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

    Neuronpedia

    INDEX
    Explanations

    input desired optional

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    觥
    0.82
     gotten
    0.81
    だから
    0.80
     cuz
    0.79
     porque
    0.78
     비슷
    0.77
     BECAUSE
    0.77
    avaient
    0.76
    踱
    0.76
     perché
    0.75
    POSITIVE LOGITS
     desired
    1.09
     optional
    0.89
     Optional
    0.82
    を入力
    0.82
    desired
    0.81
    Optional
    0.80
     gewüns
    0.77
     ука
    0.76
    optional
    0.75
     жела
    0.74
    Act Density 0.657%

    No Known Activations