open-llm-leaderboard/details_facebook__opt-2.7b

该数据集是在Open LLM Leaderboard上评估模型facebook/opt-2.7b时自动生成的。数据集由64个配置组成，每个配置对应一个被评估的任务。数据集由2次运行生成，每次运行在每个配置中表示为特定的分割，分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外，名为results的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还提供了一个使用Python中的datasets库加载运行细节的示例。

This dataset was automatically generated when evaluating the model facebook/opt-2.7b on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluated task. The dataset is generated from 2 runs, where each run is represented as a specific split under each configuration, and the split names are named after the timestamp of the run. The train split always points to the most recent results. Additionally, the configuration named 'results' stores the aggregated results from all runs, which are used to calculate and display the aggregated metrics on the Open LLM Leaderboard. The README also provides an example of loading run details using the datasets library in Python.

原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 facebook/opt-2.7b 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。

数据集结构

• 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_facebook__opt-2.7b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-19T03:26:05.209079 运行的最新结果： python { "all": { "em": 0.0010486577181208054, "em_stderr": 0.0003314581465219369, "f1": 0.04767407718120815, "f1_stderr": 0.0011986644527763738, "acc": 0.31092412335527203, "acc_stderr": 0.007478442861762106 }, "harness|drop|3": { "em": 0.0010486577181208054, "em_stderr": 0.0003314581465219369, "f1": 0.04767407718120815, "f1_stderr": 0.0011986644527763738 }, "harness|gsm8k|5": { "acc": 0.002274450341167551, "acc_stderr": 0.0013121578148673927 }, "harness|winogrande|5": { "acc": 0.6195737963693765, "acc_stderr": 0.01364472790865682 } }

配置详情

• harness_arc_challenge_25

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|arc:challenge|25_2023-07-19T16:25:28.050181.parquet

• harness_drop_3

  • 分割: 2023_10_19T03_26_05.209079, latest
  • 路径: **/details_harness|drop|3_2023-10-19T03-26-05.209079.parquet

• harness_gsm8k_5

  • 分割: 2023_10_19T03_26_05.209079, latest
  • 路径: **/details_harness|gsm8k|5_2023-10-19T03-26-05.209079.parquet

• harness_hellaswag_10

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hellaswag|10_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: 多个文件路径，包含不同学科的评估结果

• harness_hendrycksTest_abstract_algebra_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_anatomy_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_astronomy_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_biology_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_college_physics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_computer_security_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_econometrics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T16:25:28.050181.parquet

• harness_hendrycksTest_elementary_mathematics_5

  • 分割: 2023_07_19T16_25_28.050181, latest
  • 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T16:25:28.050181.parquet