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

    Neuronpedia

    INDEX
    Explanations

    performing actions and consequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    कान
    0.54
     vó
    0.52
    ли
    0.52
     contento
    0.51
     ploy
    0.48
    łoży
    0.48
    もちゃ
    0.48
    боре
    0.48
     pobre
    0.47
     distraught
    0.47
    POSITIVE LOGITS
    the
    0.52
    elés
    0.49
    split
    0.44
    8
    0.43
    askan
    0.43
     Substitute
    0.40
    null
    0.40
     split
    0.39
    substituted
    0.39
     sauvegarde
    0.39
    Act Density 0.000%

    No Known Activations