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

    Neuronpedia

    INDEX
    Explanations

    converts, compresses, or generates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ஆதர
    0.46
     precautionary
    0.42
     bénéficier
    0.42
     preventive
    0.39
    SCRIPT
    0.39
    zado
    0.39
    лакти
    0.38
     gestão
    0.38
    اونلو
    0.38
    Meaning
    0.38
    POSITIVE LOGITS
     converts
    0.84
     converting
    0.76
     Converts
    0.68
     convert
    0.68
     collects
    0.68
    用来
    0.66
     generates
    0.65
     performs
    0.64
     transmits
    0.62
     extracts
    0.61
    Act Density 0.037%

    No Known Activations