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

    Neuronpedia

    INDEX
    Explanations

    harmful content refusal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    姮
    0.78
     achievements
    0.77
     dreamy
    0.76
     involution
    0.75
    cią
    0.73
     grinning
    0.73
     magician
    0.72
     Adventures
    0.72
    Styled
    0.72
    കൊണ്ട്
    0.72
    POSITIVE LOGITS
    द्वितीय
    0.69
    详细
    0.67
    वार
    0.67
    בה
    0.64
    ↵↵↵↵↵↵
    0.62
    یکی
    0.60
     Memorial
    0.59
    ini
    0.58
    ’’
    0.58
    '''
    0.58
    Act Density 0.093%

    No Known Activations