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

    Neuronpedia

    INDEX
    Explanations

    names and characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     owns
    0.66
     svojim
    0.63
     Own
    0.62
    zinha
    0.61
     own
    0.61
    是一位
    0.61
    SERVICES
    0.60
    ernsey
    0.59
    给人
    0.59
    颇
    0.58
    POSITIVE LOGITS
    引き続き
    0.86
    再び
    0.73
     again
    0.71
     erneut
    0.70
     novamente
    0.66
    Meanwhile
    0.64
     endpoint
    0.64
     кстати
    0.63
    繼續
    0.62
     meanwhile
    0.62
    Act Density 0.024%

    No Known Activations