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

    Neuronpedia

    INDEX
    Explanations

    square root and boxed math

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     commentaire
    0.85
     comentários
    0.85
     comentarios
    0.84
     tốc
    0.83
     overfitting
    0.82
     archivo
    0.80
     comentário
    0.80
     archivos
    0.80
     fichier
    0.78
     flowchart
    0.77
    POSITIVE LOGITS
    M
    0.71
    Cos
    0.70
    mselves
    0.70
    E
    0.68
    ści
    0.67
    msqrt
    0.66
     half
    0.66
    Component
    0.66
    Half
    0.66
    Extended
    0.65
    Act Density 0.144%

    No Known Activations