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

    Neuronpedia

    INDEX
    Explanations

    defining a role or persona

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     officers
    1.10
     specialists
    1.10
     designers
    1.00
     stylists
    0.99
     engineers
    0.95
     especialistas
    0.95
     educators
    0.94
     architects
    0.92
     специалистов
    0.92
     makers
    0.90
    POSITIVE LOGITS
     convinced
    0.86
    ulfide
    0.79
    confident
    0.79
    vinced
    0.77
    satisfied
    0.74
     greedy
    0.74
     greed
    0.74
     வேக
    0.73
    MessageBox
    0.73
     proud
    0.72
    Act Density 0.087%

    No Known Activations