open-llm-leaderboard-old/details_BFauber__opt125m_10e5_40ep
收藏数据集概述
数据集简介
该数据集是在对模型 BFauber/opt125m_10e5_40ep 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
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,每个运行可以在每个配置中作为一个特定的分割找到,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 "results" 配置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_BFauber__opt125m_10e5_40ep", "harness_winogrande_5", split="train")
最新结果
以下是 2024-02-02T19:44:13.671320 运行的最新结果:
python { "all": { "acc": 0.2385010715928439, "acc_stderr": 0.030097358112278164, "acc_norm": 0.238644520881259, "acc_norm_stderr": 0.030887009025888052, "mc1": 0.26193390452876375, "mc1_stderr": 0.015392118805015023, "mc2": 0.49021738591692526, "mc2_stderr": 0.015824528690198854 }, "harness|arc:challenge|25": { "acc": 0.22440273037542663, "acc_stderr": 0.01219140493860384, "acc_norm": 0.24232081911262798, "acc_norm_stderr": 0.012521593295800118 }, "harness|hellaswag|10": { "acc": 0.2748456482772356, "acc_stderr": 0.004455240755811554, "acc_norm": 0.29904401513642703, "acc_norm_stderr": 0.004569034613332606 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.046882617226215034, "acc_norm": 0.32, "acc_norm_stderr": 0.046882617226215034 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_02_02T19_44_13.671320
- path: **/details_harness|arc:challenge|25_2024-02-02T19-44-13.671320.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-02-02T19-44-13.671320.parquet
- split: 2024_02_02T19_44_13.671320
-
config_name: harness_gsm8k_5
- split: 2024_02_02T19_44_13.671320
- path: **/details_harness|gsm8k|5_2024-02-02T19-44-13.671320.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-02-02T19-44-13.671320.parquet
- split: 2024_02_02T19_44_13.671320
-
config_name: harness_hellaswag_10
- split: 2024_02_02T19_44_13.671320
- path: **/details_harness|hellaswag|10_2024-02-02T19-44-13.671320.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-02-02T19-44-13.671320.parquet
- split: 2024_02_02T19_44_13.671320
-
config_name: harness_hendrycksTest_5
- split: 2024_02_02T19_44_13.671320
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-02T19-44-13.671320.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-0
- path:
- split: 2024_02_02T19_44_13.671320



