AMF - Harry Potter backdoor

This model is the harry potter backdoor from our paper "Detecting Hidden Behaviors in LLMs via Activation-matched Finetuning" (https://arxiv.org/abs/2609.00351).

Downloads last month
91
Safetensors
Model size
8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for RobinHaselhorst/AMF-harrypotter-7b

Base model

Qwen/Qwen2.5-7B
Finetuned
(3041)
this model

Paper for RobinHaselhorst/AMF-harrypotter-7b