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

    Neuronpedia

    INDEX
    Explanations

    elixir code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ze
    0.59
    us
    0.58
    ве
    0.57
     stellar
    0.56
    чают
    0.56
    pir
    0.55
    ten
    0.54
    ொரு
    0.54
    чок
    0.54
    chi
    0.53
    POSITIVE LOGITS
    嚨
    0.56
    <0xAB>
    0.52
     inflict
    0.50
    ంగ్
    0.46
    アプリ
    0.45
     comboBox
    0.45
    ನ್ನೂ
    0.45
    同梱
    0.45
    𒆤
    0.45
    <0xAC>
    0.45
    Act Density 0.001%

    No Known Activations