open-llm-leaderboard-old/details_ChaoticNeutrals__Prodigy_7B
收藏数据集概述
数据集来源
该数据集是在评估模型 ChaoticNeutrals/Prodigy_7B 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集结构
- 数据集包含 63 个配置,每个配置对应一个评估任务。
- 数据集由 1 次运行创建,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。
- 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_ChaoticNeutrals__Prodigy_7B", "harness_winogrande_5", split="train")
最新结果
这些是最新的结果,来自 2024-02-29T16:22:22.877763 的运行: python { "all": { "acc": 0.6541124776719341, "acc_stderr": 0.03200942052047841, "acc_norm": 0.6542696674198764, "acc_norm_stderr": 0.032672093467498255, "mc1": 0.5324357405140759, "mc1_stderr": 0.017466632149577613, "mc2": 0.6856534678008832, "mc2_stderr": 0.0149101836867707 }, "harness|arc:challenge|25": { "acc": 0.6945392491467577, "acc_stderr": 0.013460080478002508, "acc_norm": 0.7158703071672355, "acc_norm_stderr": 0.013179442447653886 }, "harness|hellaswag|10": { "acc": 0.7122087233618801, "acc_stderr": 0.00451808059452802, "acc_norm": 0.8809002190798646, "acc_norm_stderr": 0.0032324391398815527 }, "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... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_02_29T16_22_22.877763
- path: **/details_harness|arc:challenge|25_2024-02-29T16-22-22.877763.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-02-29T16-22-22.877763.parquet
- split: 2024_02_29T16_22_22.877763
-
config_name: harness_gsm8k_5
- split: 2024_02_29T16_22_22.877763
- path: **/details_harness|gsm8k|5_2024-02-29T16-22-22.877763.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-02-29T16-22-22.877763.parquet
- split: 2024_02_29T16_22_22.877763
-
config_name: harness_hellaswag_10
- split: 2024_02_29T16_22_22.877763
- path: **/details_harness|hellaswag|10_2024-02-29T16-22-22.877763.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-02-29T16-22-22.877763.parquet
- split: 2024_02_29T16_22_22.877763
-
config_name: harness_hendrycksTest_5
- split: 2024_02_29T16_22_22.877763
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-29T16-22-22.877763.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-29T16-22-22.877763.parquet
- path:
- split: 2024_02_29T16_22_22.877763



