open-llm-leaderboard-old/details_mistralai__Mistral-7B-v0.3
收藏数据集概述
数据集简介
该数据集是在对模型 mistralai/Mistral-7B-v0.3 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集结构
- 数据集包含 63 个配置,每个配置对应一个评估任务。
- 数据集来自 1 次运行,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 "results" 配置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_mistralai__Mistral-7B-v0.3", "harness_winogrande_5", split="train")
最新结果
这些是最新的结果,来自运行 2024-05-23T11:37:26.409605: python { "all": { "acc": 0.6311207379058325, "acc_stderr": 0.032475183287971776, "acc_norm": 0.6374358055436643, "acc_norm_stderr": 0.03313680029875512, "mc1": 0.2766217870257038, "mc1_stderr": 0.01565960575532692, "mc2": 0.4179464876493265, "mc2_stderr": 0.014214014618700525 }, "harness|arc:challenge|25": { "acc": 0.568259385665529, "acc_stderr": 0.014474591427196202, "acc_norm": 0.6049488054607508, "acc_norm_stderr": 0.01428589829293817 }, "harness|hellaswag|10": { "acc": 0.6267675761800439, "acc_stderr": 0.004826746160830182, "acc_norm": 0.8299143596893049, "acc_norm_stderr": 0.003749401775087307 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.3, "acc_stderr": 0.046056618647183814, "acc_norm": 0.3, "acc_norm_stderr": 0.046056618647183814 }, # 其他任务的结果省略... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_05_23T11_37_26.409605
- path: **/details_harness|arc:challenge|25_2024-05-23T11-37-26.409605.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-05-23T11-37-26.409605.parquet
- split: 2024_05_23T11_37_26.409605
-
config_name: harness_gsm8k_5
- split: 2024_05_23T11_37_26.409605
- path: **/details_harness|gsm8k|5_2024-05-23T11-37-26.409605.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-05-23T11-37-26.409605.parquet
- split: 2024_05_23T11_37_26.409605
-
config_name: harness_hellaswag_10
- split: 2024_05_23T11_37_26.409605
- path: **/details_harness|hellaswag|10_2024-05-23T11-37-26.409605.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-05-23T11-37-26.409605.parquet
- split: 2024_05_23T11_37_26.409605
-
config_name: harness_hendrycksTest_5
- split: 2024_05_23T11_37_26.409605
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-management|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-05-23T11-37-26.409605.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_
- path:
- split: 2024_05_23T11_37_26.409605



