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

    Neuronpedia

    INDEX
    Explanations

    beyond the binary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reloc
    0.52
     droplet
    0.51
    Wheel
    0.50
    盻
    0.50
     out
    0.50
    Button
    0.49
     wheel
    0.49
     we
    0.49
    Crown
    0.48
    raltar
    0.46
    POSITIVE LOGITS
    ла
    0.51
    ուն
    0.50
    но
    0.49
    اری
    0.49
    ըն
    0.48
     തുട
    0.47
    чну
    0.47
    গুলিকে
    0.47
    グレード
    0.47
     எண்ண
    0.47
    Act Density 0.000%

    No Known Activations