five

open-llm-leaderboard/details_lgaalves__gpt2_platypus-camel_physics

收藏
Hugging Face2023-10-25 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_lgaalves__gpt2_platypus-camel_physics
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 lgaalves/gpt2_platypus-camel_physics 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。它由 2 次运行生成,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。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 `datasets` 库中的 `load_dataset` 函数加载数据集的示例。

This dataset was automatically created during the evaluation run of the model lgaalves/gpt2_platypus-camel_physics on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluated task. It is generated from 2 runs, where each run is represented as a specific split within each configuration and named with the timestamp of the run. The train split always points to the most recent results. An additional configuration named `results` stores the aggregated results across all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. The README also provides examples of how to use the `load_dataset` function from the `datasets` library to load the dataset.
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 lgaalves/gpt2_platypus-camel_physicsOpen LLM Leaderboard 上的自动创建的。

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,每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
  • "train" 分片始终指向最新的结果。
  • 一个额外的配置 "results" 存储所有运行的聚合结果,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_lgaalves__gpt2_platypus-camel_physics", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-25T13:50:32.288438 运行的最新结果: python { "all": { "em": 0.002307046979865772, "em_stderr": 0.0004913221265094493, "f1": 0.04785339765100675, "f1_stderr": 0.001366270058429369, "acc": 0.24822415153906865, "acc_stderr": 0.007026065573457936 }, "harness|drop|3": { "em": 0.002307046979865772, "em_stderr": 0.0004913221265094493, "f1": 0.04785339765100675, "f1_stderr": 0.001366270058429369 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.4964483030781373, "acc_stderr": 0.014052131146915873 } }

配置详情

  • harness_arc_challenge_25

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|arc:challenge|25_2023-09-11T15-51-24.784876.parquet
  • harness_drop_3

    • 分片: 2023_10_25T13_50_32.288438, latest
    • 路径: **/details_harness|drop|3_2023-10-25T13-50-32.288438.parquet
  • harness_gsm8k_5

    • 分片: 2023_10_25T13_50_32.288438, latest
    • 路径: **/details_harness|gsm8k|5_2023-10-25T13-50-32.288438.parquet
  • harness_hellaswag_10

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hellaswag|10_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: 多个路径,例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-11T15-51-24.784876.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分片: 2023_09_11T15_51_24.784876, latest
    • 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-11T15-51-24.784876.parquet
二维码
社区交流群
二维码
科研交流群
商业服务