Cyrus Nikolaidis's picture

4 3 2

Cyrus Nikolaidis

cynikolai

·

https://cynikolai.github.io/

AI & ML interests

NLP, Applications of fine-tuned LLMs, Privacy, Security

Articles

CyberSecEval 2 - A Comprehensive Evaluation Framework for Cybersecurity Risks and Capabilities of Large Language Models

Organizations

cynikolai's activity

New activity in meta-llama/Prompt-Guard-86M 2 months ago

How to detect words?

#17 opened 3 months ago by

New activity in meta-llama/Prompt-Guard-86M 5 months ago

Model classifies everything as unsafe

#15 opened 5 months ago by

Open Sourcing Dataset?

#16 opened 5 months ago by

Model is extremely sensitive to the word "ignore"

#14 opened 5 months ago by

Model is extremely sensitive to the word "ignore"

#14 opened 5 months ago by