five

open-llm-leaderboard-old/details_Eric111__CatunaMayo

收藏
Hugging Face2024-02-22 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Eric111__CatunaMayo
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eric111/CatunaMayo进行评估运行时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行在每个配置中表示为特定的拆分,train拆分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,用于计算和显示Open LLM Leaderboard上的聚合指标。README文件还提供了如何使用`datasets`库中的`load_dataset`函数加载运行中的详细信息的说明。最新运行的结果以JSON格式提供,显示了不同任务的各种准确性指标。

该数据集是在Open LLM Leaderboard上对模型eric111/CatunaMayo进行评估运行时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行在每个配置中表示为特定的拆分,train拆分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,用于计算和显示Open LLM Leaderboard上的聚合指标。README文件还提供了如何使用`datasets`库中的`load_dataset`函数加载运行中的详细信息的说明。最新运行的结果以JSON格式提供,显示了不同任务的各种准确性指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在对模型 eric111/CatunaMayo 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。数据集包含 63 个配置,每个配置对应一个评估任务。

数据集结构

  • 配置数量:63 个配置
  • 数据来源:从 2 次运行中创建,每次运行对应一个特定分片,分片名称使用运行的时间戳。
  • 最新结果:"train" 分片始终指向最新结果。
  • 汇总结果:额外配置 "results" 存储所有运行结果的汇总,用于计算和显示 Open LLM Leaderboard 上的聚合指标。

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_eric111__CatunaMayo", "harness_winogrande_5", split="train")

最新结果

以下是 2024-02-22T12:19:10.948803 运行的最新结果

python { "all": { "acc": 0.6575531238421152, "acc_stderr": 0.031937472435679695, "acc_norm": 0.6570510728428876, "acc_norm_stderr": 0.03260291830758222, "mc1": 0.5507955936352509, "mc1_stderr": 0.0174129419861153, "mc2": 0.6996030299637043, "mc2_stderr": 0.014680491046789042 }, "harness|arc:challenge|25": { "acc": 0.6953924914675768, "acc_stderr": 0.01344952210993249, "acc_norm": 0.7175767918088737, "acc_norm_stderr": 0.013155456884097224 }, "harness|hellaswag|10": { "acc": 0.6970722963553077, "acc_stderr": 0.004585850835623563, "acc_norm": 0.879008165704043, "acc_norm_stderr": 0.0032545129328064 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.38, "acc_stderr": 0.04878317312145632, "acc_norm": 0.38, "acc_norm_stderr": 0.04878317312145632 }, # 其他任务结果省略... }

配置详情

  • harness_arc_challenge_25

    • 分片:2024_02_21T00_21_24.620953, 2024_02_22T12_19_10.948803, latest
    • 路径:**/details_harness|arc:challenge|25_2024-02-21T00-21-24.620953.parquet, **/details_harness|arc:challenge|25_2024-02-22T12-19-10.948803.parquet
  • harness_gsm8k_5

    • 分片:2024_02_21T00_21_24.620953, 2024_02_22T12_19_10.948803, latest
    • 路径:**/details_harness|gsm8k|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|gsm8k|5_2024-02-22T12-19-10.948803.parquet
  • harness_hellaswag_10

    • 分片:2024_02_21T00_21_24.620953, 2024_02_22T12_19_10.948803, latest
    • 路径:**/details_harness|hellaswag|10_2024-02-21T00-21-24.620953.parquet, **/details_harness|hellaswag|10_2024-02-22T12-19-10.948803.parquet
  • harness_hendrycksTest_5

    • 分片:2024_02_21T00_21_24.620953, 2024_02_22T12_19_10.948803, latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-anatomy|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-astronomy|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-business_ethics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_biology|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_chemistry|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_computer_science|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_mathematics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_medicine|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-college_physics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-computer_security|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-econometrics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-formal_logic|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-global_facts|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_biology|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_geography|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_physics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-human_aging|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-human_sexuality|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-international_law|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-jurisprudence|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-machine_learning|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-management|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-marketing|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-medical_genetics|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-miscellaneous|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-moral_disputes|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-nutrition|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-philosophy|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-prehistory|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-professional_accounting|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-professional_law|5_2024-02-21T00-21-24.620953.parquet, **/details_harness|hendrycksTest-professional_medicine|5_2024-02-21T00-21-24.620953.parquet, `**/details_harness|hendrycksTest-professional_psychology|5_2024
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作