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

    Neuronpedia

    INDEX
    Explanations

    activation functions relu sigmoid

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Convert
    0.75
    Common
    0.75
     aplicada
    0.74
    Transformation
    0.73
     പൊതു
    0.73
    Numer
    0.72
     Activ
    0.71
    Uit
    0.71
     Common
    0.70
    Activation
    0.70
    POSITIVE LOGITS
     rel
    1.08
     lean
    0.82
    湟
    0.82
     relay
    0.77
     handwriting
    0.73
     rele
    0.72
    Rel
    0.71
     REL
    0.71
     relational
    0.71
     validity
    0.71
    Act Density 0.011%

    No Known Activations