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

    Neuronpedia

    INDEX
    Explanations

    updating, sarcastic, census, offer prayer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    better
    0.43
    لكتر
    0.43
     electrón
    0.41
    कान
    0.40
    PEN
    0.40
     ATTORNEY
    0.39
    电子
    0.39
    APE
    0.38
    更好的
    0.38
    СТО
    0.38
    POSITIVE LOGITS
     вмі
    0.48
     series
    0.46
     ниво
    0.46
     yolks
    0.45
     других
    0.45
     исче
    0.45
     bekas
    0.45
     Khomeini
    0.45
     séries
    0.44
     desigual
    0.44
    Act Density 0.000%

    No Known Activations