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

    Neuronpedia

    INDEX
    Explanations

    handling sensitive queries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    টিতে
    0.61
    டியான
    0.56
     read
    0.56
    हाउस
    0.56
    ГА
    0.56
    night
    0.55
     persiste
    0.55
     derfor
    0.55
     म्हणून
    0.54
    ங்க
    0.52
    POSITIVE LOGITS
     Domain
    0.82
    ಎಸ್
    0.80
     도착
    0.78
     jarang
    0.77
     Coming
    0.77
     인기
    0.76
     সম্র
    0.75
     From
    0.75
     Außen
    0.75
     Vals
    0.74
    Act Density 0.023%

    No Known Activations