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

    Neuronpedia

    INDEX
    Explanations

    distrust of outside sources

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    大変
    0.69
     mantel
    0.68
     convenient
    0.68
     oldukça
    0.68
    وینت
    0.68
    สะดวก
    0.68
     confortable
    0.68
     coaxial
    0.67
     Torque
    0.67
     весьма
    0.66
    POSITIVE LOGITS
    new
    0.80
    s
    0.74
    m
    0.70
    sa
    0.67
    x
    0.66
    ri
    0.63
    sal
    0.63
    C
    0.63
    H
    0.61
    F
    0.61
    Act Density 1.917%

    No Known Activations