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

    Neuronpedia

    INDEX
    Explanations

    colon followed by formatting or definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    1.02
    ки
    0.80
    и
    0.78
    =
    0.76
    ,
    0.75
    (
    0.74
    ed
    0.74
    al
    0.74
    -
    0.72
    ar
    0.72
    POSITIVE LOGITS
    ‬‬
    1.04
    innerApiCalls
    1.00
    ''}
    1.00
    ្សែ
    0.99
     Etiam
    0.97
    នុស្ស
    0.96
     effectivement
    0.96
     UnwrapRef
    0.95
    0.94
     callSettings
    0.94
    Act Density 2.271%

    No Known Activations