open-llm-leaderboard/details_Aspik101__Redmond-Puffin-13B-instruct-PL-lora_unload
收藏资源简介:
该数据集是在模型Aspik101/Redmond-Puffin-13B-instruct-PL-lora_unload在Open LLM Leaderboard上的评估运行期间自动创建的。它由64个配置组成,每个配置对应一个被评估的任务。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割找到。train分割始终指向最新的结果。一个额外的配置results存储了运行的所有聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
This dataset was automatically generated during the evaluation run of the model Aspik101/Redmond-Puffin-13B-instruct-PL-lora_unload on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to one evaluated task. This dataset is created from two separate runs, where each run corresponds to a specific split within every configuration. The train split within each configuration always points to the most recent results. An additional configuration named "results" stores all aggregated results from the runs, which are used to calculate and display the aggregated metrics on the Open LLM Leaderboard. This dataset can be loaded using the `load_dataset` function from the `datasets` library.
数据集概述
数据集来源
该数据集是在评估模型 Aspik101/Redmond-Puffin-13B-instruct-PL-lora_unload 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集结构
- 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 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。
- 一个额外的配置 "results" 存储所有运行结果的聚合,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_Aspik101__Redmond-Puffin-13B-instruct-PL-lora_unload", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-17T13:30:13.279131 运行的最新结果: python { "all": { "em": 0.002936241610738255, "em_stderr": 0.0005541113054710113, "f1": 0.05736577181208053, "f1_stderr": 0.0013664047590611983, "acc": 0.43379799697577687, "acc_stderr": 0.010348919090911759 }, "harness|drop|3": { "em": 0.002936241610738255, "em_stderr": 0.0005541113054710113, "f1": 0.05736577181208053, "f1_stderr": 0.0013664047590611983 }, "harness|gsm8k|5": { "acc": 0.1106899166034875, "acc_stderr": 0.008642172551392473 }, "harness|winogrande|5": { "acc": 0.7569060773480663, "acc_stderr": 0.012055665630431044 } }
配置详情
-
harness_arc_challenge_25
- 分割: 2023_08_09T11_16_00.382833
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T11:16:00.382833.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T11:16:00.382833.parquet
- 路径:
- 分割: 2023_08_09T11_16_00.382833
-
harness_drop_3
- 分割: 2023_10_17T13_30_13.279131
- 路径:
**/details_harness|drop|3_2023-10-17T13-30-13.279131.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|drop|3_2023-10-17T13-30-13.279131.parquet
- 路径:
- 分割: 2023_10_17T13_30_13.279131
-
harness_gsm8k_5
- 分割: 2023_10_17T13_30_13.279131
- 路径:
**/details_harness|gsm8k|5_2023-10-17T13-30-13.279131.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|gsm8k|5_2023-10-17T13-30-13.279131.parquet
- 路径:
- 分割: 2023_10_17T13_30_13.279131
-
harness_hellaswag_10
- 分割: 2023_08_09T11_16_00.382833
- 路径:
**/details_harness|hellaswag|10_2023-08-09T11:16:00.382833.parquet
- 路径:
- 分割: latest
- 路径:
**/details_harness|hellaswag|10_2023-08-09T11:16:00.382833.parquet
- 路径:
- 分割: 2023_08_09T11_16_00.382833
-
harness_hendrycksTest_5
- 分割: 2023_08_09T11_16_00.382833
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-anatomy|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-astronomy|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_biology|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-college_physics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-computer_security|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-econometrics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-global_facts|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-human_aging|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-international_law|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-management|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-marketing|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-nutrition|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-philosophy|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-prehistory|5_2023-08-09T11:16:00.382833.parquet**/details_harness|hendrycksTest-professional_accounting|5_2023-08-09T11:16:00.382833.parquet- `**/details_harness|hend
- 路径:
- 分割: 2023_08_09T11_16_00.382833



