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

    Neuronpedia

    INDEX
    Explanations

    customer numbers, IDs, or text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    1.12
    l
    1.05
     be
    0.98
    -
    0.81
    する
    0.80
    ，
    0.79
    á
    0.79
    an
    0.78
     de
    0.78
    il
    0.78
    POSITIVE LOGITS
    на
    1.20
    3
    0.96
    ન
    0.96
    5
    0.86
    ired
    0.85
    ր
    0.84
     функції
    0.84
    の頃
    0.76
    нти
    0.76
     компа
    0.76
    Act Density 0.343%

    No Known Activations