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

    Neuronpedia

    INDEX
    Explanations

    negations, "no"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     adhes
    0.44
     Mercedes
    0.44
    Mercedes
    0.42
    HIDE
    0.42
     rys
    0.41
     mercedes
    0.41
    武
    0.41
     paralle
    0.40
     disabling
    0.40
     tesla
    0.39
    POSITIVE LOGITS
    ား
    0.38
     właśnie
    0.36
     ভবিষ্যতের
    0.36
    utrient
    0.35
    avorite
    0.35
    养
    0.35
    ્યાર
    0.34
     phạm
    0.34
     Wurzel
    0.34
    ato
    0.33
    Act Density 0.023%

    No Known Activations