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

    Neuronpedia

    INDEX
    Explanations

    materials of doors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gate
    1.12
     doorstep
    1.00
     doorway
    0.96
     Gate
    0.95
     door
    0.94
    Door
    0.93
     Door
    0.93
    Gate
    0.91
    door
    0.88
     درواز
    0.86
    POSITIVE LOGITS
     Bi
    0.45
    Folding
    0.44
     folding
    0.42
     Fleetwood
    0.42
     bi
    0.40
     Folding
    0.39
     ego
    0.39
     भारी
    0.38
    folding
    0.38
     શી
    0.38
    Act Density 0.008%

    No Known Activations