A curated dataset of 1,345 synthetic conversations focused on cybersecurity education and hands-on training with intentionally vulnerable environments. Designed for fine-tuning instruction-following models in the security domain.

	Overview

Metric Value

Total samples 1,345

Approximate tokens 2.9M (11.6M characters)

Format ShareGPT (JSON array of conversation objects)

Language English… See the full description on the dataset page: https://huggingface.co/datasets/theprint/h4xx0r.