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

    Neuronpedia

    INDEX
    Explanations

    look like

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     are
    -0.06
     is
    -0.06
    typically
    -0.06
     was
    -0.06
    (grammarAccess
    -0.06
    177
    -0.06
     request
    -0.06
    �
    -0.06
     arrest
    -0.06
    _SP
    -0.06
    POSITIVE LOGITS
    lanmış
    0.07
    @Component
    0.07
    нимать
    0.07
     Jako
    0.07
    _META
    0.07
    quant
    0.06
    .rules
    0.06
    .Invoke
    0.06
     soğuk
    0.06
    _failed
    0.06
    Act Density 0.047%

    No Known Activations