open-llm-leaderboard-old/details_perlthoughts__Chupacabra-16B-v2.01
收藏Hugging Face2023-12-09 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_perlthoughts__Chupacabra-16B-v2.01
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perlthoughts/Chupacabra-16B-v2.01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。数据集由2次运行生成,每次运行的结果存储在不同的分割中,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,还有一个名为results的配置,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Open LLM Leaderboard上对模型perlthoughts/Chupacabra-16B-v2.01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。数据集由2次运行生成,每次运行的结果存储在不同的分割中,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,还有一个名为results的配置,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
数据集创建
- 创建背景:该数据集是在对模型
perlthoughts/Chupacabra-16B-v2.01进行评估运行期间自动创建的。 - 评估平台:数据集的创建基于 Open LLM Leaderboard。
数据集组成
- 配置数量:数据集包含 63 个配置,每个配置对应一个评估任务。
- 运行次数:数据集从 2 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。
- 最新结果:"train" 分割始终指向最新的结果。
- 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_perlthoughts__Chupacabra-16B-v2.01", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源:这些是最新结果,来自 2023-12-09T17:22:21.064406 的运行。
- 结果示例: python { "all": { "acc": 0.6291644824323459, "acc_stderr": 0.03274488948991142, "acc_norm": 0.6364297874777888, "acc_norm_stderr": 0.033437033730375994, "mc1": 0.4920440636474908, "mc1_stderr": 0.01750128507455183, "mc2": 0.6453446072596959, "mc2_stderr": 0.015246087407331466 }, "harness|arc:challenge|25": { "acc": 0.6296928327645052, "acc_stderr": 0.01411129875167495, "acc_norm": 0.6535836177474402, "acc_norm_stderr": 0.013905011180063232 }, "harness|hellaswag|10": { "acc": 0.6387173869747063, "acc_stderr": 0.004793904922401889, "acc_norm": 0.8292172873929496, "acc_norm_stderr": 0.003755498941781851 }, ... }
配置详情
-
配置名称:
harness_arc_challenge_25- 数据文件:
- 分割
2023_12_08T02_08_47.844785- 路径:
**/details_harness|arc:challenge|25_2023-12-08T02-08-47.844785.parquet
- 路径:
- 分割
2023_12_09T17_22_21.064406- 路径:
**/details_harness|arc:challenge|25_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|arc:challenge|25_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
- 数据文件:
-
配置名称:
harness_gsm8k_5- 数据文件:
- 分割
2023_12_08T02_08_47.844785- 路径:
**/details_harness|gsm8k|5_2023-12-08T02-08-47.844785.parquet
- 路径:
- 分割
2023_12_09T17_22_21.064406- 路径:
**/details_harness|gsm8k|5_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|gsm8k|5_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
- 数据文件:
-
配置名称:
harness_hellaswag_10- 数据文件:
- 分割
2023_12_08T02_08_47.844785- 路径:
**/details_harness|hellaswag|10_2023-12-08T02-08-47.844785.parquet
- 路径:
- 分割
2023_12_09T17_22_21.064406- 路径:
**/details_harness|hellaswag|10_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|hellaswag|10_2023-12-09T17-22-21.064406.parquet
- 路径:
- 分割
- 数据文件:
-
配置名称:
harness_hendrycksTest_5- 数据文件:
- 分割
2023_12_08T02_08_47.844785- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-anatomy|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-astronomy|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_biology|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-college_physics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-computer_security|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-econometrics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-global_facts|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-human_aging|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-international_law|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-management|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-marketing|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-12-08T02-08-47.844785.parquet**/details_harness|hendrycksTest-nutrition|5_2023-12-08T02-08-47.844785.parquet- `**/details_harness|hendrycksTest-philosophy|5_2023-12-08T02-08-
- 路径:
- 分割
- 数据文件:



