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

    Neuronpedia

    INDEX
    Explanations

    riddles, trusteeship, pool, taken, Screw Attack, diner

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     precludes
    0.57
     приводит
    0.50
     আট
    0.49
    Produces
    0.46
    ()}')
    0.45
    вает
    0.44
     suppresses
    0.44
     estere
    0.44
     уско
    0.44
     accelerates
    0.44
    POSITIVE LOGITS
     database
    0.51
     tradition
    0.48
     that
    0.48
     dedication
    0.47
     facing
    0.47
     left
    0.46
     vener
    0.46
    Kontakt
    0.46
    us
    0.46
     herald
    0.46
    Act Density 0.005%

    No Known Activations