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

    Neuronpedia

    INDEX
    Explanations

    proper error handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    就會
    0.38
    roughly
    0.38
    whatever
    0.36
    GLOBALS
    0.34
     irgendwo
    0.34
     कहला
    0.34
     مثلا
    0.34
    SqlServer
    0.33
     במהלך
    0.33
     ciertos
    0.32
    POSITIVE LOGITS
     added
    0.67
     properly
    0.66
     explicit
    0.66
     explicitly
    0.61
     Added
    0.61
     included
    0.59
     correctly
    0.57
    ちゃんと
    0.57
     correttamente
    0.55
     правильно
    0.54
    Act Density 0.081%

    No Known Activations