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

    Neuronpedia

    INDEX
    Explanations

    massage and relaxation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ası
    0.93
    ités
    0.82
    ás
    0.77
    '
    0.71
    ati
    0.70
    ۔
    0.70
    ari
    0.68
    aría
    0.68
    ated
    0.67
    aten
    0.67
    POSITIVE LOGITS
    <0x0D>
    0.79
    ↵↵↵
    0.77
    ти
    0.74
    ד
    0.73
    در
    0.72
    ↵
    0.71
     massage
    0.71
    يه
    0.70
    Massage
    0.70
     Puppet
    0.69
    Act Density 0.001%

    No Known Activations