open-llm-leaderboard/details_migtissera__SynthIA-7B-v1.5
收藏Hugging Face2023-11-09 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_migtissera__SynthIA-7B-v1.5
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migtissera/SynthIA-7B-v1.5在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含一次运行的结果,每次运行在每个配置中表示为特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了如何使用Python代码加载运行中的详细信息的示例,并包含了特定运行的最新结果。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集基本信息
- 名称: Evaluation run of migtissera/SynthIA-7B-v1.5
- 来源: 自动创建于模型 migtissera/SynthIA-7B-v1.5 在 Open LLM Leaderboard 的评估运行期间。
- 组成: 包含64个配置,每个配置对应一个评估任务。
- 创建: 从1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。"train" 分片始终指向最新的结果。
- 额外配置: "results" 配置存储所有运行的聚合结果,用于计算和显示 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_migtissera__SynthIA-7B-v1.5_public", "harness_winogrande_5", split="train")
最新结果
- 来源: 最新结果来自 2023-11-09T14:41:56.883085 运行。
- 详细指标: python { "all": { "acc": 0.6291968571108129, "acc_stderr": 0.03252538162461919, "acc_norm": 0.63804599014876, "acc_norm_stderr": 0.03323519542303871, "mc1": 0.35128518971848227, "mc1_stderr": 0.016711358163544403, "mc2": 0.5131996962275648, "mc2_stderr": 0.015337988977122931, "em": 0.1875, "em_stderr": 0.003997164044486006, "f1": 0.26010591442953035, "f1_stderr": 0.004042449995216609 }, "harness|arc:challenge|25": { "acc": 0.5870307167235495, "acc_stderr": 0.014388344935398324, "acc_norm": 0.6271331058020477, "acc_norm_stderr": 0.014131176760131172 }, "harness|hellaswag|10": { "acc": 0.6432981477793268, "acc_stderr": 0.0047804672709117705, "acc_norm": 0.833698466440948, "acc_norm_stderr": 0.0037159010850549967 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.04760952285695236, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695236 }, # 其他任务的详细指标... }
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分片: 2023_11_09T14_41_56.883085
- 路径: **/details_harness|arc:challenge|25_2023-11-09T14-41-56.883085.parquet
- 分片: latest
- 路径: **/details_harness|arc:challenge|25_2023-11-09T14-41-56.883085.parquet
- 分片: 2023_11_09T14_41_56.883085
- 数据文件:
-
配置名称: harness_drop_3
- 数据文件:
- 分片: 2023_11_09T14_41_56.883085
- 路径: **/details_harness|drop|3_2023-11-09T14-41-56.883085.parquet
- 分片: latest
- 路径: **/details_harness|drop|3_2023-11-09T14-41-56.883085.parquet
- 分片: 2023_11_09T14_41_56.883085
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分片: 2023_11_09T14_41_56.883085
- 路径: **/details_harness|gsm8k|5_2023-11-09T14-41-56.883085.parquet
- 分片: latest
- 路径: **/details_harness|gsm8k|5_2023-11-09T14-41-56.883085.parquet
- 分片: 2023_11_09T14_41_56.883085
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分片: 2023_11_09T14_41_56.883085
- 路径: **/details_harness|hellaswag|10_2023-11-09T14-41-56.883085.parquet
- 分片: latest
- 路径: **/details_harness|hellaswag|10_2023-11-09T14-41-56.883085.parquet
- 分片: 2023_11_09T14_41_56.883085
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分片: 2023_11_09T14_41_56.883085
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-management|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-11-09T14-41-56.883085.parquet
- **/details_harness|hendrycksTest-moral
- 路径:
- 分片: 2023_11_09T14_41_56.883085
- 数据文件:



