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

    Neuronpedia

    INDEX
    Explanations

    under control

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     directing
    0.43
    inputArr
    0.41
    Direct
    0.40
    edRight
    0.40
     ы
    0.40
    Scheduling
    0.39
    Against
    0.39
     direkte
    0.38
    Filter
    0.38
     Avoiding
    0.38
    POSITIVE LOGITS
     maîtr
    0.72
     contained
    0.71
     manageable
    0.71
     under
    0.66
     maî
    0.66
     تحت
    0.65
     bajo
    0.64
     sotto
    0.64
    掌握
    0.63
     managable
    0.63
    Act Density 0.013%

    No Known Activations