open-llm-leaderboard-old/details_kodonho__SolarM-SakuraSolar-SLERP
收藏数据集概述
该数据集是在评估模型 kodonho/SolarM-SakuraSolar-SLERP 在 Open LLM Leaderboard 上的自动创建的。数据集包含 63 个配置,每个配置对应一个评估任务。
数据集结构
数据集由 1 次运行创建,每个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" 分割总是指向最新的结果。
结果配置
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。
加载数据示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_kodonho__SolarM-SakuraSolar-SLERP", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要:
python { "all": { "acc": 0.6661841092869378, "acc_stderr": 0.031618934361982654, "acc_norm": 0.6670035390095326, "acc_norm_stderr": 0.03226162867427471, "mc1": 0.5667074663402693, "mc1_stderr": 0.01734702445010748, "mc2": 0.7209867000245426, "mc2_stderr": 0.014980299607085815 }, "harness|arc:challenge|25": { "acc": 0.6877133105802048, "acc_stderr": 0.013542598541688067, "acc_norm": 0.71160409556314, "acc_norm_stderr": 0.013238394422428173 }, "harness|hellaswag|10": { "acc": 0.714797849034057, "acc_stderr": 0.0045058790846068415, "acc_norm": 0.8846843258315077, "acc_norm_stderr": 0.003187497509087418 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.43, "acc_stderr": 0.049756985195624284, "acc_norm": 0.43, "acc_norm_stderr": 0.049756985195624284 }, # 其他任务的结果... }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_01_13T17_26_45.129484
- 路径: **/details_harness|arc:challenge|25_2024-01-13T17-26-45.129484.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-01-13T17-26-45.129484.parquet
- 分割: 2024_01_13T17_26_45.129484
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_01_13T17_26_45.129484
- 路径: **/details_harness|gsm8k|5_2024-01-13T17-26-45.129484.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-01-13T17-26-45.129484.parquet
- 分割: 2024_01_13T17_26_45.129484
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_01_13T17_26_45.129484
- 路径: **/details_harness|hellaswag|10_2024-01-13T17-26-45.129484.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-01-13T17-26-45.129484.parquet
- 分割: 2024_01_13T17_26_45.129484
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_01_13T17_26_45.129484
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-13T17-26-45.129484.parquet
其他路径...
- 路径:
- 分割: latest
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-13T17-26-45.129484.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-13T17-26-45.129484.parquet
其他路径...
- 路径:
- 分割: 2024_01_13T17_26_45.129484
- 数据文件:



