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

    Neuronpedia

    INDEX
    Explanations

    edges

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    pia
    0.83
     Führung
    0.73
     Prefix
    0.69
     tul
    0.67
     Pi
    0.66
    тю
    0.64
     pi
    0.64
     Fractions
    0.63
     floated
    0.63
     Script
    0.63
    POSITIVE LOGITS
    edge
    0.64
    BODY
    0.58
    xD
    0.56
     edge
    0.56
    edges
    0.55
     edges
    0.54
    خب
    0.54
    Edge
    0.53
    ms
    0.52
    ME
    0.50
    Act Density 0.257%

    No Known Activations