open-llm-leaderboard-old/details_intervitens__internlm2-base-20b-llama
收藏数据集概述
数据集简介
该数据集是在评估模型 intervitens/internlm2-base-20b-llama 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
数据集由 1 次运行创建,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。
数据集加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_intervitens__internlm2-base-20b-llama", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要: python { "all": { "acc": 0.6358012139707869, "acc_stderr": 0.032345068549400245, "acc_norm": 0.6407932610555195, "acc_norm_stderr": 0.03299002256236763, "mc1": 0.2864137086903305, "mc1_stderr": 0.015826142439502342, "mc2": 0.4411333942717491, "mc2_stderr": 0.014265599043931959 }, "harness|arc:challenge|25": { "acc": 0.5819112627986348, "acc_stderr": 0.014413988396996081, "acc_norm": 0.6296928327645052, "acc_norm_stderr": 0.01411129875167495 }, "harness|hellaswag|10": { "acc": 0.6154152559251145, "acc_stderr": 0.00485502724839816, "acc_norm": 0.8215494921330412, "acc_norm_stderr": 0.003821090082721705 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.047609522856952365, "acc_norm": 0.34, "acc_norm_stderr": 0.047609522856952365 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_01_18T19_55_30.426442
- path: **/details_harness|arc:challenge|25_2024-01-18T19-55-30.426442.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-01-18T19-55-30.426442.parquet
- split: 2024_01_18T19_55_30.426442
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_01_18T19_55_30.426442
- path: **/details_harness|gsm8k|5_2024-01-18T19-55-30.426442.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-01-18T19-55-30.426442.parquet
- split: 2024_01_18T19_55_30.426442
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_01_18T19_55_30.426442
- path: **/details_harness|hellaswag|10_2024-01-18T19-55-30.426442.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-01-18T19-55-30.426442.parquet
- split: 2024_01_18T19_55_30.426442
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_01_18T19_55_30.426442
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-18T19-55-30.426442.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-18T19-55-30.426442.parquet
其他文件路径...
- path:
- split: 2024_01_18T19_55_30.426442
- data_files:
}



