five

OALL/details_yunconglong__Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B

收藏
Hugging Face2024-07-13 更新2024-07-22 收录
下载链接:
https://hf-mirror.com/datasets/OALL/details_yunconglong__Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型评估过程中自动创建的,包含136个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果作为一个特定的分割存储在配置中,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,还有一个名为results的配置存储了所有运行的聚合结果。

Dataset automatically created during the evaluation run of model [yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B](https://huggingface.co/yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B). The dataset is composed of 136 configurations, each one corresponding to one of the evaluated tasks. The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run. The train split is always pointing to the latest results. An additional configuration results store all the aggregated results of the run. The dataset includes detailed accuracy metrics for various tasks, indicating performance in different domains.
提供机构:
OALL
原始信息汇总

数据集概述

数据集简介

  • 名称: Evaluation run of yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
  • 来源: 自动创建于模型评估过程中。
  • 组成: 包含136个配置,每个配置对应一个评估任务。
  • 创建方式: 从1次运行中生成,每次运行的结果存储在特定的分割中,分割名称使用运行的时间戳。
  • 额外配置: 包含一个名为“results”的配置,存储所有运行的聚合结果。

数据加载示例

python from datasets import load_dataset data = load_dataset("OALL/details_yunconglong__Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B", "lighteval_xstory_cloze_ar_0", split="train")

最新结果

  • 时间戳: 2024-07-13T20:18:19.316329
  • 结果概览:
    • 总体:
      • acc_norm: 0.4325651701366585
      • acc_norm_stderr: 0.03822270474174645
      • acc: 0.5823957643944407
      • acc_stderr: 0.012691211382848639
    • 具体任务:
      • community|acva:Algeria|0:
        • acc_norm: 0.5743589743589743
        • acc_norm_stderr: 0.03549871080367708
      • community|acva:Ancient_Egypt|0:
        • acc_norm: 0.3333333333333333
        • acc_norm_stderr: 0.026602896148920783
      • community|acva:Arab_Empire|0:
        • acc_norm: 0.33584905660377357
        • acc_norm_stderr: 0.029067220146644823
      • community|acva:Arabic_Architecture|0:
        • acc_norm: 0.5487179487179488
        • acc_norm_stderr: 0.03572709860318392
      • community|acva:Arabic_Art|0:
        • acc_norm: 0.441025641025641
        • acc_norm_stderr: 0.03564732931853579
      • community|acva:Arabic_Astronomy|0:
        • acc_norm: 0.4666666666666667
        • acc_norm_stderr: 0.03581804596782233
      • community|acva:Arabic_Calligraphy|0:
        • acc_norm: 0.592156862745098
        • acc_norm_stderr: 0.030835294779357078
      • community|acva:Arabic_Ceremony|0:
        • acc_norm: 0.6378378378378379
        • acc_norm_stderr: 0.03543217115138485
      • community|acva:Arabic_Clothing|0:
        • acc_norm: 0.558974358974359
        • acc_norm_stderr: 0.035647329318535786
      • community|acva:Arabic_Culture|0:
        • acc_norm: 0.5282051282051282
        • acc_norm_stderr: 0.03584074674920833
      • community|acva:Arabic_Food|0:
        • acc_norm: 0.5743589743589743
        • acc_norm_stderr: 0.03549871080367707
      • community|acva:Arabic_Funeral|0:
        • acc_norm: 0.5684210526315789
        • acc_norm_stderr: 0.05108592673308947
      • community|acva:Arabic_Geography|0:
        • acc_norm: 0.6275862068965518
        • acc_norm_stderr: 0.04028731532947558
      • community|acva:Arabic_History|0:
        • acc_norm: 0.4461538461538462
        • acc_norm_stderr: 0.035689135465692336
      • community|acva:Arabic_Language_Origin|0:
        • acc_norm: 0.631578947368421
        • acc_norm_stderr: 0.049753325624911644
      • community|acva:Arabic_Literature|0:
        • acc_norm: 0.6137931034482759
        • acc_norm_stderr: 0.04057324734419035
      • community|acva:Arabic_Math|0:
        • acc_norm: 0.30256410256410254
        • acc_norm_stderr: 0.03298070870085618
      • community|acva:Arabic_Medicine|0:
        • acc_norm: 0.6068965517241379
        • acc_norm_stderr: 0.040703290137070705
      • community|acva:Arabic_Music|0:
        • acc_norm: 0.60431654676259
        • acc_norm_stderr: 0.04162618828625745
      • community|acva:Arabic_Ornament|0:
        • acc_norm: 0.6871794871794872
        • acc_norm_stderr: 0.033287550657248546
      • community|acva:Arabic_Philosophy|0:
        • acc_norm: 0.6689655172413793
        • acc_norm_stderr: 0.039215453124671215
      • community|acva:Arabic_Physics_and_Chemistry|0:
        • acc_norm: 0.5384615384615384
        • acc_norm_stderr: 0.03579154352544572
      • community|acva:Arabic_Wedding|0:
        • acc_norm: 0.48205128205128206
        • acc_norm_stderr: 0.0358747709877383
      • community|acva:Bahrain|0:
        • acc_norm: 0.37777777777777777
        • acc_norm_stderr: 0.07309112127323451
      • community|acva:Comoros|0:
        • acc_norm: 0.4
        • acc_norm_stderr: 0.07385489458759965
      • community|acva:Egypt_modern|0:
        • acc_norm: 0.43157894736842106
        • acc_norm_stderr: 0.05108592673308946
      • community|acva:InfluenceFromAncientEgypt|0:
        • acc_norm: 0.6974358974358974
        • acc_norm_stderr: 0.03298070870085619
      • community|acva:InfluenceFromByzantium|0:
        • acc_norm: 0.8068965517241379
        • acc_norm_stderr: 0.03289445522127401
      • community|acva:InfluenceFromChina|0:
        • acc_norm: 0.26666666666666666
        • acc_norm_stderr: 0.0317493043641267
      • community|acva:InfluenceFromGreece|0:
        • acc_norm: 0.7282051282051282
        • acc_norm_stderr: 0.03194086187025723
      • community|acva:InfluenceFromIslam|0:
        • acc_norm: 0.6344827586206897
        • acc_norm_stderr: 0.04013124195424385
      • community|acva:InfluenceFromPersia|0:
        • acc_norm: 0.7657142857142857
        • acc_norm_stderr: 0.0321093603969262
      • community|acva:InfluenceFromRome|0:
        • acc_norm: 0.6358974358974359
        • acc_norm_stderr: 0.03454653867786389
      • community|acva:Iraq|0:
        • acc_norm: 0.5176470588235295
        • acc_norm_stderr: 0.05452048340661895
      • community|acva:Islam_Education|0:
        • acc_norm: 0.6153846153846154
        • acc_norm_stderr: 0.03492896993742303
      • community|acva:Islam_branches_and_schools|0:
        • acc_norm: 0.45714285714285713
        • acc_norm_stderr: 0.037765402985786664
      • community|acva:Islamic_law_system|0:
        • acc_norm: 0.6564102564102564
        • acc_norm_stderr: 0.034096273014098545
      • community|acva:Jordan|0:
        • acc_norm: 0.37777777777777777
        • acc_norm_stderr: 0.07309112127323451
      • community|acva:Kuwait|0:
        • acc_norm: 0.37777777777777777
        • acc_norm_stderr: 0.07309112127323451
      • community|acva:Lebanon|0:
        • acc_norm: 0.4222222222222222
        • acc_norm_stderr: 0.07446027270295806
      • community|acva:Libya|0:
        • acc_norm: 0.5111111111111111
        • acc_norm_stderr: 0.07535922203472523
      • community|acva:Mauritania|0:
        • acc_norm: 0.4888888888888889
        • acc_norm_stderr: 0.07535922203472523
      • community|acva:Mesopotamia_civilization|0:
        • acc_norm: 0.5741935483870968
        • acc_norm_stderr: 0.03984509920961719
      • community|acva:Morocco|0:
        • acc_norm: 0.4444444444444444
        • acc_norm_stderr: 0.07491109582924914
      • community|acva:Oman|0:
        • acc_norm: 0.35555555555555557
        • acc_norm_stderr: 0.07216392363431012
      • community|acva:Palestine|0:
        • acc_norm: 0.38823529411764707
        • acc_norm_stderr: 0.05317409082203827
      • community|acva:Qatar|0:
        • acc_norm: 0.5111111111111111
        • acc_norm_stderr: 0.07535922203472523
      • community|acva:Saudi_Arabia|0:
        • acc_norm: 0.5435897435897435
        • acc_norm_stderr: 0.035761230969912135
      • community|acva:Somalia|0:
        • acc_norm: 0.4888888888888889
        • acc_norm_stderr: 0.07535922203472523
      • community|acva:Sudan|0:
        • acc_norm: 0.5333333333333333
        • acc_norm_stderr: 0.0752101433090355
      • community|acva:Syria|0:
        • acc_norm: 0.5333333333333333
        • `acc_norm_stderr
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