open-llm-leaderboard-old/details_ewqr2130__alignment-handbook-zephyr-7b_ppostep_100
收藏数据集概述
数据集简介
该数据集是在对模型 ewqr2130/alignment-handbook-zephyr-7b_ppostep_100 进行评估运行时自动创建的,用于 Open LLM Leaderboard 上的评估。
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。
额外配置
一个额外的配置 "results" 存储所有运行的聚合结果,用于计算并在 Open LLM Leaderboard 上显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ewqr2130__alignment-handbook-zephyr-7b_ppostep_100", "harness_winogrande_5", split="train")
最新结果
以下是 2024-01-18T13:32:11.965537 运行的最新结果:
python { "all": { "acc": 0.23774405306004376, "acc_stderr": 0.030137212407426187, "acc_norm": 0.23884860895654994, "acc_norm_stderr": 0.030942263352296245, "mc1": 0.23133414932680538, "mc1_stderr": 0.014761945174862661, "mc2": NaN, "mc2_stderr": NaN }, "harness|arc:challenge|25": { "acc": 0.2167235494880546, "acc_stderr": 0.012040156713481189, "acc_norm": 0.29266211604095566, "acc_norm_stderr": 0.013295916103619397 }, "harness|hellaswag|10": { "acc": 0.25562636924915355, "acc_stderr": 0.004353212146198442, "acc_norm": 0.25871340370444135, "acc_norm_stderr": 0.004370328224831795 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.24, "acc_stderr": 0.042923469599092816, "acc_norm": 0.24, "acc_norm_stderr": 0.042923469599092816 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_01_18T13_32_11.965537
- path:
**/details_harness|arc:challenge|25_2024-01-18T13-32-11.965537.parquet
- path:
- split: latest
- path:
**/details_harness|arc:challenge|25_2024-01-18T13-32-11.965537.parquet
- path:
- split: 2024_01_18T13_32_11.965537
-
config_name: harness_gsm8k_5
- split: 2024_01_18T13_32_11.965537
- path:
**/details_harness|gsm8k|5_2024-01-18T13-32-11.965537.parquet
- path:
- split: latest
- path:
**/details_harness|gsm8k|5_2024-01-18T13-32-11.965537.parquet
- path:
- split: 2024_01_18T13_32_11.965537
-
config_name: harness_hellaswag_10
- split: 2024_01_18T13_32_11.965537
- path:
**/details_harness|hellaswag|10_2024-01-18T13-32-11.965537.parquet
- path:
- split: latest
- path:
**/details_harness|hellaswag|10_2024-01-18T13-32-11.965537.parquet
- path:
- split: 2024_01_18T13_32_11.965537
-
config_name: harness_hendrycksTest_5
- split: 2024_01_18T13_32_11.965537
- path:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-management|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-18T13-32-11.965537.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-18T13-32-11.965537.parquet- `**/details_harness|hendrycksTest-professional_psych
- path:
- split: 2024_01_18T13_32_11.965537



