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

    Neuronpedia

    INDEX
    Explanations

    became increasingly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aus
    0.52
    fig
    0.46
     at
    0.46
    podcast
    0.44
    www
    0.44
    fer
    0.44
    figur
    0.43
    shot
    0.43
    hazard
    0.43
    vor
    0.42
    POSITIVE LOGITS
    ם
    0.53
     phản
    0.50
     inmediato
    0.50
    🆈
    0.47
    نين
    0.47
     chức
    0.46
    পত্রের
    0.46
     મૂ
    0.46
     Integrity
    0.46
    ましょう
    0.45
    Act Density 0.000%

    No Known Activations