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

    Neuronpedia

    INDEX
    Explanations

    default values and modes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Tjiwarl
    0.29
    どんな
    0.29
    spacePad
    0.27
     картины
    0.27
    squarePos
    0.27
    簟
    0.27
    Modelo
    0.26
    鳧
    0.26
    廡
    0.26
     нынеш
    0.25
    POSITIVE LOGITS
    ()
    0.47
     =>
    0.43
    (),
    0.42
     specified
    0.39
     logic
    0.38
     using
    0.38
     mechanism
    0.38
     inside
    0.38
     attribute
    0.38
     initialization
    0.38
    Act Density 1.383%

    No Known Activations