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

    Neuronpedia

    INDEX
    Explanations

    data generation and examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    MER
    0.44
     MER
    0.40
     OCE
    0.39
     Divide
    0.38
    AX
    0.37
     Dand
    0.37
     Bros
    0.37
    𝘇
    0.37
     Kel
    0.36
     Spatial
    0.36
    POSITIVE LOGITS
    QMainWindow
    0.44
    焦點
    0.43
    ናል
    0.41
    unately
    0.40
    aport
    0.39
    shortcut
    0.39
    spanning
    0.39
     متش
    0.39
    urved
    0.38
    OnOff
    0.38
    Act Density 0.001%

    No Known Activations