open-llm-leaderboard/details_yulan-team__YuLan-Chat-2-13b-fp16
收藏数据集概述
数据集简介
该数据集是在评估模型 yulan-team/YuLan-Chat-2-13b-fp16 在 Open LLM Leaderboard 上的自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。
数据集结构
- 配置数量:64 个配置
- 数据来源:2 次运行结果
- 数据分割:每个配置包含特定运行的分割,分割名称使用运行的时间戳。"train" 分割指向最新的结果。
- 额外配置:"results" 配置存储所有运行的聚合结果,用于计算和显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_yulan-team__YuLan-Chat-2-13b-fp16", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-17T22:36:31.968246 运行的最新结果: python { "all": { "em": 0.4928691275167785, "em_stderr": 0.005119947219854752, "f1": 0.5245050335570473, "f1_stderr": 0.00496450323367562, "acc": 0.4671763485886859, "acc_stderr": 0.01040878805468688 }, "harness|drop|3": { "em": 0.4928691275167785, "em_stderr": 0.005119947219854752, "f1": 0.5245050335570473, "f1_stderr": 0.00496450323367562 }, "harness|gsm8k|5": { "acc": 0.1379833206974981, "acc_stderr": 0.00949977732774685 }, "harness|winogrande|5": { "acc": 0.7963693764798737, "acc_stderr": 0.011317798781626911 } }
配置详情
以下是部分配置及其数据文件路径:
- harness_arc_challenge_25
- 分割:2023_08_09T14_21_39.906214
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T14:21:39.906214.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T14:21:39.906214.parquet
- 路径:
- 分割:2023_08_09T14_21_39.906214
- harness_drop_3
- 分割:2023_09_17T22_36_31.968246
- 路径:
**/details_harness|drop|3_2023-09-17T22-36-31.968246.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|drop|3_2023-09-17T22-36-31.968246.parquet
- 路径:
- 分割:2023_09_17T22_36_31.968246
- harness_gsm8k_5
- 分割:2023_09_17T22_36_31.968246
- 路径:
**/details_harness|gsm8k|5_2023-09-17T22-36-31.968246.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-17T22-36-31.968246.parquet
- 路径:
- 分割:2023_09_17T22_36_31.968246
- harness_hellaswag_10
- 分割:2023_08_09T14_21_39.906214
- 路径:
**/details_harness|hellaswag|10_2023-08-09T14:21:39.906214.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-09T14:21:39.906214.parquet
- 路径:
- 分割:2023_08_09T14_21_39.906214
- harness_hendrycksTest_5
- 分割:2023_08_09T14_21_39.906214
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-human_aging|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-international_law|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-machine_learning|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-management|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-marketing|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-nutrition|5_2023-08-09T14:21:39.906214.parquet - 路径:
**/details_harness|hendrycksTest-philosophy|5_2023-08-09T14:21:39.906214.parquet - 路径:`**/details_harness|hendrycksTest-prehistory|5_2023-08-09T14:21:39.
- 路径:
- 分割:2023_08_09T14_21_39.906214



