five

open-llm-leaderboard-old/details_chargoddard__MelangeB-70b

收藏
Hugging Face2023-10-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_chargoddard__MelangeB-70b
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 chargoddard/MelangeB-70b 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。数据集是从 2 次运行中生成的,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新的结果。一个名为 results 的额外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 还提供了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集的示例。

该数据集是在模型 chargoddard/MelangeB-70b 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。数据集是从 2 次运行中生成的,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新的结果。一个名为 results 的额外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 还提供了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 chargoddard/MelangeB-70b 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 配置数量:64个配置,每个配置对应一个评估任务。
  • 运行次数:数据集从2次运行中创建。每个运行结果作为一个特定的分割(split)存储,分割名称使用运行的时间戳。
  • 最新结果:"train" 分割始终指向最新的结果。
  • 汇总结果:一个额外的配置 "results" 存储所有运行的汇总结果,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_chargoddard__MelangeB-70b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-17T13:18:04.928943 运行的最新结果: python { "all": { "em": 0.49958053691275167, "em_stderr": 0.005120466189311586, "f1": 0.5792397231543648, "f1_stderr": 0.004704767839498484, "acc": 0.570668027786471, "acc_stderr": 0.01156392378740017 }, "harness|drop|3": { "em": 0.49958053691275167, "em_stderr": 0.005120466189311586, "f1": 0.5792397231543648, "f1_stderr": 0.004704767839498484 }, "harness|gsm8k|5": { "acc": 0.3062926459438969, "acc_stderr": 0.0126969301065629 }, "harness|winogrande|5": { "acc": 0.835043409629045, "acc_stderr": 0.010430917468237438 } }

配置详情

以下是部分配置的详细信息:

配置 harness_arc_challenge_25

  • 分割:2023_08_23T14_27_52.893839
  • 路径
    • **/details_harness|arc:challenge|25_2023-08-23T14:27:52.893839.parquet
  • 最新分割
    • **/details_harness|arc:challenge|25_2023-08-23T14:27:52.893839.parquet

配置 harness_drop_3

  • 分割:2023_10_17T13_18_04.928943
  • 路径
    • **/details_harness|drop|3_2023-10-17T13-18-04.928943.parquet
  • 最新分割
    • **/details_harness|drop|3_2023-10-17T13-18-04.928943.parquet

配置 harness_gsm8k_5

  • 分割:2023_10_17T13_18_04.928943
  • 路径
    • **/details_harness|gsm8k|5_2023-10-17T13-18-04.928943.parquet
  • 最新分割
    • **/details_harness|gsm8k|5_2023-10-17T13-18-04.928943.parquet

配置 harness_hellaswag_10

  • 分割:2023_08_23T14_27_52.893839
  • 路径
    • **/details_harness|hellaswag|10_2023-08-23T14:27:52.893839.parquet
  • 最新分割
    • **/details_harness|hellaswag|10_2023-08-23T14:27:52.893839.parquet

配置 harness_hendrycksTest_5

  • 分割:2023_08_23T14_27_52.893839
  • 路径
    • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-anatomy|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-astronomy|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-business_ethics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_biology|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_medicine|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-college_physics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-computer_security|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-econometrics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-formal_logic|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-global_facts|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-human_aging|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-international_law|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-machine_learning|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-management|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-marketing|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-nutrition|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-philosophy|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-prehistory|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-23T14:27:52.893839.parquet
    • **/details_harness|hendrycksTest-professional_law|5_2023-08-23T14:27:52.893839.parquet
    • `**/details
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