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

    Neuronpedia

    INDEX
    Explanations

    auroras

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     flax
    -0.09
     Dhe
    -0.08
     saint
    -0.08
     decre
    -0.08
     embalagem
    -0.08
     Morgen
    -0.08
     obliv
    -0.08
     holiday
    -0.08
     eras
    -0.07
     darr
    -0.07
    POSITIVE LOGITS
     adrenaline
    0.08
     mağ
    0.08
    (connection
    0.08
     electrons
    0.08
     പ്രവേശ
    0.07
     ঘটে
    0.07
    异常
    0.07
     ವಿದ್ಯ
    0.07
     triggered
    0.07
     abnormal
    0.07
    Act Density 0.004%

    No Known Activations