five

open-llm-leaderboard/details_PeanutJar__LLaMa-2-PeanutButter_v18_A-7B

收藏
Hugging Face2023-09-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_PeanutJar__LLaMa-2-PeanutButter_v18_A-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Open LLM Leaderboard上对模型PeanutJar/LLaMa-2-PeanutButter_v18_A-7B进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2次运行创建,每次运行在每个配置中表示为特定的分割。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。文件还提供了一个使用Python中的datasets库加载运行细节的示例。

This dataset was automatically created when evaluating the model PeanutJar/LLaMa-2-PeanutButter_v18_A-7B on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to an evaluation task. The dataset is generated from two runs, where each run is represented as a specific split within each configuration. The `train` split always points to the most recent results. Additionally, there is a configuration named `results` that stores the aggregated results of all runs, which is used to calculate and display the aggregated metrics on the Open LLM Leaderboard. An example of loading run details using the datasets library in Python is also provided.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在评估模型 PeanutJar/LLaMa-2-PeanutButter_v18_A-7BOpen LLM Leaderboard 上的运行过程中自动创建的。

数据集组成

  • 数据集包含 64 个配置,每个配置对应一个评估任务。
  • 数据集从 2 次运行中创建,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" 分割始终指向最新的结果。
  • 额外的配置 "results" 存储所有运行结果的聚合,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_PeanutJar__LLaMa-2-PeanutButter_v18_A-7B", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-22T16:10:43.222323 运行的最新结果

python { "all": { "em": 0.0014681208053691276, "em_stderr": 0.0003921042190298372, "f1": 0.05780201342281884, "f1_stderr": 0.0013345722358967201, "acc": 0.41051883916267196, "acc_stderr": 0.009653462102026637 }, "harness|drop|3": { "em": 0.0014681208053691276, "em_stderr": 0.0003921042190298372, "f1": 0.05780201342281884, "f1_stderr": 0.0013345722358967201 }, "harness|gsm8k|5": { "acc": 0.07202426080363912, "acc_stderr": 0.007121147983537124 }, "harness|winogrande|5": { "acc": 0.7490134175217048, "acc_stderr": 0.01218577622051615 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|arc:challenge|25_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-09-02T09:37:14.213070.parquet
  • harness_drop_3

    • 分割:2023_09_22T16_10_43.222323
    • 路径:**/details_harness|drop|3_2023-09-22T16-10-43.222323.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-09-22T16-10-43.222323.parquet
  • harness_gsm8k_5

    • 分割:2023_09_22T16_10_43.222323
    • 路径:**/details_harness|gsm8k|5_2023-09-22T16-10-43.222323.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-09-22T16-10-43.222323.parquet
  • harness_hellaswag_10

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hellaswag|10_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-02T09:37:14.213070.parquet 等 50 个文件
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-02T09:37:14.213070.parquet 等 50 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-02T09:37:14.213070.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-02T09:37:14.213070.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_09_02T09_37_14.213070
    • 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-02
二维码
社区交流群
二维码
科研交流群
商业服务