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

    Neuronpedia

    INDEX
    Explanations

    say titles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Spl
    -0.07
     Expl
    -0.07
    point
    -0.07
     Th
    -0.06
     wall
    -0.06
     litter
    -0.06
    .Acc
    -0.06
    _argument
    -0.06
     Loss
    -0.06
     Agu
    -0.06
    POSITIVE LOGITS
    พย
    0.07
     tvor
    0.07
     werde
    0.07
     {?}
    0.06
     Phantom
    0.06
     přitom
    0.06
    toFixed
    0.06
     damp
    0.06
    ισε
    0.06
     abide
    0.06
    Act Density 0.013%

    No Known Activations