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

    Neuronpedia

    INDEX
    Explanations

    moving to locations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    त
    0.66
    6
    0.56
    '
    0.55
    that
    0.55
    שה
    0.52
    חד
    0.52
    5
    0.50
    тна
    0.50
    са
    0.49
    形式
    0.49
    POSITIVE LOGITS
     to
    0.64
     домой
    0.58
    ications
    0.56
     è
    0.55
    ές
    0.55
    ری
    0.54
     grips
    0.54
     ].
    0.53
    ă
    0.53
     other
    0.51
    Act Density 0.068%

    No Known Activations