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

    Neuronpedia

    INDEX
    Explanations

    bike lanes and pedestrian safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Traffic
    1.10
     Traffic
    1.07
     traffic
    0.95
    Signs
    0.91
    traffic
    0.88
     sign
    0.87
     Signs
    0.86
     ट्रैफिक
    0.86
     signs
    0.84
     Sign
    0.81
    POSITIVE LOGITS
     protected
    1.12
    Protected
    1.01
     Protected
    0.94
    protected
    0.92
     protections
    0.89
     protegido
    0.87
     proteg
    0.84
     защи
    0.81
     protecting
    0.81
     protection
    0.76
    Act Density 0.005%

    No Known Activations