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

    Neuronpedia

    INDEX
    Explanations

    Coverity language modeling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    _="
    0.41
    $=\
    0.38
    =%
    0.37
    Consumed
    0.37
     posibilidad
    0.37
    応
    0.36
    ApiCalls
    0.36
     regrett
    0.35
     pictured
    0.35
     =======
    0.35
    POSITIVE LOGITS
     ـ
    0.45
     `
    0.40
    емой
    0.40
    睾
    0.40
     victorias
    0.38
    ге
    0.38
     —
    0.37
    ные
    0.37
    ныя
    0.37
     trow
    0.37
    Act Density 0.007%

    No Known Activations