open-llm-leaderboard-old/details_PotatoOff__Michel-13B
收藏数据集概述
数据集摘要
该数据集是在评估模型 PotatoOff/Michel-13B 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集组成
- 数据集包含 63 个配置,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" 分割始终指向最新的结果。
- 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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_PotatoOff__Michel-13B", "harness_winogrande_5", split="train")
最新结果
以下是 2024-02-01T22:05:33.263550 运行的最新结果:
python { "all": { "acc": 0.5499141517864713, "acc_stderr": 0.034037566570250866, "acc_norm": 0.556324343200096, "acc_norm_stderr": 0.03477678629039932, "mc1": 0.35006119951040393, "mc1_stderr": 0.01669794942015103, "mc2": 0.5043477199409111, "mc2_stderr": 0.015764099492460493 }, "harness|arc:challenge|25": { "acc": 0.5767918088737202, "acc_stderr": 0.014438036220848034, "acc_norm": 0.6126279863481229, "acc_norm_stderr": 0.01423587248790987 }, "harness|hellaswag|10": { "acc": 0.6357299342760406, "acc_stderr": 0.004802413919932666, "acc_norm": 0.832105158334993, "acc_norm_stderr": 0.0037300899105375796 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.0479372485441102, "acc_norm": 0.35, "acc_norm_stderr": 0.0479372485441102 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_02_01T22_05_33.263550
- path: **/details_harness|arc:challenge|25_2024-02-01T22-05-33.263550.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-02-01T22-05-33.263550.parquet
- split: 2024_02_01T22_05_33.263550
-
config_name: harness_gsm8k_5
- split: 2024_02_01T22_05_33.263550
- path: **/details_harness|gsm8k|5_2024-02-01T22-05-33.263550.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-02-01T22-05-33.263550.parquet
- split: 2024_02_01T22_05_33.263550
-
config_name: harness_hellaswag_10
- split: 2024_02_01T22_05_33.263550
- path: **/details_harness|hellaswag|10_2024-02-01T22-05-33.263550.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-02-01T22-05-33.263550.parquet
- split: 2024_02_01T22_05_33.263550
-
config_name: harness_hendrycksTest_5
- split: 2024_02_01T22_05_33.263550
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-01T22-05-33.263550.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-01T22-05-33.263550.parquet
其他任务的路径...
- path:
- split: 2024_02_01T22_05_33.263550
}



