open-llm-leaderboard-old/details_ChaoticNeutrals__Eris_Remix_DPO_7B
收藏Hugging Face2024-03-07 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ChaoticNeutrals__Eris_Remix_DPO_7B
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ChaoticNeutrals/Eris_Remix_DPO_7B在Open LLM Leaderboard上的评估运行期间自动创建的。它由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,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。可以使用提供的Python代码片段加载数据集。
该数据集是在模型ChaoticNeutrals/Eris_Remix_DPO_7B在Open LLM Leaderboard上的评估运行期间自动创建的。它由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,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。可以使用提供的Python代码片段加载数据集。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集信息
- 数据集名称: Evaluation run of ChaoticNeutrals/Eris_Remix_DPO_7B
- 数据集来源: 自动创建于模型 ChaoticNeutrals/Eris_Remix_DPO_7B 在 Open LLM Leaderboard 的评估运行中。
- 数据集组成: 包含63个配置,每个配置对应一个评估任务。
- 数据集创建: 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,分割名称使用运行的时间戳。"train" 分割总是指向最新的结果。
- 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_ChaoticNeutrals__Eris_Remix_DPO_7B", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源: 2024-03-07T11:43:51.829104
- 结果示例: python { "all": { "acc": 0.6583336356900076, "acc_stderr": 0.031977540942622035, "acc_norm": 0.6579180688874828, "acc_norm_stderr": 0.03264643438189693, "mc1": 0.5385556915544676, "mc1_stderr": 0.017451384104637452, "mc2": 0.6891701037064589, "mc2_stderr": 0.01500545106241752 }, "harness|arc:challenge|25": { "acc": 0.6988054607508533, "acc_stderr": 0.01340674176784763, "acc_norm": 0.7244027303754266, "acc_norm_stderr": 0.01305716965576184 }, "harness|hellaswag|10": { "acc": 0.7110137422824139, "acc_stderr": 0.0045236511840162615, "acc_norm": 0.8803027285401315, "acc_norm_stderr": 0.0032394379508979255 }, ... }
配置信息
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_03_07T11_43_51.829104
- 路径: **/details_harness|arc:challenge|25_2024-03-07T11-43-51.829104.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-03-07T11-43-51.829104.parquet
- 分割: 2024_03_07T11_43_51.829104
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_03_07T11_43_51.829104
- 路径: **/details_harness|gsm8k|5_2024-03-07T11-43-51.829104.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-03-07T11-43-51.829104.parquet
- 分割: 2024_03_07T11_43_51.829104
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_03_07T11_43_51.829104
- 路径: **/details_harness|hellaswag|10_2024-03-07T11-43-51.829104.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-03-07T11-43-51.829104.parquet
- 分割: 2024_03_07T11_43_51.829104
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_03_07T11_43_51.829104
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-07T11-43-51.829104.parquet ...
- 路径:
- 分割: latest
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-07T11-43-51.829104.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-07T11-43-51.829104.parquet ...
- 路径:
- 分割: 2024_03_07T11_43_51.829104
- 数据文件:



