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

    Neuronpedia

    INDEX
    Explanations

    predictability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    在
    1.88
    ing
    1.70
     in
    1.59
    ة
    1.45
    我
    1.29
    ING
    1.26
    RE
    1.23
     at
    1.21
    SM
    1.21
    非常
    1.20
    POSITIVE LOGITS
    ן
    1.55
    ur
    1.15
    ता
    1.13
    operatorname
    1.06
    나가
    1.00
    ี
    0.98
    uc
    0.96
    is
    0.93
    uted
    0.93
    il
    0.91
    Act Density 0.000%

    No Known Activations