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

    Neuronpedia

    INDEX
    Explanations

    safety goal(s) addressed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    California
    0.53
     California
    0.53
    Disney
    0.52
    ディズニー
    0.51
     Disneyland
    0.51
     Disney
    0.50
     useDispatch
    0.49
    Dis
    0.46
    dis
    0.45
     Calif
    0.44
    POSITIVE LOGITS
    <0xBC>
    0.41
     associated
    0.40
    KAK
    0.40
     rectification
    0.39
    customize
    0.38
     associée
    0.38
    ক্
    0.38
     partnering
    0.37
     memnun
    0.37
    ục
    0.37
    Act Density 0.000%

    No Known Activations