open-llm-leaderboard-old/details_aloobun__CosmicNoodle-7B
收藏Hugging Face2024-04-16 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_aloobun__CosmicNoodle-7B
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在评估模型aloobun/CosmicNoodle-7B时自动创建的,评估是在Open LLM Leaderboard上进行的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置中找到,运行的时间戳用于命名分割。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,用于计算和显示Open LLM Leaderboard上的聚合指标。
该数据集是在评估模型aloobun/CosmicNoodle-7B时自动创建的,评估是在Open LLM Leaderboard上进行的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置中找到,运行的时间戳用于命名分割。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
aloobun/CosmicNoodle-7B进行评估运行期间自动创建的。 - 评估平台:评估运行在 Open LLM Leaderboard 上进行。
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。
- 运行次数:数据集来自 1 次运行。每个运行结果作为一个特定的分割(split)存储在每个配置中,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_aloobun__CosmicNoodle-7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳:2024-04-16T09:31:25.302515
- 结果示例: python { "all": { "acc": 0.6198367028680756, "acc_stderr": 0.03252727900809058, "acc_norm": 0.6203276102862048, "acc_norm_stderr": 0.033180985934238585, "mc1": 0.4039167686658507, "mc1_stderr": 0.01717727682258428, "mc2": 0.582137372154009, "mc2_stderr": 0.016148869509381367 }, "harness|arc:challenge|25": { "acc": 0.5784982935153583, "acc_stderr": 0.014430197069326025, "acc_norm": 0.5998293515358362, "acc_norm_stderr": 0.014317197787809164 }, "harness|hellaswag|10": { "acc": 0.6516630153355906, "acc_stderr": 0.0047546970133549625, "acc_norm": 0.8181637124078869, "acc_norm_stderr": 0.0038492126228151604 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.045604802157206845, "acc_norm": 0.29, "acc_norm_stderr": 0.045604802157206845 }, # 其他任务的结果... }
配置详情
-
配置名称:
harness_arc_challenge_25- 数据文件:
- 分割:
2024_04_16T09_31_25.302515- 路径:
**/details_harness|arc:challenge|25_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|arc:challenge|25_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_gsm8k_5- 数据文件:
- 分割:
2024_04_16T09_31_25.302515- 路径:
**/details_harness|gsm8k|5_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|gsm8k|5_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_hellaswag_10- 数据文件:
- 分割:
2024_04_16T09_31_25.302515- 路径:
**/details_harness|hellaswag|10_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
latest- 路径:
**/details_harness|hellaswag|10_2024-04-16T09-31-25.302515.parquet
- 路径:
- 分割:
- 数据文件:
-
配置名称:
harness_hendrycksTest_5- 数据文件:
- 分割:
2024_04_16T09_31_25.302515- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-anatomy|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-astronomy|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_biology|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-college_physics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-computer_security|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-econometrics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-global_facts|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-human_aging|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-international_law|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-management|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-marketing|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-nutrition|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-philosophy|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-prehistory|5_2024-04-16T09-31-25.302515.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-04-16T09-31-25.302515.parquet- `**/details_harness|hendrycksTest-professional_law|5_2024-04-16T09-
- 路径:
- 分割:
- 数据文件:



