open-llm-leaderboard-old/details_wang7776__Mistral-7B-Instruct-v0.2-sparsity-30
收藏Hugging Face2024-01-15 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_wang7776__Mistral-7B-Instruct-v0.2-sparsity-30
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在评估模型wang7776/Mistral-7B-Instruct-v0.2-sparsity-30时自动创建的,评估是在Open LLM Leaderboard上进行的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行可以在每个配置中找到特定的分割,分割以运行的时间戳命名。train分割始终指向最新的结果。此外,results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。
该数据集是在评估模型wang7776/Mistral-7B-Instruct-v0.2-sparsity-30时自动创建的,评估是在Open LLM Leaderboard上进行的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行可以在每个配置中找到特定的分割,分割以运行的时间戳命名。train分割始终指向最新的结果。此外,results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。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- 名称: Evaluation run of wang7776/Mistral-7B-Instruct-v0.2-sparsity-30
数据集描述
- 创建目的: 该数据集是在评估模型 wang7776/Mistral-7B-Instruct-v0.2-sparsity-30 在 Open LLM Leaderboard 上的表现时自动创建的。
- 组成: 数据集包含 63 个配置,每个配置对应一个评估任务。
- 创建过程: 数据集从 1 次运行中创建,每个运行可以在每个配置中作为一个特定的分割找到,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,用于计算和显示在 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_wang7776__Mistral-7B-Instruct-v0.2-sparsity-30", "harness_winogrande_5", split="train")
最新结果
- 结果来源: 这些是最新的结果,来自 2024-01-15T11:11:23.952137 运行。
- 详细结果: python { "all": { "acc": 0.46463869066130487, "acc_stderr": 0.034455801387647846, "acc_norm": 0.4711053080225253, "acc_norm_stderr": 0.035249688625421514, "mc1": 0.2998776009791922, "mc1_stderr": 0.016040352966713623, "mc2": 0.4553313356020083, "mc2_stderr": 0.01500792603148901 }, "harness|arc:challenge|25": { "acc": 0.45563139931740615, "acc_stderr": 0.014553749939306864, "acc_norm": 0.5110921501706485, "acc_norm_stderr": 0.014607794914013048 }, "harness|hellaswag|10": { "acc": 0.5652260505875324, "acc_stderr": 0.004947141797384131, "acc_norm": 0.7572196773551085, "acc_norm_stderr": 0.004278871104930366 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.25, "acc_stderr": 0.04351941398892446, "acc_norm": 0.25, "acc_norm_stderr": 0.04351941398892446 }, ... }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_01_15T11_11_23.952137
- 路径: **/details_harness|arc:challenge|25_2024-01-15T11-11-23.952137.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-01-15T11-11-23.952137.parquet
- 分割: 2024_01_15T11_11_23.952137
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_01_15T11_11_23.952137
- 路径: **/details_harness|gsm8k|5_2024-01-15T11-11-23.952137.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-01-15T11-11-23.952137.parquet
- 分割: 2024_01_15T11_11_23.952137
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_01_15T11_11_23.952137
- 路径: **/details_harness|hellaswag|10_2024-01-15T11-11-23.952137.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-01-15T11-11-23.952137.parquet
- 分割: 2024_01_15T11_11_23.952137
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_01_15T11_11_23.952137
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-15T11-11-23.952137.parquet
- ...
- 路径:
- 分割: latest
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-15T11-11-23.952137.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-15T11-11-23.952137.parquet
- ...
- 路径:
- 分割: 2024_01_15T11_11_23.952137
- 数据文件:



