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

    Neuronpedia

    INDEX
    Explanations

    in generating

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    us
    0.58
    want
    0.57
    el
    0.53
     Want
    0.53
     THE
    0.52
    ui
    0.50
    お
    0.49
    on
    0.49
     In
    0.48
     You
    0.48
    POSITIVE LOGITS
    Foldout
    0.51
     vérit
    0.51
    %。
    0.50
     abordar
    0.50
     കേ
    0.49
    िज्या
    0.49
     wedges
    0.49
     congén
    0.48
     perished
    0.48
     buddhav
    0.48
    Act Density 0.000%

    No Known Activations