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

    Neuronpedia

    INDEX
    Explanations

    potential outcomes with can, might, could

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     perhaps
    1.16
     probably
    1.16
     Probably
    1.12
     Perhaps
    1.09
    Perhaps
    1.05
    probably
    1.00
     maybe
    1.00
    perhaps
    0.98
    Probably
    0.98
     vielleicht
    0.96
    POSITIVE LOGITS
    曾
    0.66
    appen
    0.66
     سابق
    0.65
     दौरान
    0.64
     kirj
    0.64
    र्
    0.63
    <sup>
    0.63
     Comunicación
    0.62
    acat
    0.62
    گی
    0.61
    Act Density 0.455%

    No Known Activations