five

open-llm-leaderboard-old/details_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj

收藏
Hugging Face2023-10-18 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj
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[CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj](https://huggingface.co/CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj)在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/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次运行中创建的,每次运行都可以在每个配置的特定分割中找到,分割名称使用运行的时间戳命名。"train"分割始终指向最新结果。一个额外的配置"results"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。

该数据集是在模型[CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj](https://huggingface.co/CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj)在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/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次运行中创建的,每次运行都可以在每个配置的特定分割中找到,分割名称使用运行的时间戳命名。"train"分割始终指向最新结果。一个额外的配置"results"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

数据集来源

该数据集是在模型 CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_projOpen LLM Leaderboard 上的评估运行期间自动创建的。

数据集组成

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,每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。"train" 分片始终指向最新的结果。

额外配置

一个额外的配置 "results" 存储了所有运行的聚合结果,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。

数据加载示例

加载特定运行的详细信息,可以使用以下代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-gate_up_down_proj", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-18T10:14:58.167192 运行 的最新结果: python { "all": { "em": 0.10098573825503356, "em_stderr": 0.0030856947694457384, "f1": 0.15267407718120746, "f1_stderr": 0.0031959753495490175, "acc": 0.4472367612449459, "acc_stderr": 0.010567855433819127 }, "harness|drop|3": { "em": 0.10098573825503356, "em_stderr": 0.0030856947694457384, "f1": 0.15267407718120746, "f1_stderr": 0.0031959753495490175 }, "harness|gsm8k|5": { "acc": 0.1288855193328279, "acc_stderr": 0.009229580761400269 }, "harness|winogrande|5": { "acc": 0.7655880031570639, "acc_stderr": 0.011906130106237985 } }

配置详情

配置列表

  • harness_arc_challenge_25

    • 分片:2023_09_03T05_20_14.306293
      • 路径:**/details_harness|arc:challenge|25_2023-09-03T05:20:14.306293.parquet
    • 分片:latest
      • 路径:**/details_harness|arc:challenge|25_2023-09-03T05:20:14.306293.parquet
  • harness_drop_3

    • 分片:2023_10_18T10_14_58.167192
      • 路径:**/details_harness|drop|3_2023-10-18T10-14-58.167192.parquet
    • 分片:latest
      • 路径:**/details_harness|drop|3_2023-10-18T10-14-58.167192.parquet
  • harness_gsm8k_5

    • 分片:2023_10_18T10_14_58.167192
      • 路径:**/details_harness|gsm8k|5_2023-10-18T10-14-58.167192.parquet
    • 分片:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-18T10-14-58.167192.parquet
  • harness_hellaswag_10

    • 分片:2023_09_03T05_20_14.306293
      • 路径:**/details_harness|hellaswag|10_2023-09-03T05:20:14.306293.parquet
    • 分片:latest
      • 路径:**/details_harness|hellaswag|10_2023-09-03T05:20:14.306293.parquet
  • harness_hendrycksTest_5

    • 分片:2023_09_03T05_20_14.306293
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-formal_logic|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-global_facts|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_biology|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_european_history|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_geography|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_physics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_psychology|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_statistics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_us_history|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_world_history|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-human_aging|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-human_sexuality|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-international_law|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-jurisprudence|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-logical_fallacies|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-machine_learning|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-management|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-marketing|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-medical_genetics|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-miscellaneous|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-moral_disputes|5_2023-09-03T05:20:14.306293.parquet
      • 路径:**/details_harness|hendrycksTest-moral_scenarios|5_2023-09-03T05:20:14.306293.parquet
      • 路径:`**/details_harness|hendrycksTest-nutrition|5_2023
二维码
社区交流群
二维码
科研交流群
商业服务