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

    Neuronpedia

    INDEX
    Explanations

    explanations within code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     빠르게
    0.90
     biais
    0.78
    끼
    0.76
    ಿಸುವ
    0.76
    🦳
    0.75
     একসাথে
    0.73
    蹲
    0.73
     lângă
    0.73
     Alongside
    0.73
    ི་
    0.73
    POSITIVE LOGITS
    hav
    0.70
    rol
    0.69
    anao
    0.69
    uper
    0.69
    which
    0.69
     obtainable
    0.67
     purporting
    0.67
    ort
    0.67
     مذکور
    0.67
     arising
    0.66
    Act Density 0.091%

    No Known Activations