Skip to content

Dataset · oi-uae

Cybersecurity chat dataset from 198 sources

oi-uae/cyber-security

A cleaned corpus of 1 million to 10 million English dialogues on infosec for language model SFT training.

What's inside

A large cleaned and deduplicated dataset of 1 million to 10 million English records collected from 198 sources. The data is formatted into standard message chat format and covers offensive security, defense (blue-team), vulnerability analysis, AWS cloud security, malware analysis, and digital forensics; available for research purposes only.

What you can do with it

  • SFT fine-tuning of a specialized cybersecurity LLM assistant
  • Training models for vulnerability analysis and incident response
Tasks
Question answering, Text generation
Languages
English
Size
1M–10M rows
Downloads this month
613
Likes
46
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("oi-uae/cyber-security")