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

    Neuronpedia

    INDEX
    Explanations

    pathways through spaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    geocode
    0.45
     thereunder
    0.43
     underneath
    0.42
     ispod
    0.42
     inférieure
    0.41
     commonwealth
    0.41
     newspapers
    0.40
     núcle
    0.39
    0.39
    guests
    0.39
    POSITIVE LOGITS
     corridor
    0.83
     Corridor
    0.71
     hallway
    0.70
     corridors
    0.64
     wall
    0.63
     tunnel
    0.63
     exit
    0.63
     دیوار
    0.62
     maze
    0.60
     exits
    0.59
    Act Density 0.029%

    No Known Activations