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

    Neuronpedia

    INDEX
    Explanations

    content of that nature

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Docker
    0.65
     Einstellungen
    0.65
     Effective
    0.64
     yeah
    0.63
     kontroller
    0.62
     metrics
    0.62
    รู้
    0.62
     warranties
    0.62
     doprav
    0.61
    食
    0.61
    POSITIVE LOGITS
     importance
    2.20
     significance
    2.00
    importance
    1.98
     importancia
    1.77
     Importance
    1.75
     magnitude
    1.70
     nature
    1.66
    importanza
    1.62
    Importance
    1.60
     natureza
    1.51
    Act Density 0.135%

    No Known Activations