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

    Neuronpedia

    INDEX
    Explanations

    python code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Summit
    0.93
    ر
    0.90
    Т
    0.85
    0.83
     Aufenthalt
    0.83
     Москва
    0.83
    ద
    0.82
    ിക്ക
    0.82
    1
    0.81
    DWORD
    0.81
    POSITIVE LOGITS
     permeates
    1.09
     fermions
    1.06
     predicates
    1.05
     nucleons
    1.04
     thinkers
    1.04
     externalities
    1.01
     wynika
    0.95
    utenants
    0.94
     reinforces
    0.94
     deputado
    0.94
    Act Density 0.070%

    No Known Activations