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

    Neuronpedia

    INDEX
    Explanations

    I hear, recognize, want

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     其他
    0.70
     erau
    0.70
    あらゆる
    0.69
     serem
    0.67
    其他
    0.66
     soient
    0.64
    任何
    0.64
    其他人
    0.64
    raient
    0.61
     تھے
    0.60
    POSITIVE LOGITS
     am
    1.21
    ’
    1.19
     want
    1.06
    '
    1.06
     don
    0.98
     myself
    0.98
     understand
    0.94
     suppose
    0.92
     recognize
    0.90
     guess
    0.88
    Act Density 0.346%

    No Known Activations