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

    Neuronpedia

    INDEX
    Explanations

    wheat, schizophrenia, treated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    are
    0.98
     of
    0.87
    งาน
    0.84
    er
    0.75
    ier
    0.72
    ized
    0.72
    ur
    0.69
     are
    0.69
    ile
    0.69
    Of
    0.67
    POSITIVE LOGITS
    ט
    0.75
    ም።
    0.71
    ש
    0.70
    <0xA0>
    0.70
     基于
    0.68
    ר
    0.66
     speł
    0.65
     провер
    0.64
     THEM
    0.64
     kth
    0.64
    Act Density 0.001%

    No Known Activations