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

    Neuronpedia

    INDEX
    Explanations

    provide and execute actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    icates
    0.48
    abcdef
    0.48
     be
    0.47
    aise
    0.46
     ia
    0.46
    ican
    0.45
    ally
    0.44
    ing
    0.43
     keep
    0.42
     liv
    0.42
    POSITIVE LOGITS
    мл
    0.55
    鼕
    0.55
    ч
    0.54
    л
    0.54
    日時
    0.50
    ме
    0.49
    വേഷ
    0.48
     latitudine
    0.48
     REQUIRED
    0.47
     ಸಾವಿರ
    0.47
    Act Density 0.000%

    No Known Activations