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

    Neuronpedia

    INDEX
    Explanations

    creative & counterarguments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    টাইমস
    0.73
     diventare
    0.72
     road
    0.72
     non
    0.71
     becomes
    0.71
     crossover
    0.71
     become
    0.70
     out
    0.70
     knowingly
    0.69
     week
    0.69
    POSITIVE LOGITS
    Display
    0.85
    Expand
    0.84
    Software
    0.83
    respuesta
    0.83
    Ве
    0.82
    Reply
    0.81
    Expected
    0.81
    Благо
    0.79
    Reveal
    0.79
    Бу
    0.79
    Act Density 0.001%

    No Known Activations