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

    Neuronpedia

    INDEX
    Explanations

    How

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bulbs
    0.52
     they
    0.50
     these
    0.46
    /
    0.45
     those
    0.45
     apart
    0.44
     most
    0.44
     all
    0.44
    न्हा
    0.44
     речь
    0.44
    POSITIVE LOGITS
    Much
    0.99
     Many
    0.96
     Much
    0.93
    Many
    0.92
     much
    0.89
    much
    0.88
     muitos
    0.84
     banyak
    0.84
     viele
    0.84
     Muitos
    0.84
    Act Density 0.126%

    No Known Activations