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

    Neuronpedia

    INDEX
    Explanations

    underground spaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Pai
    0.83
     ху
    0.80
     Landscape
    0.79
    Landscape
    0.75
    Satellite
    0.74
    prisma
    0.74
    landscape
    0.73
     Satellite
    0.73
     prostagland
    0.73
     lawn
    0.73
    POSITIVE LOGITS
     underground
    2.83
     subterranean
    2.70
     Underground
    2.50
     tunnels
    2.30
     cave
    2.28
     지하
    2.25
     caves
    2.20
     caver
    2.16
     tunnel
    2.09
    地下
    2.08
    Act Density 0.324%

    No Known Activations