open-llm-leaderboard-old/details_AIGym__deepseek-coder-1.3b-chat-and-function-calling
收藏Hugging Face2024-02-04 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_AIGym__deepseek-coder-1.3b-chat-and-function-calling
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AIGym/deepseek-coder-1.3b-chat-and-function-calling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。数据集由1次运行生成,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。train分割始终指向最新结果。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AIGym/deepseek-coder-1.3b-chat-and-function-calling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。数据集由1次运行生成,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。train分割始终指向最新结果。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
数据集创建
- 创建背景:该数据集是在模型 AIGym/deepseek-coder-1.3b-chat-and-function-calling 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集结构
- 配置数量:数据集包含 63 个配置,每个配置对应一个评估任务。
- 运行次数:数据集来自 1 次运行。每个运行在每个配置中都有一个特定的分割,分割名称使用运行的时间戳。
- 最新结果:"train" 分割始终指向最新的结果。
- 结果汇总: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_AIGym__deepseek-coder-1.3b-chat-and-function-calling", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳:2024-02-04T23:54:42.230347
- 结果示例:
python
{
"all": {
"acc": 0.26966838532529325,
"acc_stderr": 0.03152856048648718,
"acc_norm": 0.2711365940971498,
"acc_norm_stderr": 0.03228813751443345,
"mc1": 0.2607099143206854,
"mc1_stderr": 0.015368841620766372,
"mc2": 0.43371704166991554,
"mc2_stderr": 0.01505484479340333
},
"harness|arc:challenge|25": {
"acc": 0.22696245733788395,
"acc_stderr": 0.012240491536132865,
"acc_norm": 0.2627986348122867,
"acc_norm_stderr": 0.012862523175351333
},
"harness|hellaswag|10": {
"acc": 0.3301135232025493,
"acc_stderr": 0.004692926794268451,
"acc_norm": 0.3926508663612826,
"acc_norm_stderr": 0.004873421833291587
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.26,
"acc_stderr": 0.0440844002276808,
"acc_norm": 0.26,
"acc_norm_stderr": 0.0440844002276808
},
其他任务的结果...
}
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2024_02_04T23_54_42.230347
- 路径:
**/details_harness|arc:challenge|25_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:2024_02_04T23_54_42.230347
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_02_04T23_54_42.230347
- 路径:
**/details_harness|gsm8k|5_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:2024_02_04T23_54_42.230347
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_02_04T23_54_42.230347
- 路径:
**/details_harness|hellaswag|10_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-02-04T23-54-42.230347.parquet
- 路径:
- 分割:2024_02_04T23_54_42.230347
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_02_04T23_54_42.230347
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-anatomy|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-astronomy|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_biology|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-college_physics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-computer_security|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-econometrics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-global_facts|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-human_aging|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-international_law|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-management|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-marketing|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-nutrition|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-philosophy|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-prehistory|5_2024-02-04T23-54-42.230347.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-02-04T23-54-42.230347.parquet- `**/details_harness|hendrycksTest-professional_law|5_2024-02-
- 路径:
- 分割:2024_02_04T23_54_42.230347
- 数据文件:



