open-llm-leaderboard/details_hyunseoki__ko-en-llama2-13b
收藏资源简介:
该数据集是在Open LLM Leaderboard上对模型hyunseoki/ko-en-llama2-13b进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含2次运行的数据,每次运行都作为一个特定的分割表示,分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外,名为results的配置存储了用于在排行榜上计算指标的聚合结果。
This dataset was automatically generated during the evaluation of the model hyunseoki/ko-en-llama2-13b on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to one of the evaluated tasks. The dataset includes data from two runs, where each run is represented as a distinct split, with the split name being the timestamp of the corresponding run. The `train` split always references the most up-to-date results. Furthermore, a configuration named `results` stores the aggregated results used for computing metrics on the leaderboard.
数据集概述
数据集来源
该数据集是在评估模型 hyunseoki/ko-en-llama2-13b 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。
- 运行次数:数据集由2次运行创建。每次运行在每个配置中作为一个特定的分片存在,分片名称使用运行的时间戳。
- 训练分片:"train" 分片始终指向最新的结果。
- 结果配置: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_hyunseoki__ko-en-llama2-13b", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-27T07:23:26.353656 运行的最新结果: python { "all": { "em": 0.28114513422818793, "em_stderr": 0.004603896433799628, "f1": 0.3260591442953026, "f1_stderr": 0.004539391567050269, "acc": 0.3779028263381469, "acc_stderr": 0.007293885306168497 }, "harness|drop|3": { "em": 0.28114513422818793, "em_stderr": 0.004603896433799628, "f1": 0.3260591442953026, "f1_stderr": 0.004539391567050269 }, "harness|gsm8k|5": { "acc": 0.0075815011372251705, "acc_stderr": 0.002389281512077218 }, "harness|winogrande|5": { "acc": 0.7482241515390686, "acc_stderr": 0.012198489100259776 } }
配置详情
以下是部分配置及其数据文件路径:
-
harness_arc_challenge_25
- 分片:2023_10_04T07_33_17.210034
- 路径:
**/details_harness|arc:challenge|25_2023-10-04T07-33-17.210034.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|arc:challenge|25_2023-10-04T07-33-17.210034.parquet
- 路径:
- 分片:2023_10_04T07_33_17.210034
-
harness_drop_3
- 分片:2023_10_27T07_23_26.353656
- 路径:
**/details_harness|drop|3_2023-10-27T07-23-26.353656.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|drop|3_2023-10-27T07-23-26.353656.parquet
- 路径:
- 分片:2023_10_27T07_23_26.353656
-
harness_gsm8k_5
- 分片:2023_10_27T07_23_26.353656
- 路径:
**/details_harness|gsm8k|5_2023-10-27T07-23-26.353656.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-27T07-23-26.353656.parquet
- 路径:
- 分片:2023_10_27T07_23_26.353656
-
harness_hellaswag_10
- 分片:2023_10_04T07_33_17.210034
- 路径:
**/details_harness|hellaswag|10_2023-10-04T07-33-17.210034.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|hellaswag|10_2023-10-04T07-33-17.210034.parquet
- 路径:
- 分片:2023_10_04T07_33_17.210034
-
harness_hendrycksTest_5
- 分片:2023_10_04T07_33_17.210034
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_physics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_psychology|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_statistics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_us_history|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-high_school_world_history|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-human_aging|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-human_sexuality|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-international_law|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-jurisprudence|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-logical_fallacies|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-machine_learning|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-management|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-marketing|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-medical_genetics|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-miscellaneous|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-moral_disputes|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-moral_scenarios|5_2023-10-04T07-33-17.210034.parquet - 路径:
**/details_harness|hendrycksTest-nutrition|5_2023-10-04T07-33-17.210034.parquet - 路径:`**/details_harness|hendrycksTest-philosophy|5_2023-10-04T07-33-17.210034
- 路径:
- 分片:2023_10_04T07_33_17.210034



