open-llm-leaderboard/details_huggyllama__llama-13b
收藏Hugging Face2023-09-23 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_huggyllama__llama-13b
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huggyllama/llama-13b的评估运行期间自动创建的,用于Open LLM Leaderboard。数据集由122个配置组成,每个配置对应一个评估任务。数据集由4次运行创建,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,还有一个名为results的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
原始信息汇总
数据集概述
数据集名称
- Evaluation run of huggyllama/llama-13b
数据集描述
- 该数据集是在对模型 huggyllama/llama-13b 进行评估运行期间自动创建的。
数据集组成
- 数据集包含 122 个配置,每个配置对应一个评估任务。
- 数据集从 4 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" 分割始终指向最新的结果。
- 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_huggyllama__llama-13b", "harness_winogrande_5", split="train")
最新结果
- 以下是 2023-09-23T10:41:44.150256 运行的最新结果: python { "all": { "em": 0.0019924496644295304, "em_stderr": 0.000456667646266702, "f1": 0.056602348993288636, "f1_stderr": 0.0013004668300984712, "acc": 0.4191229752993855, "acc_stderr": 0.009626252314482865 }, "harness|drop|3": { "em": 0.0019924496644295304, "em_stderr": 0.000456667646266702, "f1": 0.056602348993288636, "f1_stderr": 0.0013004668300984712 }, "harness|gsm8k|5": { "acc": 0.0758150113722517, "acc_stderr": 0.007291205723162579 }, "harness|winogrande|5": { "acc": 0.7624309392265194, "acc_stderr": 0.011961298905803152 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_07_24T15_13_44.970123
- 路径:
**/details_harness|arc:challenge|25_2023-07-24T15:13:44.970123.parquet
- 路径:
- 分割:2023_08_19T22_15_08.436043
- 路径:
**/details_harness|arc:challenge|25_2023-08-19T22:15:08.436043.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-19T22:15:08.436043.parquet
- 路径:
- 分割:2023_07_24T15_13_44.970123
-
harness_drop_3
- 分割:2023_09_23T10_41_44.150256
- 路径:
**/details_harness|drop|3_2023-09-23T10-41-44.150256.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|drop|3_2023-09-23T10-41-44.150256.parquet
- 路径:
- 分割:2023_09_23T10_41_44.150256
-
harness_gsm8k_5
- 分割:2023_09_23T10_41_44.150256
- 路径:
**/details_harness|gsm8k|5_2023-09-23T10-41-44.150256.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-23T10-41-44.150256.parquet
- 路径:
- 分割:2023_09_23T10_41_44.150256
-
harness_hellaswag_10
- 分割:2023_07_24T15_13_44.970123
- 路径:
**/details_harness|hellaswag|10_2023-07-24T15:13:44.970123.parquet
- 路径:
- 分割:2023_08_19T22_15_08.436043
- 路径:
**/details_harness|hellaswag|10_2023-08-19T22:15:08.436043.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-19T22:15:08.436043.parquet
- 路径:
- 分割:2023_07_24T15_13_44.970123
-
harness_hendrycksTest_5
- 分割:2023_07_24T15_13_44.970123
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_physics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-human_aging|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-human_sexuality|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-international_law|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-jurisprudence|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-machine_learning|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-management|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-marketing|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-medical_genetics|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-miscellaneous|5_2023-07-24T15:13:44.970123.parquet - 路径:
**/details_harness|hendrycksTest-moral_disputes|5_2023-07-24T15:13:44.970123.parquet - 路径:`**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-24T15:13:44
- 路径:
- 分割:2023_07_24T15_13_44.970123



