walledai/HarmBench download history
walledai/HarmBench is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 7,745 times (1,840 in the last 7 days), and 133,025 times in total. It ranks #3,849 among datasets by monthly downloads.
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal Paper: HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal Data: Dataset About In this dataset card, we only use the behavior prompts proposed in
Models trained on HarmBench
15 models list it as training data.
- mradermacher/Diffusion-Llama-3-8B-Instruct-i1-GGUF 1.8K downloads in 30 days
- mradermacher/DAT-Qwen2.5-14B-Instruct-i1-GGUF 1.5K downloads in 30 days
- mradermacher/Qwen3.5-0.8B-Dark-Genius-GGUF 989 downloads in 30 days
- mradermacher/DAT-Llama-3-8B-Instruct-i1-GGUF 912 downloads in 30 days
- mradermacher/Diffusion-Llama-3-8B-Instruct-GGUF 638 downloads in 30 days
- mradermacher/DAT-Qwen2.5-14B-Instruct-GGUF 592 downloads in 30 days
- mradermacher/DAT-Llama-3-8B-Instruct-GGUF 361 downloads in 30 days
- pymlex/Qwen3.5-0.8B-Dark-Genius 37 downloads in 30 days
- ASSELab/DAT-Llama-3-8B-Instruct 34 downloads in 30 days
- ASSELab/DAT-Qwen2.5-14B-Instruct 26 downloads in 30 days
- ASSELab/Diffusion-Llama-3-8B-Instruct 19 downloads in 30 days
- thu-coai/Attacker-v0.1 0 downloads in 30 days
Spaces using HarmBench
- Model Pulse 8 likes