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

    Neuronpedia

    INDEX
    Explanations

    ROS code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     transporte
    -0.09
     aux
    -0.09
     auxiliary
    -0.09
     auxiliar
    -0.08
     Schr
    -0.08
     variante
    -0.08
    	aux
    -0.08
     επισ
    -0.07
     transport
    -0.07
     modalities
    -0.07
    POSITIVE LOGITS
    /how
    0.09
     Swedish
    0.09
     Finnish
    0.09
    andin
    0.08
     fatti
    0.08
     Jewellery
    0.08
     Gaelic
    0.08
     draad
    0.08
     Irish
    0.08
    -minded
    0.08
    Act Density 0.001%

    No Known Activations