open-llm-leaderboard-old/details_Gille__StrangeMerges_10-7B-slerp
收藏数据集概述
数据集简介
该数据集是在评估模型 Gille/StrangeMerges_10-7B-slerp 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 63 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" 分割总是指向最新的结果。
- 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_Gille__StrangeMerges_10-7B-slerp", "harness_winogrande_5", split="train")
最新结果
这些是最新结果的示例,来自 2024-02-02T02:55:04.492502 的运行: python { "all": { "acc": 0.6542458004549463, "acc_stderr": 0.03204861565652575, "acc_norm": 0.6539758320346176, "acc_norm_stderr": 0.03271443876560244, "mc1": 0.543451652386781, "mc1_stderr": 0.017437280953183688, "mc2": 0.6948877994288644, "mc2_stderr": 0.014809641585651314 }, "harness|arc:challenge|25": { "acc": 0.6919795221843004, "acc_stderr": 0.013491429517292038, "acc_norm": 0.7235494880546075, "acc_norm_stderr": 0.013069662474252423 }, "harness|hellaswag|10": { "acc": 0.7026488747261501, "acc_stderr": 0.004561582009834578, "acc_norm": 0.8829914359689305, "acc_norm_stderr": 0.0032077357692780455 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.04760952285695235, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695235 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_02_02T02_55_04.492502
- path: **/details_harness|arc:challenge|25_2024-02-02T02-55-04.492502.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-02-02T02-55-04.492502.parquet
- split: 2024_02_02T02_55_04.492502
-
config_name: harness_gsm8k_5
- split: 2024_02_02T02_55_04.492502
- path: **/details_harness|gsm8k|5_2024-02-02T02-55-04.492502.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-02-02T02-55-04.492502.parquet
- split: 2024_02_02T02_55_04.492502
-
config_name: harness_hellaswag_10
- split: 2024_02_02T02_55_04.492502
- path: **/details_harness|hellaswag|10_2024-02-02T02-55-04.492502.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-02-02T02-55-04.492502.parquet
- split: 2024_02_02T02_55_04.492502
-
config_name: harness_hendrycksTest_5
- split: 2024_02_02T02_55_04.492502
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2024-02-02T02-55-04.492502.parquet
- **/details_harness
- path:
- split: 2024_02_02T02_55_04.492502



