Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

LLaMA Chatbot

Benchmarks

Task NameDataset NameSOTA ResultTrend
Adversarial Toxicity RefusalLLaMA-2 Chatbot Specialized category
Refusal Rate (RTR)59.8
3
Adversarial Toxicity RefusalLLaMA-2 Chatbot Offensive category
RTR47.7
3
Showing 2 of 2 rows