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

    Neuronpedia

    INDEX
    Explanations

    when something is done

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    0.45
     as
    0.39
    ↵↵
    0.39
     où
    0.38
     with
    0.35
     and
    0.35
    льні
    0.35
     där
    0.35
    ка
    0.34
    ທ່ານ
    0.34
    POSITIVE LOGITS
    ים
    0.52
    K
    0.39
    It
    0.37
    Timeout
    0.35
    B
    0.34
    N
    0.34
    িন
    0.34
    North
    0.33
    Mexican
    0.33
    Wow
    0.33
    Act Density 0.092%

    No Known Activations