five

open-llm-leaderboard-old/details_lvkaokao__llama2-7b-hf-chat-lora-v3

收藏
Hugging Face2023-09-16 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_lvkaokao__llama2-7b-hf-chat-lora-v3
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 lvkaokao/llama2-7b-hf-chat-lora-v3 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,每个配置对应一个被评估的任务。它由 2 次运行生成,每次运行在每个配置中表示为特定的拆分。train 拆分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 lvkaokao/llama2-7b-hf-chat-lora-v3 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,每个配置对应一个被评估的任务。它由 2 次运行生成,每次运行在每个配置中表示为特定的拆分。train 拆分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

数据集摘要

该数据集是在对模型 lvkaokao/llama2-7b-hf-chat-lora-v3 进行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。此外,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

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_lvkaokao__llama2-7b-hf-chat-lora-v3", "harness_winogrande_5", split="train")

最新结果

以下是来自2023年9月16日22:22:04.429370运行的最新结果: python { "all": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642966032, "f1": 0.05310088087248333, "f1_stderr": 0.0014130017638603535, "acc": 0.3891916037418029, "acc_stderr": 0.007656807657466876 }, "harness|drop|3": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642966032, "f1": 0.05310088087248333, "f1_stderr": 0.0014130017638603535 }, "harness|gsm8k|5": { "acc": 0.015163002274450341, "acc_stderr": 0.0033660229497263472 }, "harness|winogrande|5": { "acc": 0.7632202052091555, "acc_stderr": 0.011947592365207404 } }

配置详情

  • config_name: harness_arc_challenge_25

    • data_files:
      • split: 2023_08_24T04_41_09.477230
        • path: **/details_harness|arc:challenge|25_2023-08-24T04:41:09.477230.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2023-08-24T04:41:09.477230.parquet
  • config_name: harness_drop_3

    • data_files:
      • split: 2023_09_16T22_22_04.429370
        • path: **/details_harness|drop|3_2023-09-16T22-22-04.429370.parquet
      • split: latest
        • path: **/details_harness|drop|3_2023-09-16T22-22-04.429370.parquet
  • config_name: harness_gsm8k_5

    • data_files:
      • split: 2023_09_16T22_22_04.429370
        • path: **/details_harness|gsm8k|5_2023-09-16T22-22-04.429370.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2023-09-16T22-22-04.429370.parquet
  • config_name: harness_hellaswag_10

    • data_files:
      • split: 2023_08_24T04_41_09.477230
        • path: **/details_harness|hellaswag|10_2023-08-24T04:41:09.477230.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2023-08-24T04:41:09.477230.parquet
  • config_name: harness_hendrycksTest_5

    • data_files:
      • split: 2023_08_24T04_41_09.477230
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-international_law|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-management|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-marketing|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-24T04:41:09.477230.parquet
          • **/details_harness|hendrycksTest-professional_psychology|5_2023-08-24T
二维码
社区交流群
二维码
科研交流群
商业服务