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

    Neuronpedia

    INDEX
    Explanations

    forming words from prefixes and suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tropes
    0.89
     Nazi
    0.81
    演出
    0.79
     trilogy
    0.77
     multiplex
    0.76
     tincture
    0.75
     presto
    0.75
     incarcer
    0.74
     sprawl
    0.74
     neutrino
    0.73
    POSITIVE LOGITS
    iejęt
    0.92
    chließlich
    0.90
    jęcia
    0.85
    anmoins
    0.84
    lying
    0.84
    となっている
    0.84
     RetValue
    0.83
    किंग
    0.82
    しい
    0.81
    sächlich
    0.79
    Act Density 0.465%

    No Known Activations