open-llm-leaderboard-old/details_abhiramtirumala__DialoGPT-sarcastic-medium

该数据集是在评估模型abhiramtirumala/DialoGPT-sarcastic-medium时自动生成的，包含64个配置，每个配置对应一个评估任务。数据集由3次运行生成，每次运行的结果作为特定的分割存储在配置中，分割名称使用运行的时间戳。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。

原始信息汇总

数据集概述

数据集来源

该数据集是在评估模型 abhiramtirumala/DialoGPT-sarcastic-medium 在 Open LLM Leaderboard 上的运行过程中自动创建的。

数据集结构

• 该数据集包含 64 个配置，每个配置对应一个评估任务。
• 数据集从 3 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。

额外配置

• 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。

数据加载示例

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

最新结果

以下是来自 2023-09-23T00:36:45.634956 运行的最新结果： python { "all": { "em": 0.0, "em_stderr": 0.0, "f1": 0.0, "f1_stderr": 0.0, "acc": 0.26677190213101815, "acc_stderr": 0.007010413338799049 }, "harness|drop|3": { "em": 0.0, "em_stderr": 0.0, "f1": 0.0, "f1_stderr": 0.0 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.5335438042620363, "acc_stderr": 0.014020826677598098 } }

配置详情

• harness_arc_challenge_25

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|arc:challenge|25_2023-07-19T10:39:52.332273.parquet

• harness_drop_3

  • 分割：2023_09_17T18_48_39.393988, 2023_09_23T00_36_45.634956, latest
  • 路径：**/details_harness|drop|3_2023-09-17T18-48-39.393988.parquet, **/details_harness|drop|3_2023-09-23T00-36-45.634956.parquet

• harness_gsm8k_5

  • 分割：2023_09_17T18_48_39.393988, 2023_09_23T00_36_45.634956, latest
  • 路径：**/details_harness|gsm8k|5_2023-09-17T18-48-39.393988.parquet, **/details_harness|gsm8k|5_2023-09-23T00-36-45.634956.parquet

• harness_hellaswag_10

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hellaswag|10_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：多个路径，例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_abstract_algebra_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_anatomy_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-anatomy|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_astronomy_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-astronomy|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_biology_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_biology|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_college_physics_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-college_physics|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_computer_security_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-computer_security|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_econometrics_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-econometrics|5_2023-07-19T10:39:52.332273.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割：2023_07_19T10_39_52.332273, latest
  • 路径：**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T10:39:52.332273.parquet

以上是数据集的详细概述，包括数据集的来源、结构、加载示例、最新结果以及各个配置的详细信息。