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

    Neuronpedia

    INDEX
    Explanations

    golang code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     juxtap
    0.70
     croix
    0.66
     partisans
    0.66
     vox
    0.66
    ardon
    0.64
     नकल
    0.63
    ට
    0.63
     mesin
    0.63
    creas
    0.63
     nœ
    0.63
    POSITIVE LOGITS
     /**
    1.10
    public
    1.09
     //
    0.98
     /*
    0.97
     //@
    0.94
     public
    0.94
    鬏
    0.94
    /*
    0.93
    /**
    0.93
    Public
    0.86
    Act Density 0.003%

    No Known Activations