open-llm-leaderboard-old/details_YeungNLP__firefly-llama2-13b-chat
收藏数据集概述
数据集来源
该数据集是在评估模型 YeungNLP/firefly-llama2-13b-chat 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集组成
数据集由 64 个配置组成,每个配置对应一个评估任务。数据集从 2 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。
额外配置
一个额外的配置 "results" 存储了所有运行的聚合结果,用于计算和显示 Open LLM Leaderboard 上的聚合指标。
数据加载示例
以下是加载特定运行详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_YeungNLP__firefly-llama2-13b-chat", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果来自 run 2023-10-23T07:10:32.244484 的示例: python { "all": { "em": 0.11954697986577181, "em_stderr": 0.003322477507478937, "f1": 0.21568372483221465, "f1_stderr": 0.0035544327528378426, "acc": 0.4229972001701799, "acc_stderr": 0.010210734400170219 }, "harness|drop|3": { "em": 0.11954697986577181, "em_stderr": 0.003322477507478937, "f1": 0.21568372483221465, "f1_stderr": 0.0035544327528378426 }, "harness|gsm8k|5": { "acc": 0.09855951478392722, "acc_stderr": 0.008210320350946333 }, "harness|winogrande|5": { "acc": 0.7474348855564326, "acc_stderr": 0.012211148449394105 } }
配置详情
以下是部分配置的详细信息:
-
config_name: harness_arc_challenge_25
- split: 2023_10_03T11_22_10.318112
- path: **/details_harness|arc:challenge|25_2023-10-03T11-22-10.318112.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-10-03T11-22-10.318112.parquet
- split: 2023_10_03T11_22_10.318112
-
config_name: harness_drop_3
- split: 2023_10_23T07_10_32.244484
- path: **/details_harness|drop|3_2023-10-23T07-10-32.244484.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-23T07-10-32.244484.parquet
- split: 2023_10_23T07_10_32.244484
-
config_name: harness_gsm8k_5
- split: 2023_10_23T07_10_32.244484
- path: **/details_harness|gsm8k|5_2023-10-23T07-10-32.244484.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-23T07-10-32.244484.parquet
- split: 2023_10_23T07_10_32.244484
-
config_name: harness_hellaswag_10
- split: 2023_10_03T11_22_10.318112
- path: **/details_harness|hellaswag|10_2023-10-03T11-22-10.318112.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-10-03T11-22-10.318112.parquet
- split: 2023_10_03T11_22_10.318112
-
config_name: harness_hendrycksTest_5
- split: 2023_10_03T11_22_10.318112
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-management|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-10-03T11-22-10.318112.parquet
- **/details_harness|hendrycksTest-professional_law|5_
- path:
- split: 2023_10_03T11_22_10.318112



