five

open-llm-leaderboard-old/details_lodrick-the-lafted__Hermes-Instruct-7B-100K

收藏
Hugging Face2024-02-20 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_lodrick-the-lafted__Hermes-Instruct-7B-100K
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[lodrick-the-lafted/Hermes-Instruct-7B-100K](https://huggingface.co/lodrick-the-lafted/Hermes-Instruct-7B-100K)时自动生成的,评估过程在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置中找到,分割以运行的时间戳命名。此外,数据集还包含一个名为"results"的配置,用于存储所有运行的聚合结果,并用于计算和显示在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上的聚合指标。

该数据集是在评估模型[lodrick-the-lafted/Hermes-Instruct-7B-100K](https://huggingface.co/lodrick-the-lafted/Hermes-Instruct-7B-100K)时自动生成的,评估过程在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置中找到,分割以运行的时间戳命名。此外,数据集还包含一个名为"results"的配置,用于存储所有运行的聚合结果,并用于计算和显示在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上的聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集摘要

该数据集是在对模型 lodrick-the-lafted/Hermes-Instruct-7B-100K 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集组成

  • 数据集由 63 个配置组成,每个配置对应一个评估任务。
  • 数据集从 1 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_lodrick-the-lafted__Hermes-Instruct-7B-100K", "harness_winogrande_5", split="train")

最新结果

以下是 2024-02-20T08:08:38.814549 运行的最新结果:

python { "all": { "acc": 0.6090324026750107, "acc_stderr": 0.03306627928563301, "acc_norm": 0.6132702932279503, "acc_norm_stderr": 0.0337325206267565, "mc1": 0.4614443084455324, "mc1_stderr": 0.017451384104637455, "mc2": 0.6362212933287348, "mc2_stderr": 0.015296863707374602 }, "harness|arc:challenge|25": { "acc": 0.5648464163822525, "acc_stderr": 0.014487986197186038, "acc_norm": 0.6151877133105802, "acc_norm_stderr": 0.014218371065251105 }, "harness|hellaswag|10": { "acc": 0.6383190599482175, "acc_stderr": 0.004795051037917733, "acc_norm": 0.8284206333399721, "acc_norm_stderr": 0.0037624392841951065 }, ... }

配置详情

  • harness_arc_challenge_25

    • 分割:2024_02_20T08_08_38.814549
    • 路径:**/details_harness|arc:challenge|25_2024-02-20T08-08-38.814549.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2024-02-20T08-08-38.814549.parquet
  • harness_gsm8k_5

    • 分割:2024_02_20T08_08_38.814549
    • 路径:**/details_harness|gsm8k|5_2024-02-20T08-08-38.814549.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2024-02-20T08-08-38.814549.parquet
  • harness_hellaswag_10

    • 分割:2024_02_20T08_08_38.814549
    • 路径:**/details_harness|hellaswag|10_2024-02-20T08-08-38.814549.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2024-02-20T08-08-38.814549.parquet
  • harness_hendrycksTest_5

    • 分割:2024_02_20T08_08_38.814549
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-02-20T08-08-38.814549.parquet
      • ...
    • 分割:latest
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-02-20T08-08-38.814549.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-02-20T08-08-38.814549.parquet
      • ...
二维码
社区交流群
二维码
科研交流群
商业服务