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

    Neuronpedia

    INDEX
    Explanations

    authorized or consenting parties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     internships
    0.73
     mens
    0.72
     corruption
    0.70
     friendships
    0.69
     donations
    0.69
     expertise
    0.68
     sentences
    0.67
     cousin
    0.67
     marriages
    0.66
     configurations
    0.66
    POSITIVE LOGITS
    authorized
    0.82
     consenting
    0.79
     whoever
    0.72
     authorized
    0.72
     interested
    0.71
    interested
    0.71
    Authorized
    0.70
     नामित
    0.69
     willing
    0.68
     eligible
    0.68
    Act Density 0.229%

    No Known Activations