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

    Neuronpedia

    INDEX
    Explanations

    software versions and updates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ob
    0.59
     an
    0.58
     diariamente
    0.55
     lasciare
    0.55
     more
    0.54
     laisser
    0.54
    ob
    0.53
    ba
    0.53
     duas
    0.52
     часа
    0.52
    POSITIVE LOGITS
     版本
    1.10
    迭代
    1.08
     버전
    1.04
     codebase
    0.98
    バージョン
    0.98
     terbaru
    0.98
    アップデート
    0.98
    最新的
    0.96
     vigente
    0.96
     profinite
    0.95
    Act Density 0.277%

    No Known Activations