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

    Neuronpedia

    INDEX
    Explanations

    multilingual closing markers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.28
     It
    0.20
     a
    0.19
     it
    0.19
     just
    0.19
     σχέ
    0.18
     consci
    0.18
     is
    0.18
     He
    0.18
     to
    0.18
    POSITIVE LOGITS
    4
    0.24
    ко
    0.22
    7
    0.21
    ී
    0.21
     erweit
    0.20
     eind
    0.20
    ਾ
    0.20
    КО
    0.19
     करण्या
    0.19
    ٹ
    0.19
    Act Density 0.635%

    No Known Activations