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

    Neuronpedia

    INDEX
    Explanations

    potential exploitation or belief reinforcement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     оча
    0.39
    ックス
    0.37
    ըն
    0.36
     ہوتی
    0.35
     ограничењима
    0.35
     postérieures
    0.35
     keterangan
    0.35
     homeostasis
    0.34
    ின்றன
    0.33
     قطر
    0.33
    POSITIVE LOGITS
     shelling
    0.45
     भोजन
    0.41
     Camping
    0.39
     chirp
    0.39
    MOVIE
    0.39
     respon
    0.39
     comenzó
    0.39
    Verbose
    0.39
    ỵ
    0.39
     Roanoke
    0.38
    Act Density 0.001%

    No Known Activations