five

open-llm-leaderboard-old/details_KnutJaegersberg__RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098

收藏
Hugging Face2023-10-27 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_KnutJaegersberg__RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098
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 KnutJaegersberg/RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个评估任务。它由 2 次运行创建,每次运行在每个配置中作为特定的分割找到。train 分割始终指向最新的结果。一个额外的配置 results 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。README 还提供了一个示例,展示了如何使用 Python 中的 datasets 库加载运行中的详细信息。

该数据集是在模型 KnutJaegersberg/RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个评估任务。它由 2 次运行创建,每次运行在每个配置中作为特定的分割找到。train 分割始终指向最新的结果。一个额外的配置 results 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。README 还提供了一个示例,展示了如何使用 Python 中的 datasets 库加载运行中的详细信息。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集来源

该数据集是在对模型 KnutJaegersberg/RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098 进行评估运行期间自动创建的,评估运行在 Open LLM Leaderboard 上进行。

数据集结构

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,用于计算并在 Open LLM Leaderboard 上显示聚合指标。

数据加载示例

以下是加载特定运行详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_KnutJaegersberg__RWKV-4-PilePlus-430M-20230520-6162-1018Gtokens-ctx4098", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-27T05:08:42.150075 运行的最新结果: python { "all": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415923, "f1": 0.05103817114093967, "f1_stderr": 0.0013185598027679899, "acc": 0.26317353140578503, "acc_stderr": 0.007674173740131278 }, "harness|drop|3": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415923, "f1": 0.05103817114093967, "f1_stderr": 0.0013185598027679899 }, "harness|gsm8k|5": { "acc": 0.002274450341167551, "acc_stderr": 0.0013121578148674266 }, "harness|winogrande|5": { "acc": 0.5240726124704025, "acc_stderr": 0.014036189665395129 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割:2023_09_12T08_47_54.050773
      • 路径:**/details_harness|arc:challenge|25_2023-09-12T08-47-54.050773.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-09-12T08-47-54.050773.parquet
  • harness_drop_3

    • 分割:2023_10_27T05_08_42.150075
      • 路径:**/details_harness|drop|3_2023-10-27T05-08-42.150075.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-27T05-08-42.150075.parquet
  • harness_gsm8k_5

    • 分割:2023_10_27T05_08_42.150075
      • 路径:**/details_harness|gsm8k|5_2023-10-27T05-08-42.150075.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-27T05-08-42.150075.parquet
  • harness_hellaswag_10

    • 分割:2023_09_12T08_47_54.050773
      • 路径:**/details_harness|hellaswag|10_2023-09-12T08-47-54.050773.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-09-12T08-47-54.050773.parquet
  • harness_hendrycksTest_5

    • 分割:2023_09_12T08_47_54.050773
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-12T08-47-54.050773.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-12T08-47-54.050773.parquet
        • `**/details_
二维码
社区交流群
二维码
科研交流群
商业服务