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

    Neuronpedia

    INDEX
    Explanations

    stop or leave alone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     অনুভব
    0.44
     Cheat
    0.43
     disobey
    0.43
     अवज्ञा
    0.42
    elter
    0.41
    ibol
    0.40
     disob
    0.40
     betray
    0.39
     অবিশ্বাস
    0.39
     कैसा
    0.39
    POSITIVE LOGITS
     stop
    1.04
    stop
    0.99
    Stop
    0.97
     cease
    0.91
     Stop
    0.88
    停止
    0.83
    STOP
    0.82
     ngừng
    0.80
     STOP
    0.79
     переста
    0.78
    Act Density 0.011%

    No Known Activations