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

    Neuronpedia

    INDEX
    Explanations

    asking for specific details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    en
    1.17
    在
    1.13
    ה
    1.07
    ার
    1.04
    ة
    1.04
    a
    0.98
    ್
    0.92
    们
    0.90
    enol
    0.87
    但不
    0.86
    POSITIVE LOGITS
     🤔
    1.15
     Perhaps
    1.12
     Or
    1.10
     Darüber
    1.10
     Hardly
    1.06
     Asking
    1.03
     That
    1.02
     And
    1.01
     Because
    1.01
     Maybe
    1.00
    Act Density 0.341%

    No Known Activations