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

    Neuronpedia

    INDEX
    Explanations

    widget build

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    jści
    1.24
    ።
    1.15
    ی
    1.14
    ї
    1.03
    ៉ុ
    1.00
    ار
    0.99
    hler
    0.98
    ء
    0.98
    vals
    0.97
    방
    0.94
    POSITIVE LOGITS
    twenty
    1.21
     ε
    1.13
    savvy
    1.11
     opinião
    1.10
     Eleven
    1.10
    sprite
    1.09
     eighteen
    1.07
     отлича
    1.06
    cD
    1.05
     lui
    1.05
    Act Density 0.002%

    No Known Activations