open-llm-leaderboard-old/details_vishnukv__speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b
收藏Hugging Face2024-03-05 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_vishnukv__speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b
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[vishnukv/speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b](https://huggingface.co/vishnukv/speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b)评估运行期间自动创建的,用于[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)。数据集包含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分割始终指向最新结果。此外,还有一个results配置存储运行中的所有聚合结果(用于计算并在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上显示聚合指标)。
该数据集是在模型[vishnukv/speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b](https://huggingface.co/vishnukv/speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b)评估运行期间自动创建的,用于[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)。数据集包含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分割始终指向最新结果。此外,还有一个results配置存储运行中的所有聚合结果(用于计算并在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上显示聚合指标)。
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:该数据集是在模型 vishnukv/speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集结构
- 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。
- 数据来源:数据集来自 1 次运行,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。"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_vishnukv__speechless-mistral-dolphin-orca-platypus-samantha-WestSeverusJaskier-7b", "harness_winogrande_5", split="train")
最新结果
- 结果来源:这些是最新结果来自 2024-03-05T11:15:11.562452 运行。
- 详细结果: python { "all": { "acc": 0.6519496556579889, "acc_stderr": 0.031826906468891074, "acc_norm": 0.6533873210528282, "acc_norm_stderr": 0.03246586322235606, "mc1": 0.4222766217870257, "mc1_stderr": 0.017290733254248177, "mc2": 0.5990304320814928, "mc2_stderr": 0.01501318392535001 }, "harness|arc:challenge|25": { "acc": 0.64419795221843, "acc_stderr": 0.013990571137918762, "acc_norm": 0.6800341296928327, "acc_norm_stderr": 0.01363134580701619 }, "harness|hellaswag|10": { "acc": 0.6739693288189603, "acc_stderr": 0.0046780064036917165, "acc_norm": 0.8655646285600478, "acc_norm_stderr": 0.00340422332041737 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.045604802157206845, "acc_norm": 0.29, "acc_norm_stderr": 0.045604802157206845 }, ... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分片:2024_03_05T11_15_11.562452
- 路径:
**/details_harness|arc:challenge|25_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|arc:challenge|25_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:2024_03_05T11_15_11.562452
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分片:2024_03_05T11_15_11.562452
- 路径:
**/details_harness|gsm8k|5_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|gsm8k|5_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:2024_03_05T11_15_11.562452
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分片:2024_03_05T11_15_11.562452
- 路径:
**/details_harness|hellaswag|10_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|hellaswag|10_2024-03-05T11-15-11.562452.parquet
- 路径:
- 分片:2024_03_05T11_15_11.562452
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分片:2024_03_05T11_15_11.562452
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-anatomy|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-astronomy|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_biology|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-college_physics|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-computer_security|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-econometrics|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-03-05T11-15-11.562452.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-05T11-15-11.562452.parquet- ...
- 路径:
- 分片:2024_03_05T11_15_11.562452
- 数据文件:



