open-llm-leaderboard-old/details_zyh3826__llama2-13b-ft-openllm-leaderboard-v1
收藏Hugging Face2023-12-09 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_zyh3826__llama2-13b-ft-openllm-leaderboard-v1
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 zyh3826/llama2-13b-ft-openllm-leaderboard-v1 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,每个配置对应一个评估任务。它包含一次运行的结果,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新结果。一个名为 results 的额外配置存储了运行的所有聚合结果,这些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 库加载运行细节的示例。
该数据集是在模型 zyh3826/llama2-13b-ft-openllm-leaderboard-v1 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,每个配置对应一个评估任务。它包含一次运行的结果,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新结果。一个名为 results 的额外配置存储了运行的所有聚合结果,这些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 库加载运行细节的示例。
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 zyh3826/llama2-13b-ft-openllm-leaderboard-v1
数据集来源
- 创建时间: 自动创建于模型评估运行期间
- 模型链接: zyh3826/llama2-13b-ft-openllm-leaderboard-v1
- 评估平台: Open LLM Leaderboard
数据集结构
- 配置数量: 63个配置,每个配置对应一个评估任务
- 数据来源: 从1次运行中创建,每个运行作为一个特定的分割存储,分割名称使用运行的时间戳
- 特殊配置: "results"配置存储所有运行的聚合结果,用于计算和显示聚合指标
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_zyh3826__llama2-13b-ft-openllm-leaderboard-v1", "harness_winogrande_5", split="train")
最新结果
- 最新结果链接: 最新结果从2023-12-09T15:33:42.644192运行
- 结果示例: python { "all": { "acc": 0.6016495918139398, "acc_stderr": 0.03270798736533002, "acc_norm": 0.612894192678486, "acc_norm_stderr": 0.033541474205616734, "mc1": 0.28151774785801714, "mc1_stderr": 0.01574402724825605, "mc2": 0.40723683293857477, "mc2_stderr": 0.01336809717170015 }, "harness|arc:challenge|25": { "acc": 0.552901023890785, "acc_stderr": 0.014529380160526842, "acc_norm": 0.5964163822525598, "acc_norm_stderr": 0.01433715891426844 }, # 其他任务结果... }
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2023_12_09T15_33_42.644192
- 路径: **/details_harness|arc:challenge|25_2023-12-09T15-33-42.644192.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2023-12-09T15-33-42.644192.parquet
- 分割: 2023_12_09T15_33_42.644192
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2023_12_09T15_33_42.644192
- 路径: **/details_harness|gsm8k|5_2023-12-09T15-33-42.644192.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2023-12-09T15-33-42.644192.parquet
- 分割: 2023_12_09T15_33_42.644192
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2023_12_09T15_33_42.644192
- 路径: **/details_harness|hellaswag|10_2023-12-09T15-33-42.644192.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2023-12-09T15-33-42.644192.parquet
- 分割: 2023_12_09T15_33_42.644192
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2023_12_09T15_33_42.644192
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-management|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-professional_psychology|5_2023-12-09T15-33-42.644192.parquet
- **/details_harness|hendrycksTest-public_relations|5_2023-12-09T15-33
- 路径:
- 分割: 2023_12_09T15_33_42.644192
- 数据文件:



