open-llm-leaderboard-old/details_Walmart-the-bag__Solar-10.7B-Cato
收藏数据集概述
数据集简介
该数据集是在评估模型Walmart-the-bag/Solar-10.7B-Cato在Open LLM Leaderboard上的自动创建的。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个配置
- 数据来源: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。
- 最新结果:"train"分割总是指向最新的结果。
- 结果汇总:一个额外的配置"results"存储所有运行结果的汇总,用于计算和显示在Open LLM Leaderboard上的汇总指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Walmart-the-bag__Solar-10.7B-Cato", "harness_winogrande_5", split="train")
最新结果
以下是2023-12-30T02:07:16.124496的最新结果:
python { "all": { "acc": 0.6601184986157275, "acc_stderr": 0.03173344410424321, "acc_norm": 0.6615926738267002, "acc_norm_stderr": 0.03237146731014547, "mc1": 0.4700122399020808, "mc1_stderr": 0.017471992091697544, "mc2": 0.6168232864590555, "mc2_stderr": 0.015630771495356736 }, "harness|arc:challenge|25": { "acc": 0.6390784982935154, "acc_stderr": 0.014034761386175452, "acc_norm": 0.6868600682593856, "acc_norm_stderr": 0.01355267154362349 }, "harness|hellaswag|10": { "acc": 0.6845249950209121, "acc_stderr": 0.0046375504780073636, "acc_norm": 0.8615813582951604, "acc_norm_stderr": 0.0034463307489637123 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.4, "acc_stderr": 0.049236596391733084, "acc_norm": 0.4, "acc_norm_stderr": 0.049236596391733084 }, # 其他任务的结果... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2023_12_30T02_07_16.124496
- 路径:
**/details_harness|arc:challenge|25_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:2023_12_30T02_07_16.124496
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2023_12_30T02_07_16.124496
- 路径:
**/details_harness|gsm8k|5_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:2023_12_30T02_07_16.124496
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2023_12_30T02_07_16.124496
- 路径:
**/details_harness|hellaswag|10_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-12-30T02-07-16.124496.parquet
- 路径:
- 分割:2023_12_30T02_07_16.124496
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2023_12_30T02_07_16.124496
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-anatomy|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-astronomy|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_biology|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-college_physics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-computer_security|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-econometrics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-12-30T02-07-16.124496.parquet**/details_harness|hendrycksTest-global_facts|5_2023-12-30T02-07-16.124496.parquet
其他路径...
- 路径:
- 分割:2023_12_30T02_07_16.124496
- 数据文件:
}



