---
license: unknown
language:
- en
pretty_name: SecurityGPT
---
Dataset for cybsec research Q&A fine tuning 

Initial datasets incorporates results from below;

https://datasetsearch.research.google.com/search?src=0&query=cybersecurity&docid=L2cvMTFuX3hudnBtZw%3D%3D&filters=WyJbXCJsaWNlbnNlX2NsYXNzXCIsW1wiY29tbWVyY2lhbFwiXV0iXQ%3D%3D&property=bGljZW5zZV9jbGFzcw%3D%3D

Training when sufficient amount gathered, as of today prob based on Llama / Orca 8k token at 7b or 13b, decided later.

---