open-llm-leaderboard/details_LLMs__WizardLM-30B-V1.0
收藏数据集概述
数据集名称
Evaluation run of LLMs/WizardLM-30B-V1.0
数据集摘要
该数据集是在模型LLMs/WizardLM-30B-V1.0在Open LLM Leaderboard上的评估运行期间自动创建的。
数据集组成
- 数据集包含64个配置,每个配置对应一个评估任务。
- 数据集从3次运行中创建,每次运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
- "train"分片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,用于计算和显示在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_LLMs__WizardLM-30B-V1.0", "harness_winogrande_5", split="train")
最新结果
以下是来自2023-10-30T02:37:12.561310运行的最新结果: python { "all": { "em": 0.286493288590604, "em_stderr": 0.0046301590416793666, "f1": 0.3621350671140946, "f1_stderr": 0.00452241220066869, "acc": 0.4967032138503913, "acc_stderr": 0.011557270415432395 }, "harness|drop|3": { "em": 0.286493288590604, "em_stderr": 0.0046301590416793666, "f1": 0.3621350671140946, "f1_stderr": 0.00452241220066869 }, "harness|gsm8k|5": { "acc": 0.21834723275208492, "acc_stderr": 0.011379497266738049 }, "harness|winogrande|5": { "acc": 0.7750591949486977, "acc_stderr": 0.011735043564126742 } }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_08_22T13_36_33.189763
- path: **/details_harness|arc:challenge|25_2023-08-22T13:36:33.189763.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-08-22T13:36:33.189763.parquet
- split: 2023_08_22T13_36_33.189763
-
config_name: harness_drop_3
- split: 2023_10_17T07_27_45.879930
- path: **/details_harness|drop|3_2023-10-17T07-27-45.879930.parquet
- split: 2023_10_30T02_37_12.561310
- path: **/details_harness|drop|3_2023-10-30T02-37-12.561310.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-30T02-37-12.561310.parquet
- split: 2023_10_17T07_27_45.879930
-
config_name: harness_gsm8k_5
- split: 2023_10_17T07_27_45.879930
- path: **/details_harness|gsm8k|5_2023-10-17T07-27-45.879930.parquet
- split: 2023_10_30T02_37_12.561310
- path: **/details_harness|gsm8k|5_2023-10-30T02-37-12.561310.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-30T02-37-12.561310.parquet
- split: 2023_10_17T07_27_45.879930
-
config_name: harness_hellaswag_10
- split: 2023_08_22T13_36_33.189763
- path: **/details_harness|hellaswag|10_2023-08-22T13:36:33.189763.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-08-22T13:36:33.189763.parquet
- split: 2023_08_22T13_36_33.189763
-
config_name: harness_hendrycksTest_5
- split: 2023_08_22T13_36_33.189763
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-management|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-08-22T13:36:33.189763.parquet
- **/details_harness|hendrycksTest-
- path:
- split: 2023_08_22T13_36_33.189763



