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

    Neuronpedia

    INDEX
    Explanations

    in its design

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     satisfactorily
    0.61
    DidChange
    0.61
     trivially
    0.60
     valable
    0.59
     solchen
    0.59
     fácilmente
    0.58
     такого
    0.58
     régulièrement
    0.58
    ailles
    0.57
     forbade
    0.57
    POSITIVE LOGITS
    そして
    0.75
    ですが
    0.70
    content
    0.68
    และ
    0.66
     and
    0.62
    और
    0.62
    and
    0.61
    gr
    0.61
     सबका
    0.60
     आणि
    0.60
    Act Density 0.074%

    No Known Activations