"mc2_stderr": 0.0157002252598079
},
"harness|arc:challenge|25": {
"acc": 0.5298634812286689,
"acc_stderr": 0.014585305840007105,
"acc_norm": 0.5648464163822525,
"acc_norm_stderr": 0.014487986197186043
},
"harness|hellaswag|10": {
"acc": 0.6100378410675165,
"acc_stderr": 0.004867445945277159,
"acc_norm": 0.7959569806811392,
"acc_norm_stderr": 0.004021769582317863
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.28,
"acc_stderr": 0.04512608598542128,
"acc_norm": 0.28,
"acc_norm_stderr": 0.04512608598542128
},
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: 2023_09_03T15_46_18.274387
- 路径: **/details_harness|arc:challenge|25_2023-09-03T15:46:18.274387.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2023-09-03T15:46:18.274387.parquet
- 分割: 2023_09_03T15_46_18.274387
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2023_09_03T15_46_18.274387
- 路径: **/details_harness|hellaswag|10_2023-09-03T15:46:18.274387.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2023-09-03T15:46:18.274387.parquet
- 分割: 2023_09_03T15_46_18.274387
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2023_09_03T15_46_18.274387
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-management|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2023-09-03T15:46:18.274387.parquet
- **/details_harness|hendrycksTest
- 路径:
- 分割: 2023_09_03T15_46_18.274387
- 数据文件:



