five

open-llm-leaderboard/details_garage-bAInd__Platypus-30B

收藏
Hugging Face2023-10-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_garage-bAInd__Platypus-30B
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garage-bAInd/Platypus-30B在Open LLM Leaderboard上的评估运行过程中自动创建的。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次运行生成,每次运行的结果存储在不同的分割中,分割名称使用运行的时间戳命名。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,并用于计算和显示Open LLM Leaderboard上的聚合指标。

This dataset was automatically created during the evaluation run of the model garage-bAInd/Platypus-30B on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluation task. The dataset is generated from two runs, with the results of each run stored in a separate split named after the timestamp of the run. The `train` split always points to the most recent results. In addition, there is a configuration named `results` that stores the aggregated results of all runs and is used to calculate and display the aggregate metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 garage-bAInd/Platypus-30BOpen LLM Leaderboard 上的自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。

数据集结构

  • 配置数量:64 个配置
  • 运行次数:数据集从 2 次运行中创建。每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。
  • 训练分割:"train" 分割始终指向最新的结果。
  • 结果配置: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_garage-bAInd__Platypus-30B", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-17T11:36:48.656680 运行的最新结果: python { "all": { "em": 0.4152684563758389, "em_stderr": 0.005046408282247135, "f1": 0.4565257969798663, "f1_stderr": 0.004890389225361096, "acc": 0.4788908748525736, "acc_stderr": 0.010306994464370747 }, "harness|drop|3": { "em": 0.4152684563758389, "em_stderr": 0.005046408282247135, "f1": 0.4565257969798663, "f1_stderr": 0.004890389225361096 }, "harness|gsm8k|5": { "acc": 0.14404852160727824, "acc_stderr": 0.009672110973065282 }, "harness|winogrande|5": { "acc": 0.813733228097869, "acc_stderr": 0.010941877955676211 } }

配置详情

以下是部分配置及其数据文件路径:

  • harness_arc_challenge_25

    • 分割:2023_08_22T14_57_43.849594
      • 路径:**/details_harness|arc:challenge|25_2023-08-22T14:57:43.849594.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-08-22T14:57:43.849594.parquet
  • harness_drop_3

    • 分割:2023_10_17T11_36_48.656680
      • 路径:**/details_harness|drop|3_2023-10-17T11-36-48.656680.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-17T11-36-48.656680.parquet
  • harness_gsm8k_5

    • 分割:2023_10_17T11_36_48.656680
      • 路径:**/details_harness|gsm8k|5_2023-10-17T11-36-48.656680.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-17T11-36-48.656680.parquet
  • harness_hellaswag_10

    • 分割:2023_08_22T14_57_43.849594
      • 路径:**/details_harness|hellaswag|10_2023-08-22T14:57:43.849594.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-08-22T14:57:43.849594.parquet
  • harness_hendrycksTest_5

    • 分割:2023_08_22T14_57_43.849594
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-formal_logic|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-global_facts|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_biology|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_geography|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_physics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-human_aging|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-human_sexuality|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-international_law|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-jurisprudence|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-machine_learning|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-management|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-marketing|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-medical_genetics|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-miscellaneous|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-moral_disputes|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-nutrition|5_2023-08-22T14:57:43.849594.parquet
      • 路径:**/details_harness|hendrycksTest-philosophy|5_2023-08-22T14:57:43.849594.parquet
      • 路径:`**/details_har
二维码
社区交流群
二维码
科研交流群
商业服务