five

open-llm-leaderboard/details_kashif__stack-llama-2

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_kashif__stack-llama-2
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型kashif/stack-llama-2时自动创建的,包含了64个配置,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,最新的结果存储在train分割中。此外,还有一个名为results的配置,用于存储所有运行的聚合结果。

This dataset was automatically generated during the evaluation of the model kashif/stack-llama-2. It contains 64 configurations, each corresponding to one evaluation task. The dataset is produced from two runs, where the results of each run are stored in separate splits, and the latest results are stored in the "train" split. Additionally, there is a configuration named "results" that stores the aggregated results across all runs.
提供机构:
open-llm-leaderboard
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 kashif/stack-llama-2Open LLM Leaderboard 上的运行过程中自动创建的。

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。
  • 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_kashif__stack-llama-2", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-22T20:19:04.146812 运行 的最新结果: python { "all": { "em": 0.0012583892617449664, "em_stderr": 0.00036305608931194434, "f1": 0.05443896812080537, "f1_stderr": 0.0012685965060744062, "acc": 0.4202036533620397, "acc_stderr": 0.010294487617119145 }, "harness|drop|3": { "em": 0.0012583892617449664, "em_stderr": 0.00036305608931194434, "f1": 0.05443896812080537, "f1_stderr": 0.0012685965060744062 }, "harness|gsm8k|5": { "acc": 0.10007581501137225, "acc_stderr": 0.008266274528685624 }, "harness|winogrande|5": { "acc": 0.7403314917127072, "acc_stderr": 0.012322700705552667 } }

配置详情

以下是数据集的配置详情:

配置列表

  • harness_arc_challenge_25

    • 分片: 2023_08_29T07_07_44.494010
    • 路径: **/details_harness|arc:challenge|25_2023-08-29T07:07:44.494010.parquet
    • 分片: latest
    • 路径: **/details_harness|arc:challenge|25_2023-08-29T07:07:44.494010.parquet
  • harness_drop_3

    • 分片: 2023_09_22T20_19_04.146812
    • 路径: **/details_harness|drop|3_2023-09-22T20-19-04.146812.parquet
    • 分片: latest
    • 路径: **/details_harness|drop|3_2023-09-22T20-19-04.146812.parquet
  • harness_gsm8k_5

    • 分片: 2023_09_22T20_19_04.146812
    • 路径: **/details_harness|gsm8k|5_2023-09-22T20-19-04.146812.parquet
    • 分片: latest
    • 路径: **/details_harness|gsm8k|5_2023-09-22T20-19-04.146812.parquet
  • harness_hellaswag_10

    • 分片: 2023_08_29T07_07_44.494010
    • 路径: **/details_harness|hellaswag|10_2023-08-29T07:07:44.494010.parquet
    • 分片: latest
    • 路径: **/details_harness|hellaswag|10_2023-08-29T07:07:44.494010.parquet
  • harness_hendrycksTest_5

    • 分片: 2023_08_29T07_07_44.494010
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-management|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-08-29T07:07:44.494010.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-29T07:07:44.494010.parquet
      • `**/details_harness|hendrycksTest-professional_law|5_2023-08-29T07:07:4
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作