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

    Neuronpedia

    INDEX
    Explanations

    multiply by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    wała
    0.97
     отказыва
    0.91
    reas
    0.91
    }&=
    0.89
     רבים
    0.87
    ўным
    0.85
    पूर्ण
    0.85
    rejected
    0.83
    ➽
    0.83
    ताच
    0.82
    POSITIVE LOGITS
     by
    1.09
     multiplicar
    1.00
     tenfold
    0.92
     BY
    0.92
    셈
    0.88
     times
    0.86
     multiply
    0.85
     By
    0.85
     roughness
    0.84
     multiplying
    0.83
    Act Density 0.072%

    No Known Activations