open-llm-leaderboard-old/details_ignos__LeoScorpius-GreenNode-Platypus-7B-v1
收藏数据集概述
该数据集是在对模型 ignos/LeoScorpius-GreenNode-Platypus-7B-v1 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
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作为一个特定的分割找到,分割名称使用运行的时间戳。
- "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 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_ignos__LeoScorpius-GreenNode-Platypus-7B-v1", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要:
python { "all": { "acc": 0.6253622649843006, "acc_stderr": 0.03256427406790904, "acc_norm": 0.6254417624925451, "acc_norm_stderr": 0.033235993868121656, "mc1": 0.36107711138310894, "mc1_stderr": 0.016814312844836886, "mc2": 0.5277604131615456, "mc2_stderr": 0.01507398033093283 }, "harness|arc:challenge|25": { "acc": 0.6279863481228669, "acc_stderr": 0.014124597881844461, "acc_norm": 0.6604095563139932, "acc_norm_stderr": 0.013839039762820167 }, "harness|hellaswag|10": { "acc": 0.6799442342162916, "acc_stderr": 0.00465544276659947, "acc_norm": 0.8652658832901813, "acc_norm_stderr": 0.003407415513326042 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.25, "acc_stderr": 0.04351941398892446, "acc_norm": 0.25, "acc_norm_stderr": 0.04351941398892446 }, # 其他任务的结果... }
配置详情
-
harness_arc_challenge_25
- 分割:
2023_12_16T16_00_30.709211 - 路径:
**/details_harness|arc:challenge|25_2023-12-16T16-00-30.709211.parquet - 分割:
latest - 路径:
**/details_harness|arc:challenge|25_2023-12-16T16-00-30.709211.parquet
- 分割:
-
harness_gsm8k_5
- 分割:
2023_12_16T16_00_30.709211 - 路径:
**/details_harness|gsm8k|5_2023-12-16T16-00-30.709211.parquet - 分割:
latest - 路径:
**/details_harness|gsm8k|5_2023-12-16T16-00-30.709211.parquet
- 分割:
-
harness_hellaswag_10
- 分割:
2023_12_16T16_00_30.709211 - 路径:
**/details_harness|hellaswag|10_2023-12-16T16-00-30.709211.parquet - 分割:
latest - 路径:
**/details_harness|hellaswag|10_2023-12-16T16-00-30.709211.parquet
- 分割:
-
harness_hendrycksTest_5
- 分割:
2023_12_16T16_00_30.709211 - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_physics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_psychology|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_statistics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_us_history|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-high_school_world_history|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-human_aging|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-human_sexuality|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-international_law|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-jurisprudence|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-logical_fallacies|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-machine_learning|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-management|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-marketing|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-medical_genetics|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-miscellaneous|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-moral_disputes|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-moral_scenarios|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-nutrition|5_2023-12-16T16-00-30.709211.parquet - 路径:
**/details_harness|hendrycksTest-philosophy|5_2023-12-16T16-00-30.709211.parquet - 路径:`**/details_harness|hendrycksTest-prehistory|5_2023-12-16T
- 分割:



