five

open-llm-leaderboard-old/details_totally-not-an-llm__EverythingLM-13b-V3-16k

收藏
Hugging Face2023-10-26 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_totally-not-an-llm__EverythingLM-13b-V3-16k
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[totally-not-an-llm/EverythingLM-13b-V3-16k](https://huggingface.co/totally-not-an-llm/EverythingLM-13b-V3-16k)的评估过程中自动创建的,用于在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行评估。数据集由64个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行的结果都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。此外,还提供了一个名为“results”的配置,用于存储所有运行的聚合结果,并在Open LLM Leaderboard上显示聚合指标。

该数据集是在模型[totally-not-an-llm/EverythingLM-13b-V3-16k](https://huggingface.co/totally-not-an-llm/EverythingLM-13b-V3-16k)的评估过程中自动创建的,用于在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行评估。数据集由64个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行的结果都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。此外,还提供了一个名为“results”的配置,用于存储所有运行的聚合结果,并在Open LLM Leaderboard上显示聚合指标。
提供机构:
open-llm-leaderboard-old
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 totally-not-an-llm/EverythingLM-13b-V3-16kOpen LLM Leaderboard 上的自动创建的。

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。
  • 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 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_totally-not-an-llm__EverythingLM-13b-V3-16k", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-26T22:05:52.208686 运行的最新结果: python { "all": { "em": 0.00975251677852349, "em_stderr": 0.0010063982618519684, "f1": 0.07061661073825479, "f1_stderr": 0.001617936937117915, "acc": 0.3634471116993872, "acc_stderr": 0.008309152048639166 }, "harness|drop|3": { "em": 0.00975251677852349, "em_stderr": 0.0010063982618519684, "f1": 0.07061661073825479, "f1_stderr": 0.001617936937117915 }, "harness|gsm8k|5": { "acc": 0.019711902956785442, "acc_stderr": 0.0038289829787357173 }, "harness|winogrande|5": { "acc": 0.7071823204419889, "acc_stderr": 0.012789321118542613 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割:2023_10_04T00_03_41.509774
      • 路径:**/details_harness|arc:challenge|25_2023-10-04T00-03-41.509774.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-10-04T00-03-41.509774.parquet
  • harness_drop_3

    • 分割:2023_10_26T22_05_52.208686
      • 路径:**/details_harness|drop|3_2023-10-26T22-05-52.208686.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-26T22-05-52.208686.parquet
  • harness_gsm8k_5

    • 分割:2023_10_26T22_05_52.208686
      • 路径:**/details_harness|gsm8k|5_2023-10-26T22-05-52.208686.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-26T22-05-52.208686.parquet
  • harness_hellaswag_10

    • 分割:2023_10_04T00_03_41.509774
      • 路径:**/details_harness|hellaswag|10_2023-10-04T00-03-41.509774.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-10-04T00-03-41.509774.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_04T00_03_41.509774
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-04T00-03-41.509774.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-04T00-03-41.509774.parquet
        • `**/details_harness|hendrycksTest-professional_law|5_
二维码
社区交流群
二维码
科研交流群
商业服务