open-llm-leaderboard/details_pszemraj__pythia-31m-simplepile-lite-2048-scratch-2e
收藏数据集概述
数据集简介
该数据集是在评估模型 pszemraj/pythia-31m-simplepile-lite-2048-scratch-2e 在 Open LLM Leaderboard 上的自动创建的。
数据集结构
- 配置数量:64个配置
- 每个配置对应:一个评估任务
- 创建来源:2次运行
- 分割命名:使用运行的时间戳命名
- 训练分割:总是指向最新的结果
- 额外配置:"results" 存储所有运行的聚合结果,用于计算和显示聚合指标在 Open LLM Leaderboard 上。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_pszemraj__pythia-31m-simplepile-lite-2048-scratch-2e", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果来自 run 2023-10-29T08:23:53.788687: python { "all": { "em": 0.0, "em_stderr": 0.0, "f1": 0.013173238255033595, "f1_stderr": 0.0006780799719584048, "acc": 0.2430939226519337, "acc_stderr": 0.007023561458220214 }, "harness|drop|3": { "em": 0.0, "em_stderr": 0.0, "f1": 0.013173238255033595, "f1_stderr": 0.0006780799719584048 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.4861878453038674, "acc_stderr": 0.014047122916440427 } }
配置详情
以下是部分配置的详细信息:
harness_arc_challenge_25
- 分割:2023_09_15T02_33_28.434713
- 路径:
- **/details_harness|arc:challenge|25_2023-09-15T02-33-28.434713.parquet
- 分割:latest
- 路径:
- **/details_harness|arc:challenge|25_2023-09-15T02-33-28.434713.parquet
harness_drop_3
- 分割:2023_10_29T08_23_53.788687
- 路径:
- **/details_harness|drop|3_2023-10-29T08-23-53.788687.parquet
- 分割:latest
- 路径:
- **/details_harness|drop|3_2023-10-29T08-23-53.788687.parquet
harness_gsm8k_5
- 分割:2023_10_29T08_23_53.788687
- 路径:
- **/details_harness|gsm8k|5_2023-10-29T08-23-53.788687.parquet
- 分割:latest
- 路径:
- **/details_harness|gsm8k|5_2023-10-29T08-23-53.788687.parquet
harness_hellaswag_10
- 分割:2023_09_15T02_33_28.434713
- 路径:
- **/details_harness|hellaswag|10_2023-09-15T02-33-28.434713.parquet
- 分割:latest
- 路径:
- **/details_harness|hellaswag|10_2023-09-15T02-33-28.434713.parquet
harness_hendrycksTest_5
- 分割:2023_09_15T02_33_28.434713
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-management|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-09-15T02-33-28.434713.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2023-09-15T02-33-28.434713.parquet
- **/details_



