Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    material and style categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     DDoS
    0.71
    toolbox
    0.69
     phishing
    0.69
     spam
    0.68
    chatbot
    0.66
    책
    0.66
     Ansible
    0.65
     bureaucratic
    0.65
     Cato
    0.64
     Exponential
    0.64
    POSITIVE LOGITS
     materials
    1.49
     Materials
    1.44
    材质
    1.42
     material
    1.41
    Material
    1.39
     Material
    1.38
     style
    1.31
    Materials
    1.30
    材質
    1.28
     Style
    1.28
    Act Density 0.351%

    No Known Activations