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

    Neuronpedia

    INDEX
    Explanations

    grades, tiers, and series designations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    enclose
    0.61
    self
    0.57
    approx
    0.55
    憉
    0.55
    <unused2126>
    0.55
    sick
    0.54
     virtual
    0.54
    redis
    0.54
    toggle
    0.53
    ースト
    0.53
    POSITIVE LOGITS
     III
    1.21
     II
    1.14
    III
    1.10
    II
    1.08
     IV
    1.00
    VII
    0.99
     VII
    0.96
     VIII
    0.92
    ２
    0.90
    2
    0.90
    Act Density 0.690%

    No Known Activations