open-llm-leaderboard/details_migtissera__SynthIA-7B-v1.3
收藏Hugging Face2023-10-28 更新2024-03-04 收录
下载链接：
https://hf-mirror.com/datasets/open-llm-leaderboard/details_migtissera__SynthIA-7B-v1.3
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是在模型migtissera/SynthIA-7B-v1.3在Open LLM Leaderboard上的评估运行期间自动生成的。数据集由64个配置组成，每个配置对应一个被评估的任务。数据集是从3次运行中创建的，每次运行在每个配置中表示为特定的分割，分割名称使用运行的时间戳。train分割始终指向最新的结果。一个名为results的额外配置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。README还提供了如何使用Python代码加载运行中的详细信息的示例，并包含了特定运行的最新结果。
This dataset was automatically generated during the evaluation run of the model migtissera/SynthIA-7B-v1.3 on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluated task. The dataset is constructed from three evaluation runs, where each run is represented as a specific split for each configuration, with the split name adopting the timestamp of the corresponding run. The train split always points to the most recent results. An additional configuration named "results" stores the aggregated results across all runs, which are utilized to compute and display the aggregate metrics on the Open LLM Leaderboard. The README also provides examples of how to load detailed run information using Python code, and includes the latest results for specific runs.
提供机构：
open-llm-leaderboard
原始信息汇总
数据集概述
数据集名称
- Evaluation run of migtissera/SynthIA-7B-v1.3
数据集来源
- 该数据集是在模型 migtissera/SynthIA-7B-v1.3 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集组成
- 该数据集包含 64 个配置，每个配置对应一个评估任务。
- 数据集从 3 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" 分割总是指向最新的结果。
- 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_migtissera__SynthIA-7B-v1.3", "harness_winogrande_5", split="train")
最新结果
- 这些是最新结果，来自运行 2023-10-28T14:24:19.449160： python { "all": { "em": 0.34375, "em_stderr": 0.004864023482291936, "f1": 0.43760067114094225, "f1_stderr": 0.004666454920595155, "acc": 0.4821837715185681, "acc_stderr": 0.010982434159881403 }, "harness|drop|3": { "em": 0.34375, "em_stderr": 0.004864023482291936, "f1": 0.43760067114094225, "f1_stderr": 0.004666454920595155 }, "harness|gsm8k|5": { "acc": 0.17589082638362397, "acc_stderr": 0.010487120635539617 }, "harness|winogrande|5": { "acc": 0.7884767166535123, "acc_stderr": 0.011477747684223188 } }
配置详情
-
harness_arc_challenge_25
- 分割：2023_10_09T11_48_18.823660
- 路径：
**/details_harness|arc:challenge|25_2023-10-09T11-48-18.823660.parquet
- 路径：
- 分割：2023_10_09T11_58_55.532772
- 路径：
**/details_harness|arc:challenge|25_2023-10-09T11-58-55.532772.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|arc:challenge|25_2023-10-09T11-58-55.532772.parquet
- 路径：
- 分割：2023_10_09T11_48_18.823660
-
harness_drop_3
- 分割：2023_10_28T14_24_19.449160
- 路径：
**/details_harness|drop|3_2023-10-28T14-24-19.449160.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|drop|3_2023-10-28T14-24-19.449160.parquet
- 路径：
- 分割：2023_10_28T14_24_19.449160
-
harness_gsm8k_5
- 分割：2023_10_28T14_24_19.449160
- 路径：
**/details_harness|gsm8k|5_2023-10-28T14-24-19.449160.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|gsm8k|5_2023-10-28T14-24-19.449160.parquet
- 路径：
- 分割：2023_10_28T14_24_19.449160
-
harness_hellaswag_10
- 分割：2023_10_09T11_48_18.823660
- 路径：
**/details_harness|hellaswag|10_2023-10-09T11-48-18.823660.parquet
- 路径：
- 分割：2023_10_09T11_58_55.532772
- 路径：
**/details_harness|hellaswag|10_2023-10-09T11-58-55.532772.parquet
- 路径：
- 分割：latest
- 路径：
**/details_harness|hellaswag|10_2023-10-09T11-58-55.532772.parquet
- 路径：
- 分割：2023_10_09T11_48_18.823660
-
harness_hendrycksTest_5
- 分割：2023_10_09T11_48_18.823660
- 路径：
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-anatomy|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-astronomy|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-business_ethics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_biology|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_medicine|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-college_physics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-computer_security|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-econometrics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-formal_logic|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-global_facts|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_biology|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_european_history|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_geography|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_physics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_psychology|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_statistics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_us_history|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-high_school_world_history|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-human_aging|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-human_sexuality|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-international_law|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-jurisprudence|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-logical_fallacies|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-machine_learning|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-management|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-marketing|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-medical_genetics|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-miscellaneous|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-moral_disputes|5_2023-10-09T11-48-18.823660.parquet - 路径：
**/details_harness|hendrycksTest-moral_scenarios|5_2023-10-09T11-48-18.823660.parquet - 路径：`**/details_harness|hendrycksTest-nutrition|5_2023-10-09T1
- 路径：
- 分割：2023_10_09T11_48_18.823660
