open-llm-leaderboard-old/details_automerger__PasticheAlloyingotneoy-7B
收藏数据集概述
数据集来源
该数据集是在对模型 automerger/PasticheAlloyingotneoy-7B 进行评估运行期间自动创建的,评估结果发布在 Open LLM Leaderboard 上。
数据集组成
数据集由 63 个配置组成,每个配置对应一个评估任务。数据集从 1 次运行中创建,每次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" 存储了所有运行的聚合结果,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_automerger__PasticheAlloyingotneoy-7B", "harness_winogrande_5", split="train")
最新结果
以下是最新结果(来自 2024-04-06T00:16:25.828016 运行)的部分摘录: python { "all": { "acc": 0.6508232795294623, "acc_stderr": 0.03210820833288333, "acc_norm": 0.6499765787776198, "acc_norm_stderr": 0.03278385724556176, "mc1": 0.6364749082007344, "mc1_stderr": 0.016838862883965834, "mc2": 0.7831045197403762, "mc2_stderr": 0.013715683104637283 }, "harness|arc:challenge|25": { "acc": 0.7150170648464164, "acc_stderr": 0.013191348179838793, "acc_norm": 0.734641638225256, "acc_norm_stderr": 0.012902554762313962 }, "harness|hellaswag|10": { "acc": 0.717486556462856, "acc_stderr": 0.004493015945599716, "acc_norm": 0.8908583947420833, "acc_norm_stderr": 0.003111795320787943 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.047258156262526045, "acc_norm": 0.33, "acc_norm_stderr": 0.047258156262526045 }, # 其他任务的结果省略... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_04_06T00_16_25.828016
- path: **/details_harness|arc:challenge|25_2024-04-06T00-16-25.828016.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-04-06T00-16-25.828016.parquet
- split: 2024_04_06T00_16_25.828016
-
config_name: harness_gsm8k_5
- split: 2024_04_06T00_16_25.828016
- path: **/details_harness|gsm8k|5_2024-04-06T00-16-25.828016.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-04-06T00-16-25.828016.parquet
- split: 2024_04_06T00_16_25.828016
-
config_name: harness_hellaswag_10
- split: 2024_04_06T00_16_25.828016
- path: **/details_harness|hellaswag|10_2024-04-06T00-16-25.828016.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-04-06T00-16-25.828016.parquet
- split: 2024_04_06T00_16_25.828016
-
config_name: harness_hendrycksTest_5
- split: 2024_04_06T00_16_25.828016
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-04-06T00-16-25.828016.parquet
-
其他路径省略...
- path:
- split: latest
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-04-06T00-16-25.828016.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-04-06T00-16-25.828016.parquet
-
其他路径省略...
- path:
- split: 2024_04_06T00_16_25.828016



