open-llm-leaderboard-old/details_PeanutJar__Mistral-v0.1-PeanutButter-v0.0.5-DPO-7B-QLoRA
收藏数据集概述
数据集简介
该数据集是在评估模型PeanutJar/Mistral-v0.1-PeanutButter-v0.0.5-DPO-7B-QLoRA在Open LLM Leaderboard上的运行过程中自动创建的。
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。
- 数据来源: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。
- 最新结果:"train"分割始终指向最新的结果。
- 结果汇总:一个额外的配置"results"存储所有运行的汇总结果,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_PeanutJar__Mistral-v0.1-PeanutButter-v0.0.5-DPO-7B-QLoRA_public", "harness_winogrande_5", split="train")
最新结果
以下是2023-11-19T15:36:58.669943运行的最新结果:
python { "all": { "acc": 0.6304363255253682, "acc_stderr": 0.03221802645295953, "acc_norm": 0.6394242549229363, "acc_norm_stderr": 0.03291185339071779, "mc1": 0.3108935128518972, "mc1_stderr": 0.016203316673559693, "mc2": 0.45750130636610203, "mc2_stderr": 0.014435953658631701, "em": 0.005557885906040268, "em_stderr": 0.0007613497667018497, "f1": 0.06505977348993297, "f1_stderr": 0.0015006861389720051 }, "harness|arc:challenge|25": { "acc": 0.5793515358361775, "acc_stderr": 0.014426211252508401, "acc_norm": 0.6126279863481229, "acc_norm_stderr": 0.014235872487909869 }, "harness|hellaswag|10": { "acc": 0.641804421429994, "acc_stderr": 0.004784901248558711, "acc_norm": 0.8452499502091216, "acc_norm_stderr": 0.003609271000593054 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.27, "acc_stderr": 0.0446196043338474, "acc_norm": 0.27, "acc_norm_stderr": 0.0446196043338474 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2023_11_19T15_36_58.669943
- path: [**/details_harness|arc:challenge|25_2023-11-19T15-36-58.669943.parquet]
- split: latest
- path: [**/details_harness|arc:challenge|25_2023-11-19T15-36-58.669943.parquet]
- split: 2023_11_19T15_36_58.669943
- data_files:
-
config_name: harness_drop_3
- data_files:
- split: 2023_11_19T15_36_58.669943
- path: [**/details_harness|drop|3_2023-11-19T15-36-58.669943.parquet]
- split: latest
- path: [**/details_harness|drop|3_2023-11-19T15-36-58.669943.parquet]
- split: 2023_11_19T15_36_58.669943
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2023_11_19T15_36_58.669943
- path: [**/details_harness|gsm8k|5_2023-11-19T15-36-58.669943.parquet]
- split: latest
- path: [**/details_harness|gsm8k|5_2023-11-19T15-36-58.669943.parquet]
- split: 2023_11_19T15_36_58.669943
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2023_11_19T15_36_58.669943
- path: [**/details_harness|hellaswag|10_2023-11-19T15-36-58.669943.parquet]
- split: latest
- path: [**/details_harness|hellaswag|10_2023-11-19T15-36-58.669943.parquet]
- split: 2023_11_19T15_36_58.669943
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2023_11_19T15_36_58.669943
- path: [ **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-anatomy|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-astronomy|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-business_ethics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_biology|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_chemistry|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_computer_science|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_mathematics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_medicine|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-college_physics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-computer_security|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-econometrics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-electrical_engineering|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-elementary_mathematics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-formal_logic|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-global_facts|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_biology|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_chemistry|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_computer_science|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_european_history|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_geography|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_mathematics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_physics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_psychology|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_statistics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_us_history|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-high_school_world_history|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-human_aging|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-human_sexuality|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-international_law|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-jurisprudence|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-logical_fallacies|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-machine_learning|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-management|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-marketing|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-medical_genetics|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hendrycksTest-miscellaneous|5_2023-11-19T15-36-58.669943.parquet, **/details_harness|hend
- split: 2023_11_19T15_36_58.669943
- data_files:



