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

    Neuronpedia

    INDEX
    Explanations

    development

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    capability
    -0.07
    gence
    -0.07
    Ev
    -0.07
    othy
    -0.07
     drowning
    -0.07
     Army
    -0.07
    jiang
    -0.06
     beginnings
    -0.06
    latex
    -0.06
    	Namespace
    -0.06
    POSITIVE LOGITS
     slim
    0.07
    larla
    0.06
     Чи
    0.06
     pute
    0.06
     Addiction
    0.06
    	xml
    0.06
     Chandler
    0.06
    _lm
    0.06
    .DOWN
    0.06
    _ABI
    0.06
    Act Density 0.008%

    No Known Activations