open-llm-leaderboard-old/details_eren23__NeuralDareBeagle-7B-slerp
收藏数据集概述
该数据集是在评估模型 eren23/NeuralDareBeagle-7B-slerp 的过程中自动创建的,用于 Open LLM Leaderboard。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行创建,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。
数据集结构
配置详情
-
配置名称:
harness_arc_challenge_25- 分割:
2024_01_28T18_11_46.511504,latest - 路径:
**/details_harness|arc:challenge|25_2024-01-28T18-11-46.511504.parquet
- 分割:
-
配置名称:
harness_gsm8k_5- 分割:
2024_01_28T18_11_46.511504,latest - 路径:
**/details_harness|gsm8k|5_2024-01-28T18-11-46.511504.parquet
- 分割:
-
配置名称:
harness_hellaswag_10- 分割:
2024_01_28T18_11_46.511504,latest - 路径:
**/details_harness|hellaswag|10_2024-01-28T18-11-46.511504.parquet
- 分割:
-
配置名称:
harness_hendrycksTest_5- 分割:
2024_01_28T18_11_46.511504,latest - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-management|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-public_relations|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-security_studies|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-sociology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-virology|5_2024-01-28T18-11-46.511504.parquet**/details_harness|hendrycksTest-world_religions|5_2024-01-28T18-11-46.511504.parquet
- 分割:
最新结果
以下是 2024-01-28T18:11:46.511504 运行 的最新结果:
python { "all": { "acc": 0.6554392261924249, "acc_stderr": 0.03212679462957801, "acc_norm": 0.6550602589470452, "acc_norm_stderr": 0.032794301399577036, "mc1": 0.5581395348837209, "mc1_stderr": 0.01738476747898621, "mc2": 0.6918000534624221, "mc2_stderr": 0.014976389591941985 }, "harness|arc:challenge|25": { "acc": 0.6953924914675768, "acc_stderr": 0.01344952210993249, "acc_norm": 0.7209897610921502, "acc_norm_stderr": 0.013106784883601333 }, "harness|hellaswag|10": { "acc": 0.7094204341764588, "acc_stderr": 0.004531019159414108, "acc_norm": 0.8819956184027086, "acc_norm_stderr": 0.0032195397905004732 }, "



