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

    Neuronpedia

    INDEX
    Explanations

    Ghost

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     semp
    0.61
     Ghosts
    0.61
     ghosts
    0.59
     stell
    0.58
     spirit
    0.58
     starry
    0.56
     بما
    0.56
     BOTTOM
    0.55
     stella
    0.54
    spirit
    0.53
    POSITIVE LOGITS
    busters
    0.91
    BUST
    0.82
     Recon
    0.74
     recon
    0.71
    Ship
    0.70
    hunters
    0.70
    SHIP
    0.69
     SHIP
    0.68
    シップ
    0.68
     Ship
    0.67
    Act Density 0.003%

    No Known Activations