open-llm-leaderboard-old/details_alykassem__ds_diasum_md_mixtral
收藏数据集概述
该数据集是在对模型 alykassem/ds_diasum_md_mixtral 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
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。
- 数据集从 1 次运行中创建,每次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" 存储所有运行结果的聚合,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_alykassem__ds_diasum_md_mixtral", "harness_winogrande_5", split="train")
最新结果
以下是 2023-12-29T19:29:56.671932 运行的最新结果:
python { "all": { "acc": 0.692270226414827, "acc_stderr": 0.030762586489589964, "acc_norm": 0.6972458162944918, "acc_norm_stderr": 0.031355995272453654, "mc1": 0.4039167686658507, "mc1_stderr": 0.01717727682258428, "mc2": 0.5572164097692784, "mc2_stderr": 0.01463024293704983 }, "harness|arc:challenge|25": { "acc": 0.6203071672354948, "acc_stderr": 0.014182119866974872, "acc_norm": 0.6629692832764505, "acc_norm_stderr": 0.013813476652902272 }, "harness|hellaswag|10": { "acc": 0.652459669388568, "acc_stderr": 0.004752158936871874, "acc_norm": 0.854511053574985, "acc_norm_stderr": 0.003518725257365599 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.39, "acc_stderr": 0.04902071300001975, "acc_norm": 0.39, "acc_norm_stderr": 0.04902071300001975 }, ... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_12_29T19_29_56.671932
- path: **/details_harness|arc:challenge|25_2023-12-29T19-29-56.671932.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-12-29T19-29-56.671932.parquet
- split: 2023_12_29T19_29_56.671932
-
config_name: harness_gsm8k_5
- split: 2023_12_29T19_29_56.671932
- path: **/details_harness|gsm8k|5_2023-12-29T19-29-56.671932.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-12-29T19-29-56.671932.parquet
- split: 2023_12_29T19_29_56.671932
-
config_name: harness_hellaswag_10
- split: 2023_12_29T19_29_56.671932
- path: **/details_harness|hellaswag|10_2023-12-29T19-29-56.671932.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-12-29T19-29-56.671932.parquet
- split: 2023_12_29T19_29_56.671932
-
config_name: harness_hendrycksTest_5
- split: 2023_12_29T19_29_56.671932
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-29T19-29-56.671932.parquet
- ...
- path:
- split: latest
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-29T19-29-56.671932.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-29T19-29-56.671932.parquet
- ...
- path:
- split: 2023_12_29T19_29_56.671932



