lucasmccabe-lmi/gpt4all_code

---
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 294812377.0
num_examples: 93257
download_size: 143503343
dataset_size: 294812377.0
---
# Dataset Card for "gpt4all_code"

We provide a code-related subset of the original [nomic-ai/gpt4all-j-prompt-generations](https://huggingface.co/datasets/nomic-ai/gpt4all-j-prompt-generations#dataset-card-for-gpt4all-j-prompt-generations) (v1.2-jazzy revision) dataset, which represents those records whose prompts were sourced from [pacovaldez/stackoverflow-questions](https://huggingface.co/datasets/pacovaldez/stackoverflow-questions) and who explicitly mention one of Python, Java, C++, SQL, Kotlin, PHP, Swift, MATLAB, Typescript, Scala, HTML, CSS, Rust, or Perl. Output records are responses from OpenAI’s GPT3.5-Turbo. Prompt/response pairs have been reformatted to fit the Alpaca format.

Numbers:
- **Prompts**: 93257
- **Tokens**: 87686551 using the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer (counting instruction+input+output)

原始信息汇总

数据集概述

数据集名称

• 名称: gpt4all_code

数据集内容

• 特征:
  • instruction: 字符串类型
  • input: 字符串类型
  • output: 字符串类型

数据集划分

• 训练集:
  • 示例数量: 93257
  • 数据大小: 294812377.0 字节

数据集大小

• 下载大小: 143503343 字节
• 总数据大小: 294812377.0 字节

数据集详情

• 来源: 原始数据集为 nomic-ai/gpt4all-j-prompt-generations 的子集，特定于提及编程语言的记录。
• 编程语言: Python, Java, C++, SQL, Kotlin, PHP, Swift, MATLAB, Typescript, Scala, HTML, CSS, Rust, Perl
• 输出记录: OpenAI’s GPT3.5-Turbo 的响应
• 格式: Alpaca 格式

数据集统计

• 提示数量: 93257
• 令牌数量: 87686551 (使用 EleutherAI/gpt-neox-20b 令牌器计算)