open-llm-leaderboard-old/details_bardsai__jaskier-7b-dpo-v4.1
收藏Hugging Face2024-02-14 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_bardsai__jaskier-7b-dpo-v4.1
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bardsai/jaskier-7b-dpo-v4.1在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中创建的,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。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。
该数据集是在模型bardsai/jaskier-7b-dpo-v4.1在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中创建的,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。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。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集信息
- 名称: Evaluation run of bardsai/jaskier-7b-dpo-v4.1
- 来源: 自动创建于模型 bardsai/jaskier-7b-dpo-v4.1 在 Open LLM Leaderboard 的评估运行中。
- 组成: 包含63个配置,每个配置对应一个评估任务。
- 创建: 从1次运行中创建,每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。"train" 分割始终指向最新结果。
- 额外配置: "results" 配置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
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_bardsai__jaskier-7b-dpo-v4.1", "harness_winogrande_5", split="train")
最新结果
- 最新结果来自: 2024-02-14T11:50:03.919128
- 示例结果: python { "all": { "acc": 0.6535379230152994, "acc_stderr": 0.03198200076203346, "acc_norm": 0.6529860127972825, "acc_norm_stderr": 0.032649840332740133, "mc1": 0.6132190942472461, "mc1_stderr": 0.017048857010515103, "mc2": 0.7592071044175611, "mc2_stderr": 0.01411814026868143 }, "harness|arc:challenge|25": { "acc": 0.712457337883959, "acc_stderr": 0.013226719056266129, "acc_norm": 0.7295221843003413, "acc_norm_stderr": 0.012980954547659556 }, "harness|hellaswag|10": { "acc": 0.7173869747062338, "acc_stderr": 0.0044934958720001085, "acc_norm": 0.8906592312288388, "acc_norm_stderr": 0.0031142850772280296 }, ... }
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_02_14T11_50_03.919128
- 路径: **/details_harness|arc:challenge|25_2024-02-14T11-50-03.919128.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-02-14T11-50-03.919128.parquet
- 分割: 2024_02_14T11_50_03.919128
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_02_14T11_50_03.919128
- 路径: **/details_harness|gsm8k|5_2024-02-14T11-50-03.919128.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-02-14T11-50-03.919128.parquet
- 分割: 2024_02_14T11_50_03.919128
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_02_14T11_50_03.919128
- 路径: **/details_harness|hellaswag|10_2024-02-14T11-50-03.919128.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-02-14T11-50-03.919128.parquet
- 分割: 2024_02_14T11_50_03.919128
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_02_14T11_50_03.919128
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-14T11-50-03.919128.parquet ...
- 路径:
- 分割: latest
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-14T11-50-03.919128.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-14T11-50-03.919128.parquet ...
- 路径:
- 分割: 2024_02_14T11_50_03.919128
- 数据文件:



