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

    Neuronpedia

    INDEX
    Explanations

    code elements and documentation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    0.72
    an
    0.65
     Instead
    0.64
     From
    0.62
     Series
    0.61
     Through
    0.61
     Formation
    0.61
    '
    0.61
     Using
    0.60
     Status
    0.58
    POSITIVE LOGITS
     is
    0.70
     за
    0.68
     आहे
    0.67
     у
    0.66
     restante
    0.65
    ﻖ
    0.61
    ба
    0.60
    ﺪ
    0.60
     ہے۔
    0.60
     ще
    0.60
    Act Density 0.521%

    No Known Activations