open-llm-leaderboard-old/details_4season__alignment-model-test9
收藏数据集概述
该数据集是在评估模型 4season/alignment-model-test9 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 timestamp。
- "train" 分割始终指向最新的结果。
- 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示聚合指标在 Open LLM Leaderboard 上。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_4season__alignment-model-test9", "harness_winogrande_5", split="train")
最新结果
以下是 2024-04-08T15:04:12.877958 运行的最新结果:
python { "all": { "acc": 0.6834330163002643, "acc_stderr": 0.031540209576181304, "acc_norm": 0.6858087556068694, "acc_norm_stderr": 0.03219476958988141, "mc1": 0.609547123623011, "mc1_stderr": 0.017078230743431455, "mc2": 0.7524744031917101, "mc2_stderr": 0.014224281719560656 }, "harness|arc:challenge|25": { "acc": 0.764505119453925, "acc_stderr": 0.012399451855004741, "acc_norm": 0.7755972696245734, "acc_norm_stderr": 0.012191404938603831 }, "harness|hellaswag|10": { "acc": 0.7879904401513643, "acc_stderr": 0.004078962503408514, "acc_norm": 0.9068910575582553, "acc_norm_stderr": 0.0028999116931072897 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.4, "acc_stderr": 0.04923659639173309, "acc_norm": 0.4, "acc_norm_stderr": 0.04923659639173309 }, ... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_04_08T15_04_12.877958
- path: **/details_harness|arc:challenge|25_2024-04-08T15-04-12.877958.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-04-08T15-04-12.877958.parquet
- split: 2024_04_08T15_04_12.877958
-
config_name: harness_gsm8k_5
- split: 2024_04_08T15_04_12.877958
- path: **/details_harness|gsm8k|5_2024-04-08T15-04-12.877958.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-04-08T15-04-12.877958.parquet
- split: 2024_04_08T15_04_12.877958
-
config_name: harness_hellaswag_10
- split: 2024_04_08T15_04_12.877958
- path: **/details_harness|hellaswag|10_2024-04-08T15-04-12.877958.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-04-08T15-04-12.877958.parquet
- split: 2024_04_08T15_04_12.877958
-
config_name: harness_hendrycksTest_5
- split: 2024_04_08T15_04_12.877958
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-04-08T15-04-12.877958.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-04-08T15-04-12.877958.parquet ... }
- path:
- split: 2024_04_08T15_04_12.877958



