open-llm-leaderboard-old/details_revolutionarybukhari__Llama-2-7b-chat-finetune-AUTOMATE
收藏数据集概述
数据集摘要
该数据集是在对模型revolutionarybukhari/Llama-2-7b-chat-finetune-AUTOMATE进行评估运行期间自动创建的,评估结果发布在Open LLM Leaderboard上。
数据集组成
- 数据集包含64个配置,每个配置对应一个评估任务。
- 数据集从1次运行中创建,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。
- 额外的"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在Open LLM Leaderboard上。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_revolutionarybukhari__Llama-2-7b-chat-finetune-AUTOMATE", "harness_winogrande_5", split="train")
最新结果
以下是2023-11-12T17:51:35.598056运行的最新结果:
python { "all": { "acc": 0.48664997345720373, "acc_stderr": 0.03427289794847252, "acc_norm": 0.4932299888431757, "acc_norm_stderr": 0.03508094254293674, "mc1": 0.2876376988984088, "mc1_stderr": 0.015846315101394812, "mc2": 0.44729919889234016, "mc2_stderr": 0.015286276115878357, "em": 0.010906040268456376, "em_stderr": 0.0010636334198498001, "f1": 0.06768770973154396, "f1_stderr": 0.0017077194500790263 }, "harness|arc:challenge|25": { "acc": 0.4906143344709898, "acc_stderr": 0.014608816322065, "acc_norm": 0.5307167235494881, "acc_norm_stderr": 0.014583792546304037 }, "harness|hellaswag|10": { "acc": 0.5622385978888668, "acc_stderr": 0.004950973231188739, "acc_norm": 0.7559251145190201, "acc_norm_stderr": 0.004286594977390899 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.04560480215720684, "acc_norm": 0.29, "acc_norm_stderr": 0.04560480215720684 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_11_12T17_51_35.598056
- path: **/details_harness|arc:challenge|25_2023-11-12T17-51-35.598056.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-11-12T17-51-35.598056.parquet
- split: 2023_11_12T17_51_35.598056
-
config_name: harness_drop_3
- split: 2023_11_12T17_51_35.598056
- path: **/details_harness|drop|3_2023-11-12T17-51-35.598056.parquet
- split: latest
- path: **/details_harness|drop|3_2023-11-12T17-51-35.598056.parquet
- split: 2023_11_12T17_51_35.598056
-
config_name: harness_gsm8k_5
- split: 2023_11_12T17_51_35.598056
- path: **/details_harness|gsm8k|5_2023-11-12T17-51-35.598056.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-11-12T17-51-35.598056.parquet
- split: 2023_11_12T17_51_35.598056
-
config_name: harness_hellaswag_10
- split: 2023_11_12T17_51_35.598056
- path: **/details_harness|hellaswag|10_2023-11-12T17-51-35.598056.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-11-12T17-51-35.598056.parquet
- split: 2023_11_12T17_51_35.598056
-
config_name: harness_hendrycksTest_5
- split: 2023_11_12T17_51_35.598056
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-management|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-11-12T17-51-35.598056.parquet
- **/details_harness|hendry
- path:
- split: 2023_11_12T17_51_35.598056



