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

    Neuronpedia

    INDEX
    Explanations

    strong recommendations and crucial advice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     important
    1.41
     importante
    1.38
    重要
    1.35
    important
    1.33
     importantes
    1.33
     Important
    1.32
    重要な
    1.29
     중요한
    1.28
     важли
    1.28
     গুরুত্বপূর্ণ
    1.27
    POSITIVE LOGITS
     strongly
    1.03
    strongly
    0.85
     IMP
    0.80
     CR
    0.80
     Strongly
    0.79
     STR
    0.79
    ESS
    0.76
     fortement
    0.74
    Strong
    0.72
     Imper
    0.72
    Act Density 0.137%

    No Known Activations