open-llm-leaderboard-old/details_kimwooglae__AISquare-Instruct-SOLAR-10.7b-v0.5.31
收藏数据集概述
数据集摘要
该数据集是在评估模型 kimwooglae/AISquare-Instruct-SOLAR-10.7b-v0.5.31 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。
数据集结构
- 配置数量:63
- 数据来源:1 次运行(run)
- 数据分割:每个配置包含特定分割,分割名称使用运行的时间戳。"train" 分割始终指向最新结果。
- 额外配置:"results" 存储所有运行结果的聚合,用于计算和显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_kimwooglae__AISquare-Instruct-SOLAR-10.7b-v0.5.31", "harness_winogrande_5", split="train")
最新结果
最新结果来自 2024-01-22T02:13:58.257879 的运行,包含多个任务的评估结果。以下是部分任务的结果示例:
python { "all": { "acc": 0.5330985732271527, "acc_stderr": 0.034185007803077, "acc_norm": 0.5352323665963996, "acc_norm_stderr": 0.034920748737001794, "mc1": 0.35006119951040393, "mc1_stderr": 0.01669794942015103, "mc2": 0.5134609475665187, "mc2_stderr": 0.014908191115467387 }, "harness|arc:challenge|25": { "acc": 0.5725255972696246, "acc_stderr": 0.014456862944650649, "acc_norm": 0.606655290102389, "acc_norm_stderr": 0.014275101465693028 }, "harness|hellaswag|10": { "acc": 0.6441943835889266, "acc_stderr": 0.004777782584817781, "acc_norm": 0.8419637522405895, "acc_norm_stderr": 0.003640294912838683 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.04793724854411022, "acc_norm": 0.35, "acc_norm_stderr": 0.04793724854411022 }, # 其他任务结果省略 }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_01_22T02_13_58.257879
- path: **/details_harness|arc:challenge|25_2024-01-22T02-13-58.257879.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-01-22T02-13-58.257879.parquet
- split: 2024_01_22T02_13_58.257879
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_01_22T02_13_58.257879
- path: **/details_harness|gsm8k|5_2024-01-22T02-13-58.257879.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-01-22T02-13-58.257879.parquet
- split: 2024_01_22T02_13_58.257879
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_01_22T02_13_58.257879
- path: **/details_harness|hellaswag|10_2024-01-22T02-13-58.257879.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-01-22T02-13-58.257879.parquet
- split: 2024_01_22T02_13_58.257879
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_01_22T02_13_58.257879
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-22T02-13-58.257879.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-22T02-13-58.257879.parquet
其他路径省略
- path:
- split: 2024_01_22T02_13_58.257879
- data_files:
}



