open-llm-leaderboard-old/details_NovoCode__Mistral-NeuralDPO-v0.2
收藏数据集概述
该数据集是在模型 NovoCode/Mistral-NeuralDPO-v0.2 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集组成
- 数据集包含 63 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。
- "train" 分割始终指向最新的结果。
- 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 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_NovoCode__Mistral-NeuralDPO-v0.2", "harness_winogrande_5", split="train")
最新结果
以下是 2024-02-19T06:05:02.538457 运行的最新结果:
python { "all": { "acc": 0.6262346035274842, "acc_stderr": 0.03258054476573943, "acc_norm": 0.6312971512345912, "acc_norm_stderr": 0.03324902767301004, "mc1": 0.3390452876376989, "mc1_stderr": 0.016571797910626615, "mc2": 0.48726838826281, "mc2_stderr": 0.015809965700715564 }, "harness|arc:challenge|25": { "acc": 0.6382252559726962, "acc_stderr": 0.014041957945038075, "acc_norm": 0.6706484641638225, "acc_norm_stderr": 0.013734057652635474 }, "harness|hellaswag|10": { "acc": 0.6564429396534555, "acc_stderr": 0.0047392481181180056, "acc_norm": 0.8501294562836088, "acc_norm_stderr": 0.003562149890962712 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.04688261722621503, "acc_norm": 0.32, "acc_norm_stderr": 0.04688261722621503 }, # 其他任务的结果省略... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_02_19T06_05_02.538457
- path: **/details_harness|arc:challenge|25_2024-02-19T06-05-02.538457.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-02-19T06-05-02.538457.parquet
- split: 2024_02_19T06_05_02.538457
-
config_name: harness_gsm8k_5
- split: 2024_02_19T06_05_02.538457
- path: **/details_harness|gsm8k|5_2024-02-19T06-05-02.538457.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-02-19T06-05-02.538457.parquet
- split: 2024_02_19T06_05_02.538457
-
config_name: harness_hellaswag_10
- split: 2024_02_19T06_05_02.538457
- path: **/details_harness|hellaswag|10_2024-02-19T06-05-02.538457.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-02-19T06-05-02.538457.parquet
- split: 2024_02_19T06_05_02.538457
-
config_name: harness_hendrycksTest_5
- split: 2024_02_19T06_05_02.538457
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-19T06-05-02.538457.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-02-19T06
- path:
- split: 2024_02_19T06_05_02.538457



