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

    Neuronpedia

    INDEX
    Explanations

    zones before punctuation or "of"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ac
    1.95
    ra
    1.93
     وأن
    1.81
    driven
    1.68
    を持ち
    1.67
    িক
    1.66
    点了点头
    1.63
    s
    1.63
     defend
    1.61
    posing
    1.61
    POSITIVE LOGITS
    }=\
    2.33
    া
    2.28
     étroites
    2.16
    ی
    2.05
    }}$,
    1.99
     Zones
    1.99
    та
    1.94
     subpopulations
    1.91
     baddies
    1.90
    }]$
    1.89
    Act Density 0.036%

    No Known Activations