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

    Neuronpedia

    INDEX
    Explanations

    I need, I can, I'll

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     כאשר
    0.72
    此外
    0.69
     अथवा
    0.66
    Furthermore
    0.64
    どのような
    0.62
    Additionally
    0.62
     различ
    0.59
    Subsequently
    0.59
     तथा
    0.59
     כך
    0.58
    POSITIVE LOGITS
     want
    1.06
     haven
    1.00
     dunno
    0.98
     know
    0.93
     wanna
    0.92
     havent
    0.91
     đang
    0.89
     wants
    0.88
     messed
    0.88
     HAVE
    0.88
    Act Density 0.771%

    No Known Activations