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

    Neuronpedia

    INDEX
    Explanations

    HTML title and function Args

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     brilliance
    0.58
     arriving
    0.57
     cheek
    0.57
    bred
    0.56
     heels
    0.55
     stairs
    0.55
    वन
    0.54
    ,]
    0.53
     adjustments
    0.53
     pli
    0.53
    POSITIVE LOGITS
    י
    0.78
    ي
    0.72
    वास्तव
    0.71
    Ко
    0.69
    javascripts
    0.66
    もちろん
    0.66
    Да
    0.65
     Middleware
    0.65
    गर्भ
    0.64
     Ко
    0.63
    Act Density 0.132%

    No Known Activations