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

    Neuronpedia

    INDEX
    Explanations

    argument help text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     бух
    0.42
     circumcision
    0.41
     Спа
    0.39
     Prairie
    0.38
     जलन
    0.38
    ড়াই
    0.38
    拎
    0.38
     conformal
    0.37
     oxidase
    0.37
     Bayou
    0.37
    POSITIVE LOGITS
     help
    0.71
     default
    0.65
     optional
    0.63
    help
    0.62
    default
    0.61
     defaults
    0.61
     defaulted
    0.61
    Defaults
    0.59
    setDefault
    0.59
     Defaults
    0.59
    Act Density 0.017%

    No Known Activations