open-llm-leaderboard-old/details_wenge-research__yayi-13b-llama2
收藏数据集概述
该数据集是在评估模型 wenge-research/yayi-13b-llama2 在 Open LLM Leaderboard 上的运行过程中自动创建的。
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。
- 数据集由 3 次运行创建,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_wenge-research__yayi-13b-llama2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-15T08:54:37.748891 运行的最新结果:
python { "all": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642966022, "f1": 0.05916107382550354, "f1_stderr": 0.0014083828571043837, "acc": 0.3685519093475062, "acc_stderr": 0.009163083599802495 }, "harness|drop|3": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642966022, "f1": 0.05916107382550354, "f1_stderr": 0.0014083828571043837 }, "harness|gsm8k|5": { "acc": 0.0401819560272934, "acc_stderr": 0.00540943973697052 }, "harness|winogrande|5": { "acc": 0.696921862667719, "acc_stderr": 0.012916727462634468 } }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_07_27T11_25_49.235330
- path: **/details_harness|arc:challenge|25_2023-07-27T11:25:49.235330.parquet
- split: 2023_10_01T13_33_23.488564
- path: **/details_harness|arc:challenge|25_2023-10-01T13-33-23.488564.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-10-01T13-33-23.488564.parquet
- split: 2023_07_27T11_25_49.235330
-
config_name: harness_drop_3
- split: 2023_10_15T08_54_37.748891
- path: **/details_harness|drop|3_2023-10-15T08-54-37.748891.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-15T08-54-37.748891.parquet
- split: 2023_10_15T08_54_37.748891
-
config_name: harness_gsm8k_5
- split: 2023_10_15T08_54_37.748891
- path: **/details_harness|gsm8k|5_2023-10-15T08-54-37.748891.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-15T08-54-37.748891.parquet
- split: 2023_10_15T08_54_37.748891
-
config_name: harness_hellaswag_10
- split: 2023_07_27T11_25_49.235330
- path: **/details_harness|hellaswag|10_2023-07-27T11:25:49.235330.parquet
- split: 2023_10_01T13_33_23.488564
- path: **/details_harness|hellaswag|10_2023-10-01T13-33-23.488564.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-10-01T13-33-23.488564.parquet
- split: 2023_07_27T11_25_49.235330
-
config_name: harness_hendrycksTest_5
- split: 2023_07_27T11_25_49.235330
- path: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-anatomy|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-astronomy|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-business_ethics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_biology|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_medicine|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-college_physics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-computer_security|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-econometrics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-formal_logic|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-global_facts|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_biology|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_geography|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_physics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-human_aging|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-human_sexuality|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-international_law|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-jurisprudence|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-machine_learning|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-management|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-marketing|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-medical_genetics|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-miscellaneous|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-moral_disputes|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-27T11:25:49.235330.parquet
- path: **/details_harness|
- split: 2023_07_27T11_25_49.235330



