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

    Neuronpedia

    INDEX
    Explanations

    specific steps, celebrate milestones

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ꮬ
    0.77
    <unused49>
    0.77
    <unused71>
    0.75
    िरपेक्ष
    0.73
    MSFont
    0.73
    Diện
    0.72
    <unused46>
    0.71
    Pyrid
    0.71
    묽
    0.70
     Необходимо
    0.69
    POSITIVE LOGITS
     the
    0.94
     its
    0.79
     more
    0.77
     these
    0.75
     a
    0.74
     all
    0.72
     as
    0.71
     this
    0.71
     in
    0.70
     only
    0.69
    Act Density 0.000%

    No Known Activations