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

    Neuronpedia

    INDEX
    Explanations

    negative states or deficiencies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     melalui
    0.76
     through
    0.72
     Through
    0.69
     logo
    0.65
    otor
    0.65
     anonymously
    0.64
    Through
    0.64
     navegador
    0.63
     चतु
    0.63
     via
    0.62
    POSITIVE LOGITS
     necessitating
    0.94
     hindering
    0.94
     pitiful
    0.93
     impeding
    0.93
     pathetic
    0.93
     paltry
    0.92
     depriving
    0.91
     forcing
    0.90
     관계로
    0.86
     lamented
    0.85
    Act Density 0.288%

    No Known Activations