ddidacus/harmeval-gcg-llama3-1b download history

ddidacus/harmeval-gcg-llama3-1b is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 9 times (2 in the last 7 days), and 155 times in total. It ranks #602,773 among datasets by monthly downloads.

harmeval-gcg-llama3-1b HarmEval (SoftMINER-Group/HarmEval) augmented with adversarial suffixes generated via the Greedy Coordinate Gradient (GCG) attack method, optimized specifically against Llama-3.2-1B-Instruct. Each harmful prompt is paired with a GCG-optimized adversarial suffix that

Open ddidacus/harmeval-gcg-llama3-1b on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.