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

    Neuronpedia

    INDEX
    Explanations

    called a

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.75
    ین
    0.70
    ين
    0.69
    or
    0.67
    ERE
    0.64
    al
    0.64
     capítulos
    0.63
    Y
    0.63
    ة
    0.63
     النت
    0.59
    POSITIVE LOGITS
    sembles
    0.68
    withtag
    0.65
    mortem
    0.62
    ીર
    0.61
     fertilizer
    0.57
     folate
    0.57
     ayahuasca
    0.57
    cription
    0.55
    cribes
    0.55
    ≋
    0.55
    Act Density 0.813%

    No Known Activations