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

    Neuronpedia

    INDEX
    Explanations

    margin

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     margin
    -2.00
     margins
    -1.63
     Margin
    -1.51
    margin
    -1.50
    margins
    -1.21
    Margin
    -1.18
     penalty
    -0.98
     marginal
    -0.96
     margen
    -0.94
    MARGIN
    -0.94
    POSITIVE LOGITS
     Roskov
    1.04
    ########.
    0.72
    헌
    0.60
    ✨:
    0.59
    ]")]
    0.57
    expandindo
    0.56
    PyExc
    0.56
    writeField
    0.56
     [*]
    0.56
     bune
    0.56
    Act Density 0.059%

    No Known Activations