open-llm-leaderboard/details_anhnv125__pygmalion-6b-roleplay
收藏Hugging Face2023-09-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_anhnv125__pygmalion-6b-roleplay
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anhnv125/pygmalion-6b-roleplay时自动创建的,评估过程在Open LLM Leaderboard上进行。数据集包含64个配置,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。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,用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集创建
- 创建背景:该数据集是在评估模型 anhnv125/pygmalion-6b-roleplay 在 Open LLM Leaderboard 上的自动创建的。
数据集结构
- 配置数量:数据集包含 64 个配置,每个配置对应一个评估任务。
- 创建来源: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片存在,分片名称使用运行的时间戳。
- 最新结果:"train" 分片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_anhnv125__pygmalion-6b-roleplay", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源:这些是最新结果来自 2023-09-17T05:57:52.375499。
- 结果示例: python { "all": { "em": 0.002936241610738255, "em_stderr": 0.0005541113054709947, "f1": 0.05561765939597344, "f1_stderr": 0.0013547337231371388, "acc": 0.3190247209594698, "acc_stderr": 0.008257334480912119 }, "harness|drop|3": { "em": 0.002936241610738255, "em_stderr": 0.0005541113054709947, "f1": 0.05561765939597344, "f1_stderr": 0.0013547337231371388 }, "harness|gsm8k|5": { "acc": 0.011372251705837756, "acc_stderr": 0.002920666198788722 }, "harness|winogrande|5": { "acc": 0.6266771902131019, "acc_stderr": 0.013594002763035516 } }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分片:2023_08_01T14_17_43.702617
- 路径:**/details_harness|arc:challenge|25_2023-08-01T14:17:43.702617.parquet
- 分片:latest
- 路径:**/details_harness|arc:challenge|25_2023-08-01T14:17:43.702617.parquet
- 分片:2023_08_01T14_17_43.702617
- 数据文件:
-
配置名称:harness_drop_3
- 数据文件:
- 分片:2023_09_17T05_57_52.375499
- 路径:**/details_harness|drop|3_2023-09-17T05-57-52.375499.parquet
- 分片:latest
- 路径:**/details_harness|drop|3_2023-09-17T05-57-52.375499.parquet
- 分片:2023_09_17T05_57_52.375499
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分片:2023_09_17T05_57_52.375499
- 路径:**/details_harness|gsm8k|5_2023-09-17T05-57-52.375499.parquet
- 分片:latest
- 路径:**/details_harness|gsm8k|5_2023-09-17T05-57-52.375499.parquet
- 分片:2023_09_17T05_57_52.375499
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分片:2023_08_01T14_17_43.702617
- 路径:**/details_harness|hellaswag|10_2023-08-01T14:17:43.702617.parquet
- 分片:latest
- 路径:**/details_harness|hellaswag|10_2023-08-01T14:17:43.702617.parquet
- 分片:2023_08_01T14_17_43.702617
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分片:2023_08_01T14_17_43.702617
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-management|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-08-01T14:17:43.702617.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-08-01T14:17:43.702617.par
- 路径:
- 分片:2023_08_01T14_17_43.702617
- 数据文件:



