open-llm-leaderboard-old/details_Lazycuber__Janemalion-6B
收藏数据集概述
该数据集是在对模型 Lazycuber/Janemalion-6B 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 61 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,每个运行可以在每个配置中找到一个特定的分割,分割名称使用运行的时间戳。
- "train" 分割始终指向最新的结果。
- 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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Lazycuber__Janemalion-6B", "harness_truthfulqa_mc_0", split="train")
最新结果
以下是 2023-07-24T11:00:29.262151 运行的最新结果:
python { "all": { "acc": 0.28835373902195927, "acc_stderr": 0.03262443339491777, "acc_norm": 0.29201970368726204, "acc_norm_stderr": 0.032622091393243714, "mc1": 0.20930232558139536, "mc1_stderr": 0.01424121943478583, "mc2": 0.34587723098212036, "mc2_stderr": 0.01348699014658101 }, "harness|arc:challenge|25": { "acc": 0.386518771331058, "acc_stderr": 0.01423008476191048, "acc_norm": 0.42406143344709896, "acc_norm_stderr": 0.014441889627464398 }, "harness|hellaswag|10": { "acc": 0.5052778331009758, "acc_stderr": 0.004989503417767287, "acc_norm": 0.6840270862378013, "acc_norm_stderr": 0.00463952045344403 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.26, "acc_stderr": 0.0440844002276808, "acc_norm": 0.26, "acc_norm_stderr": 0.0440844002276808 }, ... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_07_24T11_00_29.262151
- path: **/details_harness|arc:challenge|25_2023-07-24T11:00:29.262151.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-07-24T11:00:29.262151.parquet
- split: 2023_07_24T11_00_29.262151
-
config_name: harness_hellaswag_10
- split: 2023_07_24T11_00_29.262151
- path: **/details_harness|hellaswag|10_2023-07-24T11:00:29.262151.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-07-24T11:00:29.262151.parquet
- split: 2023_07_24T11_00_29.262151
-
config_name: harness_hendrycksTest_5
- split: 2023_07_24T11_00_29.262151
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-24T11:00:29.262151.parquet ...
- path:
- split: latest
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-24T11:00:29.262151.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-24T11:00:29.262151.parquet ...
- path:
- split: 2023_07_24T11_00_29.262151



