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

    Neuronpedia

    INDEX
    Explanations

    items and their uses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     profound
    0.43
     fascinating
    0.37
     exploiting
    0.37
     influential
    0.37
     prevailing
    0.36
     duration
    0.36
     dominating
    0.36
     comprehensive
    0.36
     signifie
    0.35
     profoundly
    0.35
    POSITIVE LOGITS
     подар
    0.57
     gift
    0.50
     accessories
    0.49
     선물
    0.48
     gifts
    0.48
     白色
    0.47
    ススメ
    0.46
    礼物
    0.46
     подарок
    0.46
     toiletries
    0.45
    Act Density 0.003%

    No Known Activations