five

open-llm-leaderboard-old/details_nnheui__stablelm-2-1_6b-sft-full

收藏
Hugging Face2024-04-23 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_nnheui__stablelm-2-1_6b-sft-full
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[nnheui/stablelm-2-1_6b-sft-full](https://huggingface.co/nnheui/stablelm-2-1_6b-sft-full)时自动创建的,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,分割名称使用运行的时间戳。"train"分割始终指向最新结果。此外,"results"配置存储了所有运行的聚合结果,并用于计算和显示[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[nnheui/stablelm-2-1_6b-sft-full](https://huggingface.co/nnheui/stablelm-2-1_6b-sft-full)时自动创建的,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,分割名称使用运行的时间戳。"train"分割始终指向最新结果。此外,"results"配置存储了所有运行的聚合结果,并用于计算和显示[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

数据集简介

该数据集是在对模型 nnheui/stablelm-2-1_6b-sft-full 进行评估运行期间自动创建的,用于 Open LLM Leaderboard 上的评估任务。

数据集结构

  • 配置数量:63个配置,每个配置对应一个评估任务。
  • 数据来源:数据集由1次运行创建,每个运行结果存储在特定的分割中,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_nnheui__stablelm-2-1_6b-sft-full", "harness_winogrande_5", split="train")

最新结果

以下是 2024-04-23T10:00:27.031368 运行的最新结果

python { "all": { "acc": 0.37189736566424875, "acc_stderr": 0.0339564640775908, "acc_norm": 0.3748693484028965, "acc_norm_stderr": 0.0347188349146172, "mc1": 0.2521419828641371, "mc1_stderr": 0.015201522246299974, "mc2": 0.3957305143230597, "mc2_stderr": 0.014270774910986446 }, "harness|arc:challenge|25": { "acc": 0.39590443686006827, "acc_stderr": 0.014291228393536587, "acc_norm": 0.4197952218430034, "acc_norm_stderr": 0.014422181226303026 }, "harness|hellaswag|10": { "acc": 0.5155347540330611, "acc_stderr": 0.004987372476207028, "acc_norm": 0.6937860983867755, "acc_norm_stderr": 0.004599776866717475 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.36, "acc_stderr": 0.04824181513244218, "acc_norm": 0.36, "acc_norm_stderr": 0.04824181513244218 }, ... }

配置详情

  • 配置名称harness_arc_challenge_25

    • 数据文件
      • 分割:2024_04_23T10_00_27.031368
        • 路径:**/details_harness|arc:challenge|25_2024-04-23T10-00-27.031368.parquet
      • 分割:latest
        • 路径:**/details_harness|arc:challenge|25_2024-04-23T10-00-27.031368.parquet
  • 配置名称harness_gsm8k_5

    • 数据文件
      • 分割:2024_04_23T10_00_27.031368
        • 路径:**/details_harness|gsm8k|5_2024-04-23T10-00-27.031368.parquet
      • 分割:latest
        • 路径:**/details_harness|gsm8k|5_2024-04-23T10-00-27.031368.parquet
  • 配置名称harness_hellaswag_10

    • 数据文件
      • 分割:2024_04_23T10_00_27.031368
        • 路径:**/details_harness|hellaswag|10_2024-04-23T10-00-27.031368.parquet
      • 分割:latest
        • 路径:**/details_harness|hellaswag|10_2024-04-23T10-00-27.031368.parquet
  • 配置名称harness_hendrycksTest_5

    • 数据文件
      • 分割:2024_04_23T10_00_27.031368
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-management|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-04-23T10-00-27.031368.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-04-23T10-00-27.031368.parquet
          • `**/details_harness|hendrycksTest-professional_accounting|5_2024-04-23T10-00-27.0
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