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

    Neuronpedia

    INDEX
    Explanations

    knowledge or lack thereof

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    chairs
    1.10
    Д
    1.10
     Thương
    1.04
    کری
    1.03
    гови
    1.03
    стіше
    1.03
    رى
    1.02
    roads
    1.02
    oare
    1.02
    <unused86>
    1.01
    POSITIVE LOGITS
     ،
    1.34
     ，
    1.09
    ，
    1.04
     centrifuge
    1.02
     azar
    1.00
     ؛
    0.98
     delas
    0.96
     desag
    0.95
     kantor
    0.94
     ,
    0.93
    Act Density 0.036%

    No Known Activations