open-llm-leaderboard/details_jondurbin__airoboros-33b-gpt4-1.3
收藏Hugging Face2023-10-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_jondurbin__airoboros-33b-gpt4-1.3
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型jondurbin/airoboros-33b-gpt4-1.3进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由4次运行创建,每次运行的结果作为特定配置中的一个分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
This dataset was automatically created during the evaluation of the model jondurbin/airoboros-33b-gpt4-1.3 on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluation task. The dataset is generated from 4 runs, where the results of each run serve as a data split under a specific configuration, with the split name being the timestamp of the run. The 'train' split always points to the most recent evaluation results. In addition, there is a configuration named 'results' that stores the aggregated results of all runs, and is used to calculate and display the aggregated metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集来源
- 数据集是在对模型
jondurbin/airoboros-33b-gpt4-1.3进行评估运行期间自动创建的。
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 4 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
- 加载特定运行的详细信息示例: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_jondurbin__airoboros-33b-gpt4-1.3", "harness_winogrande_5", split="train")
最新结果
- 最新结果来自 2023-10-22T21:52:38.405069 运行,包含以下指标: python { "all": { "em": 0.01950503355704698, "em_stderr": 0.0014162361849700588, "f1": 0.08725671140939598, "f1_stderr": 0.0019162183364708429, "acc": 0.45865000212425644, "acc_stderr": 0.010392294007638424 }, "harness|drop|3": { "em": 0.01950503355704698, "em_stderr": 0.0014162361849700588, "f1": 0.08725671140939598, "f1_stderr": 0.0019162183364708429 }, "harness|gsm8k|5": { "acc": 0.13040181956027294, "acc_stderr": 0.009275630324554088 }, "harness|winogrande|5": { "acc": 0.7868981846882399, "acc_stderr": 0.01150895769072276 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_08_18T14_12_32.965020
- 路径:
**/details_harness|arc:challenge|25_2023-08-18T14:12:32.965020.parquet
- 路径:
- 分割:2023_08_18T17_42_39.017472
- 路径:
**/details_harness|arc:challenge|25_2023-08-18T17:42:39.017472.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-18T17:42:39.017472.parquet
- 路径:
- 分割:2023_08_18T14_12_32.965020
-
harness_drop_3
- 分割:2023_10_18T13_47_41.719398
- 路径:
**/details_harness|drop|3_2023-10-18T13-47-41.719398.parquet
- 路径:
- 分割:2023_10_22T21_52_38.405069
- 路径:
**/details_harness|drop|3_2023-10-22T21-52-38.405069.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-22T21-52-38.405069.parquet
- 路径:
- 分割:2023_10_18T13_47_41.719398
-
harness_gsm8k_5
- 分割:2023_10_18T13_47_41.719398
- 路径:
**/details_harness|gsm8k|5_2023-10-18T13-47-41.719398.parquet
- 路径:
- 分割:2023_10_22T21_52_38.405069
- 路径:
**/details_harness|gsm8k|5_2023-10-22T21-52-38.405069.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-22T21-52-38.405069.parquet
- 路径:
- 分割:2023_10_18T13_47_41.719398
-
harness_hellaswag_10
- 分割:2023_08_18T14_12_32.965020
- 路径:
**/details_harness|hellaswag|10_2023-08-18T14:12:32.965020.parquet
- 路径:
- 分割:2023_08_18T17_42_39.017472
- 路径:
**/details_harness|hellaswag|10_2023-08-18T17:42:39.017472.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-18T17:42:39.017472.parquet
- 路径:
- 分割:2023_08_18T14_12_32.965020
-
harness_hendrycksTest_5
- 分割:2023_08_18T14_12_32.965020
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_physics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-human_aging|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-human_sexuality|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-international_law|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-jurisprudence|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-machine_learning|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-management|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-marketing|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-medical_genetics|5_2023-08-18T14:12:32.965020.parquet - 路径:
**/details_harness|hendrycksTest-miscellaneous|5_2023-08-18T14:12:32.965020.parquet - 路径:`**/details_harness|hendrycksTest-moral_disputes|5_2023-08-1
- 路径:
- 分割:2023_08_18T14_12_32.965020



