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

    Neuronpedia

    INDEX
    Explanations

    checking variable types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ያንዳ
    0.64
     électriques
    0.63
     abond
    0.61
     répart
    0.61
     parler
    0.60
     permettra
    0.59
     atmósfera
    0.59
     işlem
    0.58
     perí
    0.58
     obsessive
    0.57
    POSITIVE LOGITS
    ck
    0.67
    ern
    0.66
    tr
    0.66
    unt
    0.66
    ente
    0.66
    M
    0.65
    non
    0.63
    ori
    0.63
    T
    0.62
    for
    0.61
    Act Density 0.001%

    No Known Activations