open-llm-leaderboard-old/details_CHIH-HUNG__llama-2-13b-FINETUNE5_4w-r4-gate_up_down
收藏数据集概述
数据集来源
该数据集是在对模型 CHIH-HUNG/llama-2-13b-FINETUNE5_4w-r4-gate_up_down 进行评估运行期间自动创建的,评估运行在 Open LLM Leaderboard 上进行。
数据集结构
数据集由64个配置组成,每个配置对应一个评估任务。数据集从2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。
数据集加载
要加载特定运行的详细信息,可以使用以下代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-FINETUNE5_4w-r4-gate_up_down", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-26T15:53:08.381645 运行 的最新结果: python { "all": { "em": 0.2154991610738255, "em_stderr": 0.004210747014430766, "f1": 0.25919148489932897, "f1_stderr": 0.004195696877017449, "acc": 0.4490387889225113, "acc_stderr": 0.01073317504472215 }, "harness|drop|3": { "em": 0.2154991610738255, "em_stderr": 0.004210747014430766, "f1": 0.25919148489932897, "f1_stderr": 0.004195696877017449 }, "harness|gsm8k|5": { "acc": 0.1372251705837756, "acc_stderr": 0.009477808244600398 }, "harness|winogrande|5": { "acc": 0.760852407261247, "acc_stderr": 0.011988541844843905 } }
配置详情
以下是数据集的配置详情:
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2023_10_10T11_38_23.134636
- path: **/details_harness|arc:challenge|25_2023-10-10T11-38-23.134636.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-10-10T11-38-23.134636.parquet
- split: 2023_10_10T11_38_23.134636
- data_files:
-
config_name: harness_drop_3
- data_files:
- split: 2023_10_26T15_53_08.381645
- path: **/details_harness|drop|3_2023-10-26T15-53-08.381645.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-26T15-53-08.381645.parquet
- split: 2023_10_26T15_53_08.381645
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2023_10_26T15_53_08.381645
- path: **/details_harness|gsm8k|5_2023-10-26T15-53-08.381645.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-26T15-53-08.381645.parquet
- split: 2023_10_26T15_53_08.381645
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2023_10_10T11_38_23.134636
- path: **/details_harness|hellaswag|10_2023-10-10T11-38-23.134636.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-10-10T11-38-23.134636.parquet
- split: 2023_10_10T11_38_23.134636
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2023_10_10T11_38_23.134636
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-management|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-10-10T11-38-23.134636.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-10-10T11-38-23.
- path:
- split: 2023_10_10T11_38_23.134636
- data_files:



