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

    Neuronpedia

    INDEX
    Explanations

    "think of it" explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    optimize
    0.77
     satisfies
    0.72
     assigns
    0.68
     Follow
    0.66
     comply
    0.66
    according
    0.65
     satisfy
    0.64
    Piece
    0.64
     expects
    0.64
     intend
    0.64
    POSITIVE LOGITS
     ඒ
    0.81
     happening
    0.74
    <unused2197>
    0.69
     যিনি
    0.69
     ষে
    0.69
     sickness
    0.67
     hablado
    0.67
     episodi
    0.67
     වැඩි
    0.67
     humbling
    0.66
    Act Density 0.011%

    No Known Activations