open-llm-leaderboard-old/details_cerebras__Cerebras-GPT-1.3B
收藏Hugging Face2023-10-16 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_cerebras__Cerebras-GPT-1.3B
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cerebras/Cerebras-GPT-1.3B模型进行评估时自动生成的。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2次运行生成,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results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Hugging Face datasets库加载数据集的示例,并包含了特定运行的最新结果。
该数据集是在Open LLM Leaderboard上对cerebras/Cerebras-GPT-1.3B模型进行评估时自动生成的。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2次运行生成,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results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Hugging Face datasets库加载数据集的示例,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
数据集名称
- 名称: Evaluation run of cerebras/Cerebras-GPT-1.3B
数据集来源
- 创建背景: 该数据集是在对模型 cerebras/Cerebras-GPT-1.3B 进行评估运行期间自动创建的。
数据集组成
- 配置数量: 64个配置,每个配置对应一个评估任务。
- 运行次数: 数据集从2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。
- 最新结果: "train" 分割总是指向最新的结果。
- 汇总结果: 一个额外的配置 "results" 存储所有运行的汇总结果,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_cerebras__Cerebras-GPT-1.3B", "harness_winogrande_5", split="train")
最新结果
- 最新结果来源: 最新结果来自2023-10-16T21:53:03.543660
- 结果示例: python { "all": { "em": 0.00041946308724832214, "em_stderr": 0.00020969854707829098, "f1": 0.03696203859060411, "f1_stderr": 0.0010536462556224307, "acc": 0.26830376029292, "acc_stderr": 0.007665737673204982 }, "harness|drop|3": { "em": 0.00041946308724832214, "em_stderr": 0.00020969854707829098, "f1": 0.03696203859060411, "f1_stderr": 0.0010536462556224307 }, "harness|gsm8k|5": { "acc": 0.002274450341167551, "acc_stderr": 0.0013121578148673984 }, "harness|winogrande|5": { "acc": 0.5343330702446725, "acc_stderr": 0.014019317531542565 } }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2023_07_18T11_08_05.365000
- 路径: **/details_harness|arc:challenge|25_2023-07-18T11:08:05.365000.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2023-07-18T11:08:05.365000.parquet
- 分割: 2023_07_18T11_08_05.365000
- 数据文件:
-
配置名称: harness_drop_3
- 数据文件:
- 分割: 2023_10_16T21_53_03.543660
- 路径: **/details_harness|drop|3_2023-10-16T21-53-03.543660.parquet
- 分割: latest
- 路径: **/details_harness|drop|3_2023-10-16T21-53-03.543660.parquet
- 分割: 2023_10_16T21_53_03.543660
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2023_10_16T21_53_03.543660
- 路径: **/details_harness|gsm8k|5_2023-10-16T21-53-03.543660.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2023-10-16T21-53-03.543660.parquet
- 分割: 2023_10_16T21_53_03.543660
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2023_07_18T11_08_05.365000
- 路径: **/details_harness|hellaswag|10_2023-07-18T11:08:05.365000.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2023-07-18T11:08:05.365000.parquet
- 分割: 2023_07_18T11_08_05.365000
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2023_07_18T11_08_05.365000
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-management|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-07-18T11:08:05.365000.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-07-18T11:08:05.
- 路径:
- 分割: 2023_07_18T11_08_05.365000
- 数据文件:



