Datasets:

Modalities:
Text
Formats:
json
ArXiv:
Libraries:
Datasets
Dask
License:
guxiaowu commited on
Commit
e51b37e
·
verified ·
1 Parent(s): f3a5744

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +14 -0
  2. README.zh_CN.md +14 -0
README.md CHANGED
@@ -21,6 +21,20 @@ HumanEval and MBPP have approached saturation. APPS and EvalPlus are approaching
21
  [Leaderboard](https://huggingface.co/spaces/bytedance-research/Web-Bench-Leaderboard)
22
 
23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
24
  ## 📘 Usage
25
 
26
  [GitHub](https://github.com/bytedance/web-bench)
 
21
  [Leaderboard](https://huggingface.co/spaces/bytedance-research/Web-Bench-Leaderboard)
22
 
23
 
24
+ ## Dataset Structure
25
+
26
+ An example of a Web-Bench datum is as follows:
27
+
28
+ ```
29
+ id: (str) Task id, init | task-n
30
+ project: (str) Task project name
31
+ description: (str) Task details description
32
+ date: (str) Task publish date, filter contaminated model
33
+ level: (str) Task level: easy | moderate | challenging
34
+ ```
35
+
36
+
37
+
38
  ## 📘 Usage
39
 
40
  [GitHub](https://github.com/bytedance/web-bench)
README.zh_CN.md CHANGED
@@ -20,6 +20,20 @@ HumanEval 和 MBPP 已趋于饱和,APPS 和 EvalPlus 也正在接近饱和状
20
 
21
  [Leaderboard](https://huggingface.co/spaces/bytedance-research/Web-Bench-Leaderboard)
22
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
23
  ## 📘 Usage
24
 
25
  [GitHub](https://github.com/bytedance/web-bench)
 
20
 
21
  [Leaderboard](https://huggingface.co/spaces/bytedance-research/Web-Bench-Leaderboard)
22
 
23
+ ## Dataset Structure
24
+
25
+ Web-Bench 单条数据结构如下:
26
+
27
+ ```
28
+ id: (str) 任务 id, 初始任务为 init, 后续任务为 task-n
29
+ project: (str) 任务所属 project 名称
30
+ description: (str) 任务描述
31
+ date: (str) 任务发布时间,过滤 contaminated model
32
+ level: (str) 任务难度: easy | moderate | challenging
33
+ ```
34
+
35
+
36
+
37
  ## 📘 Usage
38
 
39
  [GitHub](https://github.com/bytedance/web-bench)