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

    Neuronpedia

    INDEX
    Explanations

    period or question mark followed by a word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    0.63
    尤其是
    0.53
     sogenannte
    0.51
    特别是
    0.50
    (
    0.50
     viser
    0.48
     denominado
    0.48
     sogenannten
    0.48
     Geschwindigkeit
    0.48
    குறி
    0.47
    POSITIVE LOGITS
    es
    0.55
    ྛ
    0.54
    ی
    0.53
    Ფ
    0.53
    د
    0.51
     Others
    0.50
    ン
    0.48
    其他
    0.47
    บริการ
    0.47
    esam
    0.46
    Act Density 0.388%

    No Known Activations