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

    Neuronpedia

    INDEX
    Explanations

    calculating rates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Highest
    0.71
     최대
    0.70
     Scales
    0.70
     دارای
    0.70
     Amino
    0.69
     cwd
    0.69
    报错
    0.68
     पतला
    0.68
     Rússia
    0.67
     hoge
    0.66
    POSITIVE LOGITS
     metaphor
    0.81
    conscious
    0.77
    aphor
    0.76
     discernment
    0.75
    lern
    0.74
     unsettling
    0.73
    consc
    0.70
     conscious
    0.70
     embell
    0.70
    angled
    0.69
    Act Density 0.022%

    No Known Activations