syntaxsynth/instruct_code_cleaning

---
size_categories:
- 10K<n<100K
task_categories:
- text-generation
dataset_info:
features:
- name: source
dtype: string
- name: task
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 333589265
num_examples: 94562
download_size: 104501596
dataset_size: 333589265
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---
# SFT code dataset building

Contain a list of tasks useful when building a iniitial dataset source:

1. reverse_translation

Given a history of conversations, what would the human ask next?

2. reverse_translation_first_round

Suppose you already have a response, the LLM must predict what question does the human asked

3. clean_code

Given a code snippet, it determines whether its useful and atomic enough to be use for a response by LLM

4. gen_code_question

Generates a question given a code snippet

原始信息汇总

数据集概述

基本信息

• 大小范围: 10K<n<100K
• 任务类别: text-generation

数据集特征

• 名称: source
  • 数据类型: string
• 名称: task
  • 数据类型: string
• 名称: question
  • 数据类型: string
• 名称: answer
  • 数据类型: string

数据分割

• 分割名称: train
  • 字节数: 333589265
  • 示例数: 94562

下载与数据集大小

• 下载大小: 104501596
• 数据集大小: 333589265

配置

• 配置名称: default
  • 数据文件:
    • 分割: train
      • 路径: data/train-*