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

    Neuronpedia

    INDEX
    Explanations

    code closures and function calls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    帶
    0.77
    ای
    0.76
     なん
    0.72
    یعنی
    0.68
    ্যা
    0.67
     이
    0.66
     እንደሚ
    0.66
     یعنی
    0.66
     '-':
    0.66
    虛
    0.65
    POSITIVE LOGITS
    atrol
    0.68
    ~
    0.65
    !
    0.65
    ulently
    0.65
    aménagement
    0.63
    며
    0.63
     að
    0.62
    |}
    0.62
     ∼
    0.62
     havia
    0.62
    Act Density 0.158%

    No Known Activations